This topic will be moderated VERY severely. Balance whining and effortless posting will be bannable. Keep the discussion constructive, and generally avoid bad posting as well as posting incorrect advice. Thanks.
Special thanks to Thaniri for doing the initial research and writing the body for this thread. He spent a lot of time looking at replays and making sure we had good data to start the discussion with.
First off this thread is meant to be a discussion like the mythical days of old where Protoss players explored how to deal with the fast 200/200 Roach Max. Terran and Protoss players alike should take part in this thread. It is the belief of the SC2 TLStrat team that while strong, blink is not an insta-win in PvT. With effort, research, and understanding, it should be possible to elevate our understanding of Blink builds, and their variants in PvT to the point where execution is the deciding factor in whether one wins or loses.
This being said the goals for this thread are to: examine pro players who seem to win and how they won vs blink stalkers, examine pro players who lose and why they lost (where another player seemed to have won) against the same or similar build. We want to improve our understanding of not only how to engage but where to engage, and how to defend and scout to better prepare for incoming attack. Dealing with a blink into third is going to be very different from a 2 base blink all in and there HAS to be a difference between the two from a game plan perspective. Knowing how much you need to invest to defend the eco-oriented build is key to having either a follow up attack against the protoss to exploit some weakness, or being able to invest just enough to defend that you can choose to play macro yourself as a Terran.
This thread is NOT for balance complaints, it isn't for map complaints, it isn't a place to bitch and moan after your ladder loss. This thread is a DISCUSSION of what a Terran can do, can try to do, and should strive to do to improve your chances of holding a blink attack. Instead of spending a LOT of time researching and releasing a guide, we the TLStrat group are giving the community an opportunity to take part in learning. We are opening what we hope to be a starting point to better understanding the beautiful game of SC2. And just like when we tried to understand the Roach max, or the 1-1-1, there will be a break through.
The recent nerfs to Mothership core vision and buffs to widow mines could for all we know be the same as the Immortal six range buff in WoL against the 1-1-1. Just enough to help you hold if you play it right, but not so simple that the strategy is wiped off the face of the earth. In the end Terran will still need to execute a defense, but knowing what defense and how to do it is eluding us at this moment. Some professional Terrans have begun to see the light and we need to learn from them. That is what we are here for. Learning.
This thread is dedicated to helping Terrans deal with the recent trend of Protoss players going for two base blink stalker attacks. I'm going to throw out every piece of information that could be useful because most strategies tend to be broken by one "key."
Just remember my fellow Terrans, no matter how bad it gets this song is true for all of us. (NSFW)
Personal notes and observations:
The traditional 7:30 scan needs to be changed. Since Protoss is choosing tech at 5-6 minutes, a 6:30 scan can give confirmation of what tech path Protoss is going for if the reaper fails to scout anything.
Seeing the mothership core moving out is almost a guarentee of pressure. If you see the mothership core move out early it is important to delay any attacks by moving out with a small force and killing the probe that comes out as quickly as possible.
Widow mines are potentially good for dealing with stalkers. You only need one for each possible attack route the stalkers can take and they have no way of killing the mines because they have no detection. The flip side of this is that you will cut into your bio production to be able to afford these mines.
Stop using stim to chase stalkers unless they blinked already. They have blink to retreat, don't be silly.
Gateway units in direct engagements are worse than barracks units. One of the reasons blink is so strong early is because Protoss has many mobility advantages over Terran in the form of time warp, blink, and a faster base movement speed. This allows the Protoss player to kite the bio force in open spaces and to catch them completely out of position.
Having just a few marauders is no reason to chase the stalker army. Because concussive shells is delayed the Stalkers are only taking shield damage and can continue to kite the bio force and win the war of attrition. Terran's lower movement speed means that they must be using interception angles and meet the stalkers at their destination and not in transit.
The tech lab researching stim must be in a safe location. Having stim research picked off early is almost a guarenteed loss for Terran because stim and medivacs are the queue for Protoss to end the attack if it's not an all-in.
A scout and scan at 9 minutes can determine what follow up Protoss is planning.
Supernova on his stream actually makes tanks against 2 base blink stalker attacks regularly. I don't think that tanks are a great idea against all-ins however as the stalkers will be able to pick off the tank in one or two shots and overwhelm the Terran.
Picking off the mothership core early is amazing, that means that the stalkers can ONLY attack your natural and can be easily held off.
Looking at the map, the walk around outside the main to the natural is lengthy for the blink stalkers so they will most likely be attacking into the main. This does not mean however that you don't have to defend the natural.
Time warp is used as a zoning tool in the main base, almost as a gigantic forcefield. The stalkers can do quite a bit of damage while the Terran army struggles to get into a position to defend from.
Protoss as a rule of thumb is deciding on first tech between 5-6 minutes, which is useful for scouting.
Zest's attack begins at 9 minutes, Maru's bunkers are up in time for this.
The mothership core's placement makes aggression easy to predict, if Terran can easily track the location of the mothership core they will always be in position to defend a blink in.
Stimming to chase stalkers doesn't seem to work, it's not worth wasting the medivac energy.
It’s critical to sneak out a scout to check and stay at third base at 9 minutes. This will help determine what is coming after the attack.
After 2 base blink stalker pressure, your opponent has a weak composition because stalkers are weak against bio in general.
Once the attack is over it's important to quickly secure a third, this I think is more important than denying their third.
Protoss can follow up with either an expand, tech, or second timing. In each case they have a stalker heavy composition until about 13 minutes. (The decision looks like it’s made at 8:30) Seeing Protoss take gasses 3 and 4 can be followup timing or tech.
A second timing attack can be held by bunkers and mines because Protoss is still on gateway tech. Also useful to note is that if on equal upgrades, 4 medivacs outheal the dps of a zealot ball in the early midgame. Expand or tech means army size is too small to threaten your third.
A scan at 9 minutes would be good in this game, he’s got gas #3 and #4 at 8:41 so he is already planning out his followup to the blink attack.
This game is more of a note on scouting than anything. Since Dear did go for a Twilight Council it might be confusing for some to figure out if blink or DTs are coming. The tip is that if you scout the same gas timing as Dear had and a nexus halfway done at 6:30, it's a DT build.
Notice the good bunker placement used by both Maru and Bbyong. All potential attack paths Protoss can take are blocked.
Just like in the previous game, the Protoss player is making his followup decision at around the 9 minute mark.
Terrans need to be be less cheap with scans, and more careful withh their reapers. Once the mothership core and stalker are both out you have to respect your opponent and assume that he will block all attempts to scout with the reaper.
In this case I don't like how Bbyong decides to go 5 rax before expand. Here he can't possibly expect to do damage this early so he might as well develop his economy.
Terrans need to be more diligent with scouting the map for proxy pylons and probes.
Artosis mentions that you can just fake out this attack and start a third nexus behind it. The answer is to SCOUT, and SCAN. If your 9 minute scan sees no gates, and no gas at the natural, Protoss probably took a third base.
GuMiho's answer to Classic's blink play was to go for a quick single medivac drop and incite a sort of base race.
At home GuMiho defended with tanks. This worked because Classic chose not to go all-in with his blink stalkers and only meant to use them as pressure. Since there weren't enough stalkers to quickly dispose of the tank by blinking directly on top of it, GuMiho held on with repairs and built up a bio force large enough to eventually win with overwhelming numbers.
Maps:
The green circles are where Terran should think about defending with bunkers, mines, or positioning of his bio force. Red circles are where the Protoss army tends to fall back to. Green arrows defence paths, and red arrows are attack paths. Thin red arrows are rare but possible attack paths.
This map is just bad for blink, the travel distance to defend is significantly shorter than to attack and Protoss can only attack two points. It's far too easy for Terran to position properly to defend an attack.
This map has the widest open natural of all of them, this gives Protoss a lot of different angles to attack from and is difficult for Terran to defend.
They MIGHT walk past the leftmost spot to defend and mothership core recall once they’ve done some damage, but it’s not likely.
There are two main “launching points” for a protoss attack, and that is the ramp to the natural and the high cliff to the main.
The main launching point for Protoss attack is outside of Terran's natural.
The attacking distance between the high ground and low ground is longer than the defending distance, however the ramp is vulnerable to time warp.
Protoss CAN go all the way to the third and blink up the high cliff, but you do not need to worry because that distance is just absurd to walk and it is easy to defend in time.
This map is pretty bad for blink, the main launching point for attack is down the ramp at the natural and can only really attack two very close points at the natural.
In general, the stalkers will bounce back and forth from outside the natural to the third base.
There are a lot of points on the high ground that can be taken advantage of by stalkers, so bunkers should probably be positioned closer to the command center to reduce the area you are trying to defend. [*Tthe ramp is less vulnerable to time warp on this map because there is a lot of "Terran owned" space to cross for the mothership to get into position to cast it.
The main base is absolutely huge, the stalkers have a lot of opportunity to kite around because of it, because of that chasing is almost always a bad idea.
The natural only has one path into it, so bunkers with repair should be enough to hold that.
The ramp is very vulnerable to time warp as it is easy for Protoss to move their mothership core into position uncontested.
The two main attack paths are obvious, and the third one around the back has a ludicrously long walking distance and is very easy to prepare for.
The attacking and defending distances from the natural to the nearby high ground are identical, so bunkers are needed because of how difficult it is so keep up with the speediness of the stalker army.
Mod note: this thread will be moderated -extremely- severely. Balance whinning will be bannable, no effort posts will be warned.
Additionally, you should know what you are talking about before posting. Do not recommend incorrect/bad things like one base siege expand for instance.
I'll start off a bit of discussion with my perspective on the way maps impact Blink.
While some maps are stupid good for blink, I'm not sure what kind of architecture we want from new maps. While Yeonsu and Daedelus are very good for blink I really dont want to see blink leave the metagame entirely due to making blink literally impossible on all maps. Its nice to know that Protoss has an option of aggressive play that isn't Stargate based for example - from the perspective of a protoss player anyway.
I think what is really hard for a Terran from what I've seen/discussed with others isn't just the fact that Blink is strong, its that its difficult to tell if its a 2 base all in or a pressure into expand build. I don't do blink a lot myself but there must be a difference and Thaniri points out the gas count in the natural and scout timings. Knowing how to react to an all in variant can lead to a stronger defence without overcommitting which makes an expansion to a third follow up stronger. Hopefully someone has some good insight into the difference between an all in and a pressure into third as well as how to transition beyond the initial defense as a Terran into the mid game.
This is looking excellent. As a Terran player who hates playing against Blink and figuring out whether or not it's an all-in, I'm very interested to see what comes from this discussion.
I would like to point out that on Polar Night, it is possible for Protoss to go to the third base location and blink up into the main. Although the walking distance is relatively long, I thought you might want to include it just for thoroughness.
Edit: Would also be interested in seeing discussion about when you should or should not pull SCVs to help defend. I have seen some good defense where an SCV pull seemed necessary to tank for a small bio force that would have died without the help.
The problem isn't the blink all in for terrans now as much as the "2 base blink contain."
If you're overly defensive and can't move out, a Protoss can go up to 3 bases and chrono out storm in ~ 60 seconds, transitioning very smoothly into a strong macro game.
In addition, on some maps it is almost impossible to hold onto two bases vs 2 base blink, such as on Daedalus Point, where travel distance between nat and main is relatively long, meaning we put ourselves even further behind.
So I guess the key problems right now for us as Terran is:
a) Keep up economically (minimum 2 mining bases vs his 2 or 3 mining bases)
b) Don't die to the blink pressure
Those two do not seem to be simultaneously viable on certain maps at the moment.
Perhaps the builds we use are to fault. Right now the two most common builds in TvP (for standard macro play) is either 3 rax, popularized by Polt, or 1-1-1, which has the capacity to harass.
Unfortunately, neither really solve the problem.
Polt's way of dealing with 2 base blink was sending a small bio force to attack the enemy nat while there was no Stalkers or MSC at home. Really, really smart, but seeing CJ Entus's herO intercept the force with stalkers and simply attack to win shows that it's not a foolproof strategy.
If you sit at home and defend with 3 rax, we all know what happens; you will be wittled down; your bio count will shrink from 30 to 20 to 10, until when medivacs finally come out, he can be on a third base with storm almost finished, and you will not even have a third started. Obviously, this is not viable.
If we look at 1-1-1, it is similarly at an impasse. 1-1-1 attacks generally involve a fast medivac and either widow mines or hellions. Gimmick attacks designed to reduce enemy worker count, not end the game. To defend, you will have no marauders most likely, and no way to easily transition. Stim is late, marauders are late, upgrades are late, and you more often than not will have to make a tank to survive, and tanks are such gas hogs that they will hurt every transition you make.
The first promising new build I saw was qxc's 2-1-1 Build. Although he lost with the build, the concept is very interesting: qxc gets 2 rax for fast stim and marauder incorporation, but also gets a fast Factory so he can defend with mines. In the event of blink all in, I'd imagine you just hot swap the barracks and the factory to make more mines than marines, and just push out once you have medivacs. And with the build, you can get medivacs out on time (10 min) for standard pressure against protoss anyways.
Additionally, with the new mines, you have a decent transition against templar players if they go for templars (and almost all 2 base blink transitions into templars).
I am glad someone made a thread to discuss this all in seriously. I've been a GM protoss for several seasons, in both WoL and Hots. One thing I noticed the OP didn't mention is a sensor tower. This isn't needed on all maps but on a map like yeonsu I would heavily recommend it. The reason is that if toss draws you to the blink spot by your main CC (below the 1 on the map the OP drew). They can then run to your main and normally kill off all your bunkers with the help of a FF on your ramp.
Also if you find yourself always being incorrectly positioned against this blink all in I would just start making a sensor tower on maps where you don't need it until you get better at stopping the all in.
On March 22 2014 03:44 Shalashaka wrote: The problem isn't the blink all in for terrans now as much as the "2 base blink contain."
If you're overly defensive and can't move out, a Protoss can go up to 3 bases and chrono out storm in ~ 60 seconds, transitioning very smoothly into a strong macro game.
I don't like these kinds of statements, simply because they are vague and unspecific. In addition, you're exaggerating the ability of Protoss to transition out of 2-base Blink pressure. It sounds like you're saying that they can drop a Nexus, a Templar Archives, and chronoboost out storm in sixty seconds, while still threatening to kill you. I may not know Protoss builds, but it just seems implausible that they can invest so much into tech and economy while still being able to kill you, especially if you are playing defensive.
As I see it, the core issue you're getting at is: Against a 2-base pressure build, how does Terran walk the fine line between being too greedy and being too defensive (both of which are punished by the build)?
The OP suggests that it begins with first identifying you're facing a pressure (as opposed to all-in) build, then identifying when the Protoss is beginning their transition out of the pressure. You suggest that we could modify our current build orders to better defend the pressure and put us in a stronger position after the pressure is over. For both of these approaches, we need to analyze and dissect the way Protoss is playing, rather than make blanket statements about what it seems that Protoss is able to do.
In addition, you're exaggerating the ability of Protoss to transition out of 2-base Blink pressure. Suddenly they can drop a Nexus, a Templar Archives, and chronoboost out storm in sixty seconds, while still threatening to kill you?
I'll put it this way; once a Terran scouts blink certain precautions must be taken: we have to get bunkers, regardless of build, we can't start a 3rd CC, regardless of build, once the Stalkers into blink range we have to pull scvs to repair the bunkers, and leave them there, else we could be overwhelmed.
At any point, Protoss can skip a production cycle of Stalkers and throw down a Nexus. Skip another and throw down a Templar Archives.
The Terran simply has no way of knowing when the pressure is still continuing or when Protoss is merely posturing and going full greed behind the attack.
That's all I'm saying. I'm not saying the tech magically appears, but I'm saying the reactions Terran has to do, regardless whether it's just 6 stalkers or 20 will put them behind any Protoss who plays greedy.
What other option does the Terran have? Not build the bunkers? Not pre pull scvs to defend? Start the third CC early? The protoss can very easily just keep producing stalkers and punish the weakness in the Terran's defense.
If Terran defends perfectly, then Protoss simply expands and continues transitioning vs a Terran who's no where near a 3rd base, and more often than not has surrendered their natural.
In addition, you're exaggerating the ability of Protoss to transition out of 2-base Blink pressure. Suddenly they can drop a Nexus, a Templar Archives, and chronoboost out storm in sixty seconds, while still threatening to kill you?
I'll put it this way; once a Terran scouts blink certain precautions must be taken: we have to get bunkers, regardless of build, we can't start a 3rd CC, regardless of build, once the Stalkers into blink range we have to pull scvs to repair the bunkers, and leave them there, else we could be overwhelmed.
At any point, Protoss can skip a production cycle of Stalkers and throw down a Nexus. Skip another and throw down a Templar Archives.
The Terran simply has no way of knowing when the pressure is still continuing or when Protoss is merely posturing and going full greed behind the attack.
That's all I'm saying. I'm not saying the tech magically appears, but I'm saying the reactions Terran has to do, regardless whether it's just 6 stalkers or 20 will put them behind any Protoss who plays greedy.
What other option does the Terran have? Not build the bunkers? Not pre pull scvs to defend? Start the third CC early? The protoss can very easily just keep producing stalkers and punish the weakness in the Terran's defense.
If Terran defends perfectly, then Protoss simply expands and continues transitioning vs a Terran who's no where near a 3rd base, and more often than not has surrendered their natural.
Well to be fair if protoss has 6 stalkers and then skips two waves of stalkers that means they have only 6 stalkers. To get a nexus down and then a templar archives right away they need to have the natural gasses, and they need to be skipping upgrades. So in the midgame their army is going to be quite small (6 stalkers) and their map presence fairly weak.
If they make 12 or 18 stalkers then all their tech is severly delayed and its not really a pressure build anymore, its a fairly dedicated attack. So these two things can I am sure be scouted for in some way. Not necessarily through a direct scout but through reads for example.
Learning the Protoss natural gas timings can help in these scenarios as well. Since we cant get templar if we are on 2 gas, we need at least 4. And defending against 18 stalkers is VERY different than defending against 6 stalker pressure into expansion.
That's the distinction we are trying to make here. Instead of jumping to conclusions of I NEED ALL THIS STUFF TO DEFEND - you can instead look at I need this bare minimum to defend X and if they are doing Y I can adjust. Rather than just defending Y and hoping they don't do X. If they are playing greedy behind a pressure, the pressure likely won't kill you. If they are super ahead after the pressure, you over reacted to it and played too safely. Its a fine line and its an important part of StarCraft.
Since protoss players leave pretty much nothing at home while doing a build like this if it's an all in, having a reaper out on the map should be really helpful. That means you can scout and see if they're taking a third or teching to templar and getting upgrades behind the actual push. I suppose it's easier to do that if you do a reactor into 2 reaper opening, so that you can use your "spare" reaper to do the job later.
An scv scout can probably accomplish the same thing though if there are no units at home for the protoss. But if they're indeed not going all in, there may be units at home, so that's also a sign.
On March 22 2014 04:45 vhapter wrote: Since protoss players leave pretty much nothing at home while doing a build like this if it's an all in, having a reaper out on the map should be really helpful. That means you can scout and see if they're taking a third or teching to templar and getting upgrades behind the actual push. I suppose it's easier to do that if you do a reactor into 2 reaper opening, so that you can use your "spare" reaper to do the job later.
Unfortunately, you usually have to sacrifice the Reaper to identify that it's two base blink, as scanning is unreliable on maps with very big mains, such as D point or Heavy Rain.
So in your point of view, do you think it's more important to keep the Reaper alive at the 8 min mark to determine whether it's safe to break out or not, vs determining whether it's blink or not to begin with?
On March 22 2014 04:45 vhapter wrote: Since protoss players leave pretty much nothing at home while doing a build like this if it's an all in, having a reaper out on the map should be really helpful. That means you can scout and see if they're taking a third or teching to templar and getting upgrades behind the actual push. I suppose it's easier to do that if you do a reactor into 2 reaper opening, so that you can use your "spare" reaper to do the job later.
Unfortunately, you usually have to sacrifice the Reaper to identify that it's two base blink, as scanning is unreliable on maps with very big mains, such as D point or Heavy Rain.
So in your point of view, do you think it's more important to keep the Reaper alive at the 8 min mark to determine whether it's safe to break out or not, vs determining whether it's blink or not to begin with?
And thats something I hope to figure out. Particular players like polt and maru are very good at holding blink in TvP and what is it about how they play that gives them a leg up? I mean Polt and Maru are two very different players but they both deal with blink relative to other terrans very well and pretty consistently. Maru has great unit control but polt doesn't (not anywhere near close Maru skill) but Polt does have very good game plan decision making and understanding. So how do they confirm the blink build and how do they check for a third base or transition?
Its something Terran players can learn from. Maybe there is a way to scout for the twilight that keeps the reaper alive with good control for a future poke at the third, or if you sneak an scv out on the map before stalkers take control of it and use the reaper and accept its death to try and find the tech path?
These are viable options worth exploring from the perspective of trying to understand the builds and how to react.
On March 22 2014 04:45 vhapter wrote: Since protoss players leave pretty much nothing at home while doing a build like this if it's an all in, having a reaper out on the map should be really helpful. That means you can scout and see if they're taking a third or teching to templar and getting upgrades behind the actual push. I suppose it's easier to do that if you do a reactor into 2 reaper opening, so that you can use your "spare" reaper to do the job later.
Unfortunately, you usually have to sacrifice the Reaper to identify that it's two base blink, as scanning is unreliable on maps with very big mains, such as D point or Heavy Rain.
So in your point of view, do you think it's more important to keep the Reaper alive at the 8 min mark to determine whether it's safe to break out or not, vs determining whether it's blink or not to begin with?
Of course not. It's easier to scout later to determine whether it's an all in if you have (a) another reaper out, (b) an scv out, or (c) your initial reaper survives. Sending a single scv to look for tech buildings at a key time or spotting a third behind the push should allow players to adapt more adequately.
In addition, you're exaggerating the ability of Protoss to transition out of 2-base Blink pressure. Suddenly they can drop a Nexus, a Templar Archives, and chronoboost out storm in sixty seconds, while still threatening to kill you?
I'll put it this way; once a Terran scouts blink certain precautions must be taken: we have to get bunkers, regardless of build, we can't start a 3rd CC, regardless of build, once the Stalkers into blink range we have to pull scvs to repair the bunkers, and leave them there, else we could be overwhelmed.
At any point, Protoss can skip a production cycle of Stalkers and throw down a Nexus. Skip another and throw down a Templar Archives.
The Terran simply has no way of knowing when the pressure is still continuing or when Protoss is merely posturing and going full greed behind the attack.
I believe the first thing to do, as the thread starter actively encourages, is to abandon the mindset that Terran has no way of identifying whether Protoss has stopped the attack and started to transition, or if he is going to continue the attack.
You're correct in saying that once Terran sees the Protoss is going Blink, he has to take precautions in case it's an all in. However, I have to question your assertion that the Protoss can transition at any time. Can Protoss actually transition anytime they want? If not, when is the usual time to start cutting army production, and how many Stalkers must be sacrificed to transition? If yes, what stops Terran from transitioning themselves, assuming they can scout the transition from Protoss?
Because if Protoss cuts stalker production to transition, the army will naturally become weaker, in turn taking pressure off the Terran. It then becomes the Terran's job to sniff out that Protoss is transitioning, and then begin his own transition. The OP suggests timings and methods in which to do so.
On March 22 2014 04:20 Shalashaka wrote: That's all I'm saying. I'm not saying the tech magically appears, but I'm saying the reactions Terran has to do, regardless whether it's just 6 stalkers or 20 will put them behind any Protoss who plays greedy.
You assume that the Terran response to a Blink pressure build will always put the Terran behind (i.e. you overreacted on defense). But, again, you're not being specific. If you're going to make that assertion, it would be helpful if you could concretely define when you've played so defensively that you're behind. That's not easy, but that's one of the things we're actively discussing in this thread.
And your response as Terran will change depending on what Protoss is doing. You wouldn't play the same way against someone you knew was going to pressure with 6 stalkers while going full greed behind it, as you would against someone who is warping in 20 stalkers before transitioning. In both cases, what constitutes an under/over-reaction (and, of course, a correct reaction) will be different. It depends on Protoss' commitment.
So this ultimately becomes a question of: When do we know that Protoss has stopped committing to their pressure? That's one thing the OP is trying to find out with this thread.
So this ultimately becomes a question of: When do we know that Protoss has stopped committing to their pressure? That's one thing the OP is trying to find out with this thread.
The most obvious one (but difficult to check without a Reaper) is a 3rd Nexus coming down.
After that, counting Stalkers. If the count isn't getting any bigger (blink stalkers almost never split up) then start a 3rd CC maybe?
After that, what's toss's transitions? Some can use the Blink Stalkers to snipe turrets and follow up with a DT drop. Others can go Oracle, some can go Collo, others will do Templar. Identifying all of those without wasting too many scans is also key.
And finally, we have to remember Terran will almost have certainly taken some damage from the pressure, especially on a map like Heavy Rain or Daedalus Point, where blinking into the main, then blinking out and pressuring the nat with a timewarp on the ramp is very very dangerous for the Terran player.
Trying to use a lot of scans to check for all the above when you're already economically behind is less than ideal.
I think way too much of the above hinges on just not losing the Reaper, so maybe be extremely conservative with it? That's all I can come with atm.
So this ultimately becomes a question of: When do we know that Protoss has stopped committing to their pressure? That's one thing the OP is trying to find out with this thread.
The most obvious one (but difficult to check without a Reaper) is a 3rd Nexus coming down.
After that, counting Stalkers. If the count isn't getting any bigger (blink stalkers almost never split up) then start a 3rd CC maybe?
After that, what's toss's transitions? Some can use the Blink Stalkers to snipe turrets and follow up with a DT drop. Others can go Oracle, some can go Collo, others will do Templar. Identifying all of those without wasting too many scans is also key.
And finally, we have to remember Terran will almost have certainly taken some damage from the pressure, especially on a map like Heavy Rain or Daedalus Point, where blinking into the main, then blinking out and pressuring the nat with a timewarp on the ramp is very very dangerous for the Terran player.
Trying to use a lot of scans to check for all the above when you're already economically behind is less than ideal.
I think way too much of the above hinges on just not losing the Reaper, so maybe be extremely conservative with it? That's all I can come with atm.
It might not seem like a lot but its good. Try it on ladder and see if it helps you, also try to incorporate the bunker positioning from OP stolen primarily from Maru.
On March 22 2014 02:59 ZeromuS wrote: I'll start off a bit of discussion with my perspective on the way maps impact Blink.
While some maps are stupid good for blink, I'm not sure what kind of architecture we want from new maps. While Yeonsu and Daedelus are very good for blink I really dont want to see blink leave the metagame entirely due to making blink literally impossible on all maps. Its nice to know that Protoss has an option of aggressive play that isn't Stargate based for example - from the perspective of a protoss player anyway.
I think what is really hard for a Terran from what I've seen/discussed with others isn't just the fact that Blink is strong, its that its difficult to tell if its a 2 base all in or a pressure into expand build. I don't do blink a lot myself but there must be a difference and Thaniri points out the gas count in the natural and scout timings. Knowing how to react to an all in variant can lead to a stronger defence without overcommitting which makes an expansion to a third follow up stronger. Hopefully someone has some good insight into the difference between an all in and a pressure into third as well as how to transition beyond the initial defense as a Terran into the mid game.
Yeah, this is really more the problem than actually holding the build off. Most Terrans I know / practice with can hold a comitted 2 base blink all-in with relatively high consistency, it's just figuring out exactly what variation is coming and how much you have to commit to defense that throws us off.
There really isn't much of a difference from what I've seen between pressure and comitted attack (in fact it usually seems like they can plan to commit to attack, see the Terran has adequate defense and just not warp in and transition from there), and even if they commit to attack (i.e. more than 1 round of warp ins) they don't need to do an incredible amount of damage to transition out (delay stim or medivacs and kill some army and/or scvs is usually enough to safely take a third). They can also transition into either HT or Collosi from Blink Stalkers and either works really well, so it's important you scout out what tech they're going for asap.
I've had the most success vs blink all-ins when I keep a reaper alive after confirming blink after expand and being able to poke in and see if Protoss is adding tech / a third, but it seems really unreliable to keep the reaper alive after blink is out (a blink stalker can chase down / kill a reaper, 2 is pretty much garunteed to). You can't really afford to be scanning constantly vs Protoss either, you can barely afford 3 rax production + tech as it is, a counter medivac drop can be okay but its hits pretty late (typically after you know whether or not Protoss is transitioning) and takes away vital defensive units.
Killing the stalkers seems really important too, if you kill your opponent's stalkers and they try to transition out you can usually pick them apart with drops (and often kill / deny the third) while securing your own economy. I don't know of any way to consistently do this though, really just hoping to catch them out of position / hoping they overcommit.
I've been experimenting a bit recently with a faster 3rd CC vs 2 base blink (in an attempt to mitigate SCV losses with MULES and recover more quickly vs a fast tranition), but I haven't figured out a good way to do it without cutting too much into tech or defense. If you do have 3CC vs this build though, I've come back from some pretty big deficits in worker count and been fine once I stabilized because the SCV's really improved my army retention and the triple MULEs + CC's meant I could recover my economy fairly quickly. Sometimes I just die outright to an early / aggressive commital though.
From personal experience (and watching games like Polt vs Classic G1 from IEM Cologne, Jjakji vs sOs from IEM Katowice ect) opening 1/1/1 vs a 2 base blink is only defendable on Habitation or Polar Night, all other maps it's too easy for Protoss to get around the immobility of tanks. Widow Mines typically aren't worth it when opening 1/1/1 because your bio count is going to be a lot lower, so even if you kill one Stalker and soften up the rest the Stalkers will still be present in enough force to kill your remaining bio whereas Tanks will constantly rain down heavy damage (forcing the Stalkers to commit to rush in or retreat), and Tanks are better at zoning them out because of the greater range as well. OFC if you've already made widow mines placing them ~6 range away from the cliff is ideal and moving them after every blink in so that you can pick off / soften stalkers consistently.
I personally can't get it to work, but Polt's ~15 supply counterattack seems to be really strong vs this build too as long as you can avoid the blink stalkers with it, since Protoss leaves too little back at home to defend with just one warp in (and by the time they get a second you've already done enough damage).
Also a little trick on Heavy Rain, build your production in the corner of the map between you expansions, Protoss wont be able to pick off add-ons making it really difficult for them to delay stim / medivacs.
So this ultimately becomes a question of: When do we know that Protoss has stopped committing to their pressure? That's one thing the OP is trying to find out with this thread.
The most obvious one (but difficult to check without a Reaper) is a 3rd Nexus coming down.
After that, counting Stalkers. If the count isn't getting any bigger (blink stalkers almost never split up) then start a 3rd CC maybe?
After that, what's toss's transitions? Some can use the Blink Stalkers to snipe turrets and follow up with a DT drop. Others can go Oracle, some can go Collo, others will do Templar. Identifying all of those without wasting too many scans is also key.
And finally, we have to remember Terran will almost have certainly taken some damage from the pressure, especially on a map like Heavy Rain or Daedalus Point, where blinking into the main, then blinking out and pressuring the nat with a timewarp on the ramp is very very dangerous for the Terran player.
Trying to use a lot of scans to check for all the above when you're already economically behind is less than ideal.
I think way too much of the above hinges on just not losing the Reaper, so maybe be extremely conservative with it? That's all I can come with atm.
I think what you've gone into is a little besides the point. What you're describing is identifying specifically what toss is transitioning into. You're asking, "What is Protoss transitioning into?" while what we want to know first, during the Blink pressure, is simply, "Is Protoss transitioning?"
But yes, once a transition is confirmed, it's definitely important to figure out exactly what it will be (Macro with Colo/Templar or weird cheese like oracles and DTs).
It goes without saying you have to scout this coming. You don't scout a Blink allin, you aren't going to hold it. Let's put that question aside and focus on what happens after you identify the most common 2-base blink: 6 Warp Gates, Twilight Council researching and chrono'd, plus you've seen multiple Stalkers push out.
I believe that doing what Polt does is the most consistent way of attempting to respond to a 2-base Blink play. For those of you who maybe don't remember exactly what Polt does, here's an abbreviated build order plus two example VoDs (the Rain game is against a lighter Blink pressure on 2 bases while the Classic game is a full-fledged attack on 2 bases).
Reaper FE (pulling out of gas to get a quicker CC and follow up Barracks) 3 Barracks (1 Reactor, 1 TL) Get 1 Marauder after Stim and take map control while following normal economy and teching patterns. If you can, force the Overcharge and back off, hiding the forces on the map. As soon as the Protoss pushes out for the actual timing, backstab his natural and kill lots of Probes, plus force defensive warp-ins. Defend the allin with the rest of your army plus reinforcements.
The reason I feel like this play is the most consistent is that it relies on the two of you both being very good players for a compromise situation to be reached. Versus a Protoss who does not aggressively try to maintain his map control or who doesn't account for possible counterattacks as he prepares the allin, this build has a high chance of crushing the opponent, because 7 Marines and 1 Marauder showing up with Stim in your natural before you can start your allin when you have no units at home is extremely tough to handle. This is true to such an extent that if the defensive warpins DO happen, and you eventually kill off the backstab force, you will have lost so many Probes as well as blunted the power of your frontal assault with warpins at home that the Terran has a much easier time repairing and holding as a result.
Versus a Protoss who aggressively tries to account for this play, you need to keep constant vigilance on the map, dodging his pokes to try and find your force and attempting to not get caught in the open as the Protoss tries to move across the map with their Stalker pack. We saw this result in Polt's destruction on Polar Night at the hands of CJ_herO during IEM Katowice. If both sides posture in the mid-map and get nothing done, there is a good chance that you have delayed the start of the attack enough to get out Medivacs and enough reinforcements to have a good shot at holding if they ever do reach your side and commit to the attack.
Now, when it comes to scouting if the Protoss is actually committing or not... I really feel we as Terrans underutilize our Scans when facing an allin. It's true that having MULEs to get enough income to support repairing Bunkers is important, but in a situation where you see 6-8 Stalkers poking around the fringes of the base, but they haven't committed to Blinking in, I feel it is compulsory to either use the Reaper to verify a possible third/Templar transition or Scan for the third base. If you sit at home too long versus the fake-out type of pressure, you just let them get sufficient Storms to survive your pressure and then it's really hard to ever come back.
I think the scouting tree should go something like this:
1. SCV scout after Barracks, verify expansion and go home, checking common proxy Pylon spots that are not too far from your way. 2. Standard initial Reaper scout to verify that he has double gas, a relatively quick Stalker and the MSC. 3. Follow up Reaper sometime between 5:15 and 5:45 to check his choice of tech, BUT you have to save the Reaper here, don't just blindly suicide. Count his units at the very least. 3+ Stalkers should be a warning bell for you. 4. If you haven't yet verified, make sure that you GET the safety Marauder after Stim, and then suicide the Reaper to confirm Blink (of course, if you've already confirmed and know it IS Blink, save it for the future). 5. Move out to take some map control with the Marine/Marauder pack, THEN get Bunkers as you are making progress towards the other side. 6. If he doesn't contest mid map, force an Overcharge, pick off a lazy Stalker if possible, then stash them in a place unlikely to be scouted (ideally, pick two and bounce between them for added uncertainty). 7. As he moves out (about 8:15-8:35 or so if the build is tight - I would love Protoss confirmation on my math here), backstab the natural, the goal being to force defensive warpins, kill Probes, and kill the warping units (in that order, if my analysis of Polt is correct). 8. Prepare to defend the main push at your base with SCV/bio forces, teching to Medivacs. 9. Assuming he has somewhat defended the backstab but you haven't seen any Blinking in or a vast amount of Stalkers, use the saved Reaper or Scans to check for a possible third. If there is a third, transition to bio-mine (?) and apply Maru-style pressure; if not and it's just a delayed all-in (i.e. mass Gateway/Archon with DTs?) defend appropriately.
This is based entirely on watching Polt and Maru play TvP and trying to figure out what you can do to make the Protoss have to guess what's coming (thus the hidden backstab and attempts to take map control), as well as what might force the allin to be neutered in strength.
Now has any look been given to the game Illusion played against Minigun, I beleive in a WCS qualifier match? From what I could tell it was a 4 rax (I dont know if it was blind or reactionary, perhaps he suspected blink play based on the map) before factory, but I could be wrong about that. All I know was that by the time minigun tried to move out and establish his forward pylon, Illusion had enough marauders off of 4 barracks to destroy the forward force and push Minigun back.
I dont know if this was committed blink play as I cant remember the details of the game. If anybody can comment on this game specifically or the overall viability of a 4 rax strat against blink play, that would be great.
Edit: the map was Yeonsu, if that helps anybody pinpoint the game im referring to
On March 22 2014 05:47 Jazzman88 wrote: Okay, my thoughts on 2-base Blink allins: ...
Excellent post.
What are your thoughts on being ahead/even/behind when you've defended or taken no damage from a low-committal Blink pressure build? Does Terran come out ahead, or should we consider ourselves even?
How do you go about pressuring or potentially doing damage to a Protoss who has transitioned off a Blink pressure build? I'm wondering if there are windows for Terran to exploit once Protoss has pulled back, or if the Protoss can hold you off as long as he doesn't make any major mistakes.
On March 22 2014 05:58 dday0123 wrote: I think the graphics on a couple of the maps aren't showing all blink stalker attack paths and areas that have to be defended.
Not necessarily as common of paths, but Terran can die if they aren't ready for them too.
On March 22 2014 05:47 Jazzman88 wrote: Okay, my thoughts on 2-base Blink allins: ...
Excellent post.
What are your thoughts on being ahead/even/behind when you've defended or taken no damage from a low-committal Blink pressure build? Does Terran come out ahead, or should we consider ourselves even?
How do you go about pressuring or potentially doing damage to a Protoss who has transitioned off a Blink pressure build? I'm wondering if there are windows for Terran to exploit once Protoss has pulled back, or if the Protoss can hold you off as long as he doesn't make any major mistakes.
Terran should come out ahead if you take almost no damage (i.e. minimal SCV losses) if you didn't slack on your Medivac production and have 2 Medivacs worth of units to go pressure with. If you took too many SCV losses and your Medivacs aren't hitting until 11:30, you're going to have a bad time.
If you identify it early enough and he's doing the typical Templar transition, I believe bio-mine pressure at the third is very strong, because he has no detection by going Blink pressure into Templar. However, if he chooses the more time-consuming Colossi transition, obviously Widow Mines are not ideal. However, he might be open to an SCV pull timing.
I think it boils down to the initial hold. Minimize your SCV losses, and do your damnedest to get those Medivacs out on time, while not abusing your Stim unless necessary. Even if you're forced to delay until four Medivacs to really start hitting, those four Medivacs of bio can really lay the punishment down on vanilla Gateway units. AoE tech is the obvious worry, so you just have to respond accordingly.
It's still brutally hard to judge, because these windows where he doesn't have access to detection or the appropriate AoE are small ones, especially if he NEVER intends to fully allin and takes a much faster third transition behind the Blink. Gateway count would probably be the best indicator there. For example, if you scout/Scan at 8:45 and he only has 3-4 Gates on two bases, I wouldn't worry about him going Duckdeok on your ass.
Edit: if someone who gets Blink all-inned with transition as opposed to straight up allin more often than I do can confirm my suppositions based on Polt/Maru that I'm working with here, I'd be grateful.
The problem is that it's almost impossible to hold a blink all-in without losing at least 6 or 7 SCVs, since you have to repair the bunkers to hold it off. The moment the SCVs start repairing away, you simply target the SCVs repairing, and keep up with your blink micro (note that I'm in gold, so blink micro is rather lack luster when I'm defending).
If you have outstanding micro, it's basically a super safe way to get your opponent to nearly half the worker amount that you have.
Periodically sending an scv into stalker ball can give you an idea on how much he is committing to stalkers, as in all in or not. Or if his stalkers arent there and pressuring you send it around the map to his base, this of course won't always happen so the stalker count can help. Energy being used for time warp is a clear indicator of all in for me I feel. If it isn't all in the protoss wants to keep energy for nexus cannon to defend later.
Then again unless its high level of play a protoss just might never time warp or use it with a pressure build anyway.
Hello, I'm not in a very high league but I thought that maybe that could provide an 'out of the box' point of view
First, I wondered how viable a CC 1st opening was right now in TvP and which pros and cons it would have in the case of blink all ins, or which reactions it would force or encline on the Protoss that might lead to situations that would be easier to defend than a blink all in.
Second, about transitionning, I recently saw a Day9 daily about TvP Mine aggression with the focus on Maru. The key point Day9 talked about that would enable this style to be efficient was that the Protoss shoudln't have Colossi (which is the case here) and another advantage in this case, as stated above by someone else, is that the Protoss wouldn't have detection.
On March 22 2014 08:21 Ayrkrane wrote: Hello, I'm not in a very high league but I thought that maybe that could provide an 'out of the box' point of view
First, I wondered how viable a CC 1st opening was right now in TvP and which pros and cons it would have in the case of blink all ins, or which reactions it would force or encline on the Protoss that might lead to situations that would be easier to defend than a blink all in.
Second, about transitionning, I recently saw a Day9 daily about TvP Mine aggression with the focus on Maru. The key point Day9 talked about that would enable this style to be efficient was that the Protoss shoudln't have Colossi (which is the case here) and another advantage in this case, as stated above by someone else, is that the Protoss wouldn't have detection.
If needed, I will delete this post.
Don't worry about deleting. The discussion seems to be moving well so I'm happy with that fact. Honestly there is nothing wrong with your comments! Questions are just as good if not more important than answers in discussion threads because it forces people to think about their answers even more
On March 22 2014 03:44 Shalashaka wrote: The problem isn't the blink all in for terrans now as much as the "2 base blink contain."
If you're overly defensive and can't move out, a Protoss can go up to 3 bases and chrono out storm in ~ 60 seconds, transitioning very smoothly into a strong macro game.
I don't like these kinds of statements, simply because they are vague and unspecific. In addition, you're exaggerating the ability of Protoss to transition out of 2-base Blink pressure. It sounds like you're saying that they can drop a Nexus, a Templar Archives, and chronoboost out storm in sixty seconds, while still threatening to kill you. I may not know Protoss builds, but it just seems implausible that they can invest so much into tech and economy while still being able to kill you, especially if you are playing defensive.
As I see it, the core issue you're getting at is: Against a 2-base pressure build, how does Terran walk the fine line between being too greedy and being too defensive (both of which are punished by the build)?
The OP suggests that it begins with first identifying you're facing a pressure (as opposed to all-in) build, then identifying when the Protoss is beginning their transition out of the pressure. You suggest that we could modify our current build orders to better defend the pressure and put us in a stronger position after the pressure is over. For both of these approaches, we need to analyze and dissect the way Protoss is playing, rather than make blanket statements about what it seems that Protoss is able to do.
On March 22 2014 03:44 Shalashaka wrote: The problem isn't the blink all in for terrans now as much as the "2 base blink contain."
If you're overly defensive and can't move out, a Protoss can go up to 3 bases and chrono out storm in ~ 60 seconds, transitioning very smoothly into a strong macro game.
I don't like these kinds of statements, simply because they are vague and unspecific. In addition, you're exaggerating the ability of Protoss to transition out of 2-base Blink pressure. It sounds like you're saying that they can drop a Nexus, a Templar Archives, and chronoboost out storm in sixty seconds, while still threatening to kill you. I may not know Protoss builds, but it just seems implausible that they can invest so much into tech and economy while still being able to kill you, especially if you are playing defensive.
As I see it, the core issue you're getting at is: Against a 2-base pressure build, how does Terran walk the fine line between being too greedy and being too defensive (both of which are punished by the build)?
The OP suggests that it begins with first identifying you're facing a pressure (as opposed to all-in) build, then identifying when the Protoss is beginning their transition out of the pressure. You suggest that we could modify our current build orders to better defend the pressure and put us in a stronger position after the pressure is over. For both of these approaches, we need to analyze and dissect the way Protoss is playing, rather than make blanket statements about what it seems that Protoss is able to do.
The way I look I at it, the solution to figuring out the fine line between greed and defense is to find out what Protoss is doing sooner. The key issues plaguing Terran are not that Terran has no way of holding or that if they overreact they can be outmacroed. The issue is that Terran can't seem to know reliably what to do until ~7:30-8:30 and therefore have to set up their tech structure and everything accordingly at 5:30-6:30 (i.e. whether you go 2 rax, 3 rax, fast factory, etc., etc.). The mere threat of Protoss doing a blink allin, a proxy oracle, or a gateway timing is enough to force Terran into a little box where they cannot react until it's too late.
Thaniri's suggestion to change the scan timing to 6:30 is a good one to me. By 6:30, you've ruled out proxy oracle and early gateway attacks, leaving only blink, in-base oracle, DTs, or some kind of robo play left. Distinguishing between the blink and DTs is very easy because blink will already be started at this time and halfway finished. 6:30 is quite literally a conjunction period for Protoss where almost every build branches out from. Additionally, this further supports the goal of figuring out what Protoss is doing as early as possible, allowing the Terran player to react more immediately (i.e. starting marauder production, starting mine production, or adding an extra barracks, etc., etc.).
The question of how to reliably scout the tech reliably is a different one. I like the idea of actively scouting the map with a pack of units. I also think it's pretty much mandatory for Terrans to sweep the edges of the maps and the most common proxy locations with an early marine or two (after ~8). Much like how Zergs sweep their side of the map against Protoss, this will help Terrans more reliably slow down the blink pressure. In my opinion, Terrans have been much too scared to lose units and neglect a lot of scouting around the map early in the game. Protoss pressure is almost always frontal and straight across the map with a probes sneaking around the sides. Again, being able to identify a proxied building or a very suspicious pylon gives Terran the ability to know definitively what's going on a lot earlier = a lot more flexibility on how to react.
On March 22 2014 03:44 Shalashaka wrote: The problem isn't the blink all in for terrans now as much as the "2 base blink contain."
If you're overly defensive and can't move out, a Protoss can go up to 3 bases and chrono out storm in ~ 60 seconds, transitioning very smoothly into a strong macro game.
I don't like these kinds of statements, simply because they are vague and unspecific. In addition, you're exaggerating the ability of Protoss to transition out of 2-base Blink pressure. It sounds like you're saying that they can drop a Nexus, a Templar Archives, and chronoboost out storm in sixty seconds, while still threatening to kill you. I may not know Protoss builds, but it just seems implausible that they can invest so much into tech and economy while still being able to kill you, especially if you are playing defensive.
As I see it, the core issue you're getting at is: Against a 2-base pressure build, how does Terran walk the fine line between being too greedy and being too defensive (both of which are punished by the build)?
The OP suggests that it begins with first identifying you're facing a pressure (as opposed to all-in) build, then identifying when the Protoss is beginning their transition out of the pressure. You suggest that we could modify our current build orders to better defend the pressure and put us in a stronger position after the pressure is over. For both of these approaches, we need to analyze and dissect the way Protoss is playing, rather than make blanket statements about what it seems that Protoss is able to do.
The way I look I at it, the solution to figuring out the fine line between greed and defense is to find out what Protoss is doing sooner. The key issues plaguing Terran are not that Terran has no way of holding or that if they overreact they can be outmacroed. The issue is that Terran can't seem to know reliably what to do until ~7:30-8:30 and therefore have to set up their tech structure and everything accordingly at 5:30-6:30 (i.e. whether you go 2 rax, 3 rax, fast factory, etc., etc.). The mere threat of Protoss doing a blink allin, a proxy oracle, or a gateway timing is enough to force Terran into a little box where they cannot react until it's too late.
Thaniri's suggestion to change the scan timing to 6:30 is a good one to me. By 6:30, you've ruled out proxy oracle and early gateway attacks, leaving only blink, in-base oracle, DTs, or some kind of robo play left. Distinguishing between the blink and DTs is very easy because blink will already be started at this time and halfway finished. 6:30 is quite literally a conjunction period for Protoss where almost every build branches out from. Additionally, this further supports the goal of figuring out what Protoss is doing as early as possible, allowing the Terran player to react more immediately (i.e. starting marauder production, starting mine production, or adding an extra barracks, etc., etc.).
The question of how to reliably scout the tech reliably is a different one. I like the idea of actively scouting the map with a pack of units. I also think it's pretty much mandatory for Terrans to sweep the edges of the maps and the most common proxy locations with an early marine or two (after ~8). Much like how Zergs sweep their side of the map against Protoss, this will help Terrans more reliably slow down the blink pressure. In my opinion, Terrans have been much too scared to lose units and neglect a lot of scouting around the map early in the game. Protoss pressure is almost always frontal and straight across the map with a probes sneaking around the sides. Again, being able to identify a proxied building or a very suspicious pylon gives Terran the ability to know definitively what's going on a lot earlier = a lot more flexibility on how to react.
Agreed in that finding out sooner is better. I actually just played versus a Blink pressure build off two bases (not allin). It actually worked out as a testing ground for the ideas we're bandying around here.
I opened in the style of Polt and was aggressive with the Reaper scout. After scouting the fast Twilight Council, I darted back in again about 1:30 later to see if it was 4-6 Gates or DTs. He blocked the Reaper scout with a Photon Overcharge which killed it before it saw more than 3 Gates. While I sent in the Reaper, I checked for early Pylons near my base, and found none. As a result I ruled out super fast DT play and moved cross map with about 8 Marines 1 Marauder. I caught 1 Stalker, and forced the three he warped in to back up the ramp and wait for MSC reinforcements. While that happens, I killed 12 Probes and brought the MSC down to red health. I figure at this point he's already behind so I prepare for a heavy Blink allin with double highground Bunker, prepulled SCVs at the natural, and a safety Turret for emergency DTs. I hold the pressure easily, getting an MSC kill as a result.
It turns out he was just going 3 Gate pressure with Blink, and transitioned into superfast Storm behind that. After spotting the Templar with a Scan, he forced me back across map and I enacted the Mines follow-up mentioned earlier. Worked like a charm. With no upgrades for his units, eventually the bio-mine parade and decent splits carried the day.
Observations from this game as related to your post SC2John:
1. The earliest possible detection of his plans is a huge boon. As you noted, by 6:30, he HAS to have committed to his follow-up tech unless he is stuck in 2011. Spotting the Twilight as it finished and began researching gave me lots of time to make sure I was building Marauders and hitting his natural with a poke.
2. There actually isn't a lot you can do to adjust with a 3 Rax build aside from build more Marauders. The only adjustment you need make in that case is to get gas #2 slightly earlier so you can still get Factory tech in good time. I think this is another point in its favour: you aren't going to be veering way off course, because anything you face will be responded to with just another ton of bio units.
3. Sweeping Pylon locations is a must. Even if you don't get them all in the end, a concerted effort on his part to try to get them down should tell you something, and it is NEVER bad to force him to warp in further away than he wants to. Also, Protoss do not get many early units as a rule, and he therefore can't contest all possible Pylon locations at once.
Im so glad this seems to be fruitful. To be honest a lot of us were worried the discussion would be way less "discussion" and much more shitposting.
This being said, I'm glad the earlier scouting is working out for you Jazzman. I hope some sort of reliable way to check for tech around 6:30 is figured out beyond the hope my reaper doesnt die or my scan hits the right spot coin flip
I think the timing of the post contributed to why the quality of posts is better than we expected.
I'm sorry that I missed some of the possible attack routes on my map diagrams. My reasoning for omitting some of them was that the walking distance was just too long for Protoss and Terran should be able to position in time to receive the blink.
Since I open reaper expo into 1-1-1 in TvP I'm going to try banshees and mines against blink. I'll also try the qxc build someone linked. This weekend hopefully I have played against some blink players on ladder and can try it.
The immediate weaknesses I can see coming from it are no and stim, no medivacs. However, I will have plenty of minerals to make bunkers and fill them with marines because banshee cloak tech is probably 2 minutes later than a mine drop would come. My theory is that I can hold out for those two minutes with one reactor rax, mines, and 3-4 bunkers.
edit: I watched the qxc video, his point on making two battles that protoss can't pay attention to both that you WILL win at least one.
On March 22 2014 09:57 Thaniri wrote: I think the timing of the post contributed to why the quality of posts is better than we expected.
I'm sorry that I missed some of the possible attack routes on my map diagrams. My reasoning for omitting some of them was that the walking distance was just too long for Protoss and Terran should be able to position in time to receive the blink.
Since I open reaper expo into 1-1-1 in TvP I'm going to try banshees and mines against blink. I'll also try the qxc build someone linked. This weekend hopefully I have played against some blink players on ladder and can try it.
The immediate weaknesses I can see coming from it are no and stim, no medivacs. However, I will have plenty of minerals to make bunkers and fill them with marines because banshee cloak tech is probably 2 minutes later than a mine drop would come. My theory is that I can hold out for those two minutes with one reactor rax, mines, and 3-4 bunkers.
edit: I watched the qxc video, his point on making two battles that protoss can't pay attention to both that you WILL win at least one.
How are you currently doing versus Blink with your structure? I have completed abandoned 1-1-1 openings because I have a 100% loss rate with them versus any sort of Blink build this season. I just don't feel I can get both the quality and amount of units fast enough and still be safe in a macro game.
On March 22 2014 09:57 Thaniri wrote: I think the timing of the post contributed to why the quality of posts is better than we expected.
I'm sorry that I missed some of the possible attack routes on my map diagrams. My reasoning for omitting some of them was that the walking distance was just too long for Protoss and Terran should be able to position in time to receive the blink.
Since I open reaper expo into 1-1-1 in TvP I'm going to try banshees and mines against blink. I'll also try the qxc build someone linked. This weekend hopefully I have played against some blink players on ladder and can try it.
The immediate weaknesses I can see coming from it are no and stim, no medivacs. However, I will have plenty of minerals to make bunkers and fill them with marines because banshee cloak tech is probably 2 minutes later than a mine drop would come. My theory is that I can hold out for those two minutes with one reactor rax, mines, and 3-4 bunkers.
edit: I watched the qxc video, his point on making two battles that protoss can't pay attention to both that you WILL win at least one.
How are you currently doing versus Blink with your structure? I have completed abandoned 1-1-1 openings because I have a 100% loss rate with them versus any sort of Blink build this season. I just don't feel I can get both the quality and amount of units fast enough and still be safe in a macro game.
I'll be honest, I don't play against blink nearly as much as the community would make you think it's happening.
The games either go into a silly base race or me defending successfully with mines and bunkers. The mines take out stalker shields completely and put mothership core HP into the red, so if the Protoss army takes a mine hit they have to blink back 4 stalkers immediately or if the MSC gets hit it will get focused down.
There's an interesting dynamic where Protoss is "against the clock" as it were because mines do damage every X seconds and medivacs are obviously paying for themselves by healing more over time. As the mine count builds up Protoss simply can't attack in any more because too many stalkers are shieldless.
I'm actually embarassed to draw from my own games though because I am just not mechanically where I used to be in starcraft and need some time to practice. I very rarely lose in the early or midgame against protoss but when it comes to 200/200 army micro battles my skill is just not there anymore.
Using the reaper to scout is already hit or miss, and it relies on your reaper not dieing. If the Protoss does manage to pick off your reaper, what then? Do you just end up in a lucky scan or you're blind situation?
Being so reliant on a single fragile reaper doesn't seem very solid. Is there a backup plan?
The upside as I see it is you get scouting and if it is a blink attack there is nothing at home so you can do a lot of economic damage. With one reaper it is easy to block scouting, with two you can take different paths or often just go through the blocking force.
Something to notice as well is the Pylon positioning when you initially scout the Protoss, since important tech structures have to be placed near/around that pylon (obviously). But lately Protoss' have become really good at the simcity of their initial Pylon + Gateway to wall off 1 scout path of the Reaper (and sometimes CyberCore if necessary) to make it easier to hide the important tech structure since they can just place it behind the Gateway wall and zone out the Reaper w/ the MsC and/or Stalker because the path the Reaper takes is now more predictable. So checking for these pylon locations can make a Scan more successful if necessary to scout the Blink play, or the follow up tech if the Blink was just pressure and not an all-in.
Also, like the OP mentioned being diligent around the map checking for pylon(s)/probe(s) is a great way to know pressure is incoming or delaying the pressure successfully by killing the probe or pylon. This can be accomplished by being active with the Reaper if kept alive, or blocked from further scouting in the Protoss main.
It seems once the Stalkers are in front of your base with a Pylon set up, it's too late to get any scouting done. But since ~9 mins is the attack timings most associated with 2-base Blink, sneaking out 1 unit (whether it be an SCV or 1 Marine or if the Reaper survived) to scout the Natural Gas Timings, or main base if possible for transition information could be very helpful in being able to transition yourself or identify if the Protoss is fully committed.
Just a couple thoughts I wanted to pitch in. The Pylon placements early on are pretty important, and the Protoss tech is almost always in a secure location that is simple to defend or shield from a pesky Reaper since in PvT drops are so prevalent and strong. Getting down these little nuances in the Protoss play could be key in defending/identifying which style of Blink pressure they are going for, or if at all. GL HF!
Great job OP and everyone who has compiled information here, it's nice to see the community coming together again to help solve this problem although I'll be sad because my Protoss skills will be much weaker, I'll be happy thinking/playing the game different in response to what Terran's figure out! Which is why Starcraft is so great in the first place!
Edit: Sorry for the wall of text but something else I forgot to mention: I noticed in a game I watched recently (sorry I forget who it was exactly) that early on if the reaper is defended and blocked from scouting fully, poking with just a small group of units at the natural to draw the enemy army out of position while sneaking the reaper up into the main to scout what hasn't been seen yet is something that works and is relatively safe as well. Just making sure the units don't get kited is the hard part though.
On March 22 2014 15:25 Traceback wrote: Using the reaper to scout is already hit or miss, and it relies on your reaper not dieing. If the Protoss does manage to pick off your reaper, what then? Do you just end up in a lucky scan or you're blind situation?
Being so reliant on a single fragile reaper doesn't seem very solid. Is there a backup plan?
It's not solid, that's why I recommend a scan. Even then it's not foolproof as mentioned before in the thread.
On March 22 2014 15:51 Yurie wrote: How viable is a double reaper opening?
The upside as I see it is you get scouting and if it is a blink attack there is nothing at home so you can do a lot of economic damage. With one reaper it is easy to block scouting, with two you can take different paths or often just go through the blocking force.
Double reaper has a harder time against oracles, however, kas was doing it a couple weeks back on his stream though into a 1-1-1 and managed to hold oracle harass with mines, marines, and a medivac.
The thing about getting 2-3 reapers against protoss is that it delays your stim, and stim is the critical upgrade for terran. Once terran has stim the roles of the matchup reverse and terran should always be attacking.
I think double reaper is map situational, because on maps like habitation station you can't jump up a cliff into the main, there is only one scouting path so it doesn't matter if you have one or three reapers.
On March 22 2014 06:56 SupahSang wrote: The problem is that it's almost impossible to hold a blink all-in without losing at least 6 or 7 SCVs, since you have to repair the bunkers to hold it off. The moment the SCVs start repairing away, you simply target the SCVs repairing, and keep up with your blink micro (note that I'm in gold, so blink micro is rather lack luster when I'm defending).
If you have outstanding micro, it's basically a super safe way to get your opponent to nearly half the worker amount that you have.
You should check your settings, and make sure that Health Bars for "Damaged" units is on. This is helpful because if the Stalkers are shooting at your SCV's and not the bunkers, 1) the units are doing damage to the Stalkers and the Bunker is taking no damage or very little 2) pulling away damaged SCV's and replacing them with full health SCV's is a good way to save them and still repair. The 2nd is a little harder to pull off but you should be able to save a couple.
Also, does anyone know the desired amount of SCV's to use to repair against a certain number of Stalkers? This would be helpful so Terran's aren't over committing too many SCV's for repair, or pulling too little to repair. Also, the number of SCV's should vary depending on the Stalker count of course, so keeping that in mind should be useful. Keeping a spotter or patroling unit like 1 marine in the usual attack paths or reinforcement locations the Protoss use should tip you off pretty reliably to when exactly the enemy Stalkers arrive, or when the reinforcement Pylon is going down. The Protoss shouldn't engage without a reinforcement Pylon relatively close by to warp-in most effectively. So holding off on pulling SCV's to repair in order to squeak out some more resources before committing to them to the defense could be huge.
On March 22 2014 16:05 ArTiFaKs wrote: Something to notice as well is the Pylon positioning when you initially scout the Protoss, since important tech structures have to be placed near/around that pylon (obviously). But lately Protoss' have become really good at the simcity of their initial Pylon + Gateway to wall off 1 scout path of the Reaper (and sometimes CyberCore if necessary) to make it easier to hide the important tech structure since they can just place it behind the Gateway wall and zone out the Reaper w/ the MsC and/or Stalker because the path the Reaper takes is now more predictable. So checking for these pylon locations can make a Scan more successful if necessary to scout the Blink play, or the follow up tech if the Blink was just pressure and not an all-in.
Also, like the OP mentioned being diligent around the map checking for pylon(s)/probe(s) is a great way to know pressure is incoming or delaying the pressure successfully by killing the probe or pylon. This can be accomplished by being active with the Reaper if kept alive, or blocked from further scouting in the Protoss main.
It seems once the Stalkers are in front of your base with a Pylon set up, it's too late to get any scouting done. But since ~9 mins is the attack timings most associated with 2-base Blink, sneaking out 1 unit (whether it be an SCV or 1 Marine or if the Reaper survived) to scout the Natural Gas Timings, or main base if possible for transition information could be very helpful in being able to transition yourself or identify if the Protoss is fully committed.
Just a couple thoughts I wanted to pitch in. The Pylon placements early on are pretty important, and the Protoss tech is almost always in a secure location that is simple to defend or shield from a pesky Reaper since in PvT drops are so prevalent and strong. Getting down these little nuances in the Protoss play could be key in defending/identifying which style of Blink pressure they are going for, or if at all. GL HF!
Great job OP and everyone who has compiled information here, it's nice to see the community coming together again to help solve this problem although I'll be sad because my Protoss skills will be much weaker, I'll be happy thinking/playing the game different in response to what Terran's figure out! Which is why Starcraft is so great in the first place!
Edit: Sorry for the wall of text but something else I forgot to mention: I noticed in a game I watched recently (sorry I forget who it was exactly) that early on if the reaper is defended and blocked from scouting fully, poking with just a small group of units at the natural to draw the enemy army out of position while sneaking the reaper up into the main to scout what hasn't been seen yet is something that works and is relatively safe as well. Just making sure the units don't get kited is the hard part though.
I feel so silly for not mentioning the thought process behind scan placement. You want to scan where you reaper hasn't seen, and you know for sure a pylon could power a building.
For me though, a question I want answered is whether to use the reaper as move-out scouting or to suicide for tech scouting.
As it stands I prefer keeping the reaper alive as it's the fastest unit on the map before stim and can scout the edges for pylons effectively.
On March 22 2014 06:56 SupahSang wrote: The problem is that it's almost impossible to hold a blink all-in without losing at least 6 or 7 SCVs, since you have to repair the bunkers to hold it off. The moment the SCVs start repairing away, you simply target the SCVs repairing, and keep up with your blink micro (note that I'm in gold, so blink micro is rather lack luster when I'm defending).
If you have outstanding micro, it's basically a super safe way to get your opponent to nearly half the worker amount that you have.
You should check your settings, and make sure that Health Bars for "Damaged" units is on. This is helpful because if the Stalkers are shooting at your SCV's and not the bunkers, 1) the units are doing damage to the Stalkers and the Bunker is taking no damage or very little 2) pulling away damaged SCV's and replacing them with full health SCV's is a good way to save them and still repair. The 2nd is a little harder to pull off but you should be able to save a couple.
Also, does anyone know the desired amount of SCV's to use to repair against a certain number of Stalkers? This would be helpful so Terran's aren't over committing too many SCV's for repair, or pulling too little to repair. Also, the number of SCV's should vary depending on the Stalker count of course, so keeping that in mind should be useful. Keeping a spotter or patroling unit like 1 marine in the usual attack paths or reinforcement locations the Protoss use should tip you off pretty reliably to when exactly the enemy Stalkers arrive, or when the reinforcement Pylon is going down. The Protoss shouldn't engage without a reinforcement Pylon relatively close by to warp-in most effectively. So holding off on pulling SCV's to repair in order to squeak out some more resources before committing to them to the defense could be huge.
I think 3 SCVs per bunker that is most likely to be blinked on is good (6 total), once they have 3 rounds of stalkers though you probably want 2-3 per bunker and simply transfer them from one bunker to another once they stalker location is revealed.
Thoughts on Terran scouting the Protosss followup:
Let's assume that you do your 6:30 scan. I feel like on every map, you should designate an area to scout with your reaper where he can easily get out, without covering the entire main. This would probably be natural expansion, as well as the edge of the main. The rest of the main should be covered by a (fast) scan.
If you can do this, you can keep your reaper alive (no need to sac it behind a mineral line while 3 stalkers chase it for example), while at the same time spotting the Toss tech somewhat reliably. Having the reaper alive also allows you to easily identify the followup: take a tower with it, as soon as you see the stalkers move out hide it somewhere, and go back to the natural one-two minutes later. Check for gasses, as well as a third base.
I haven't seen any Terran trying this method of scouting before, and while extremely multitask-intensive (you have to stay alive vs the blink player in the meantime), i think it really could help in defending blink builds.
Thoughts? The big assumption here is being able to cover a full Protoss base with a scan and a reaper, while keeping the reaper alive. Blink friendly maps are also reaper friendly, so i think it's doable.
On March 22 2014 19:09 Teoita wrote: Thoughts on Terran scouting the Protosss followup:
Let's assume that you do your 6:30 scan. I feel like on every map, you should designate an area to scout with your reaper where he can easily get out, without covering the entire main. This would probably be natural expansion, as well as the edge of the main. The rest of the main should be covered by a (fast) scan.
If you can do this, you can keep your reaper alive (no need to sac it behind a mineral line while 3 stalkers chase it for example), while at the same time spotting the Toss tech somewhat reliably. Having the reaper alive also allows you to easily identify the followup: take a tower with it, as soon as you see the stalkers move out hide it somewhere, and go back to the natural one-two minutes later. Check for gasses, as well as a third base.
I haven't seen any Terran trying this method of scouting before, and while extremely multitask-intensive (you have to stay alive vs the blink player in the meantime), i think it really could help in defending blink builds.
Thoughts? The big assumption here is being able to cover a full Protoss base with a scan and a reaper, while keeping the reaper alive. Blink friendly maps are also reaper friendly, so i think it's doable.
I feel this is 100% applicable, though, as you mentioned, multitask-intensive. Maps I can see this being a problem on are Habitation Station due to lower number of Reaper entry points as well as Polar Night, since he can park two Stalkers on the ridge and the MSC/3rd Stalker at the nat and it's really hard to get into the base at all. However, the rest of them should pose no problem. Yeonsu, Frost, Alterzim are easy shift-queue move commands, and Daedalus, while tricky, has such a massive amount of jumpable space in the natural that you should be able to juke him at some point in order to get in, and all you need is the very back line near the mineral line, as a Scan will cover the rest of that protrusion towards the dead airspace (worth noting that you will PROBABLY have to suicide there because you can get it but it's very hard to get back out due to limited space).
I think when it comes to scouting, it's important to remember things that signal that you're not playing against a blink all-in (or any given build for that matter). Even if you don't see a twillight, it's usually easy to spot other tech structures such as a robo or an early forge. Most protoss players don't actually hide their robo. Gas timings at the natural can also be a big tell. The earliest timing for geysers 3 and 4 is 6:00, while most players will take their gases at 6:30-7:00. Taking them later doesn't even make sense, since you should have both bases saturated by 7:40-7:50 if you went msc expand. By looking for tells like these, you might be able to rule out certain builds.
Reaper/scan is a way to figure out what protoss does.
But i figured out having mines and marauders is a fantastic response to Protoss early game :
early and constant marauder prod counters stalkers; early and constant mine prod counters oracles and blink play
Rough build :
12 rax>2 marines 12 gas 15 OC 16 supply @100 gas factory > constant mine prod @25 gas tech lab on the rax > constant marauder prod + conc add starport/2nd CC as money flows in
Pros : Blind counters oracle play and more or less everything without detection Allows for an early attack with 2 marines/2marauders and 1/2 mines Gets scouting info by attacking instead of using a scan/a reaper Constant marauder/mine pumping makes a stalker based attack very hard for the protoss
Cons : countered by 10 gate, needs probably to use autorepair scvs initial pressure less effective on bigger maps you need to be diligent with the first mine : if you don't know 100% there's no oracle, just keep it at home delays eco
On big maps, you can also go reaper>factory and follow the same pattern while forgetting about the initial push.
I think most players realise the true strengths of the blink build as described in the OP are not overpowered or imba. Its the not knowing that terran has the problems with and hence the percieved imbalance. Basically I am saying people realise its not a huge balance issue and this thread seeks to find a way for terran to see the followups of the blink build and hence the terran can react and play the game on an even footing.
Haven't even finished with the case studies and already LOVE LOVE LOVE it. Hate defending this build, now I can't wait to try some new takes.
Okay minor note. Not so minor for me personally, but maybe some others in the future will benefit. ZeromuS Please put a **NSFW** tag on that song. I was in a public setting checking out the thread, other people gathering interest on a sing-songy feel good song (first girl had said how it reminded her of Don't Worry Be Happy). Then I TAKE A LOOK AT MY ENORMOUS PENIS ... to not so fun results (the unfunny prudes I'm tellin ya). The profane lyrics weren't appreciated in that setting.
On March 23 2014 02:43 TwiggyWan wrote: Reaper/scan is a way to figure out what protoss does.
But i figured out having mines and marauders is a fantastic response to Protoss early game :
early and constant marauder prod counters stalkers; early and constant mine prod counters oracles and blink play
Rough build :
12 rax>2 marines 12 gas 15 OC 16 supply @100 gas factory > constant mine prod @25 gas tech lab on the rax > constant marauder prod + conc add starport/2nd CC as money flows in
Pros : Blind counters oracle play and more or less everything without detection Allows for an early attack with 2 marines/2marauders and 1/2 mines Gets scouting info by attacking instead of using a scan/a reaper Constant marauder/mine pumping makes a stalker based attack very hard for the protoss
Cons : countered by 10 gate, needs probably to use autorepair scvs initial pressure less effective on bigger maps you need to be diligent with the first mine : if you don't know 100% there's no oracle, just keep it at home delays eco
On big maps, you can also go reaper>factory and follow the same pattern while forgetting about the initial push.
You are horribly far behind against a passive protoss player. 2 marauders, 2 mines and 2 marines do absolutely nothing against overcharge and any kind of detection (so basically, both oracle and robo builds).
The point isn't to find a terran build that blind counters a particular kind of protoss bullshit. The hard part is to find a combination of build order(s), scouting patterns, sets of reactions and possible transitions that can account for every protoss option.
The purpose of this thread isn't to discuss balance changes, but rather to figure out how to properly understand and react to blink builds in the current game.
Hm, I thought my post was like a "thought provoking" how to improve terran vision outside their base. I really think this is the biggest problem for terran if stalkers are 10meter away. I often see terran army stand at natural and blink stalkers are at mainbase. Protoss can see everytime where terran army stand because of MSC or observer. Terran doesnt have this ability.
On March 23 2014 19:59 Dingodile wrote: I rather see the problem in "terran vision". They just dont know where Stalkers are. A Sensor tower without gas-cost might solve this problem easily.
but the sensor power already is a very powerfull tool for the terrans, so if you not plan to max it to 1 sensor tower, this would be a very dangerous change for every matchup just because its not often used doesnt mean it isnt strong, making it no gas cost can be to strong
i personaly think its not enough time after the msc change, if after lets say 2 ? 3 ? months its still the same, its time to take the next step, but without enough time its hard to judge what is the correct step. lile the thread says, immortal range +1 fixed the 1-1-1 and that was small as well
On March 23 2014 19:59 Dingodile wrote: I rather see the problem in "terran vision". They just dont know where Stalkers are. A Sensor tower without gas-cost might solve this problem easily.
but the sensor power already is a very powerfull tool for the terrans, so if you not plan to max it to 1 sensor tower, this would be a very dangerous change for every matchup just because its not often used doesnt mean it isnt strong, making it no gas cost can be to strong
i personaly think its not enough time after the msc change, if after lets say 2 ? 3 ? months its still the same, its time to take the next step, but without enough time its hard to judge what is the correct step. lile the thread says, immortal range +1 fixed the 1-1-1 and that was small as well
And +1 queen range broke the game
Yeah, the point of this thread is not really to touch on ways Blizzard can fix the game, but rather how we, as players, can use the tools given to us to solve a problem. Already there's been a lot of discussion on how to defend, and I think everyone can more or less agree that it's possible to hold the blink all-in with fair success with a solid build and good scouting. The main issues we've been discussing otherwise is:
1) How can Terran get the information that their opponent is going blink earlier? 2) How can Terran know if it's a serious commitment to blink or just a blink pressure into macro? 3) How can Terrans reliably stop or slow down the blink pressure without drastically changing their build? 4) How can Terran scout the followup of the Protoss player while simultaneously defending blink harassment?
Personally, I think the best discussion follows Polt's counterattack style and QXC's guide about it. Being active on the map between 6:30-7:30 with a small pack of units really makes any kind of blink pressure much weaker and difficult while also achieving the goal of taking the pressure off the Terran player while they tech to medivacs. It also forces pylons and warpin points back much further. Terrans need to start doing this more often after they scout blink.
OH, one other point I wanted to bring up was Thaniri's comment on intelligent scanning. Terran players really need to note the placement of the first 3 pylons in the Protoss base. There really is no guessing with your scan as long as you know where the pylons are; whatever you miss with the reaper, you can intelligently scan so that you catch the last pylon or two. Very smart.
OH, one other point I wanted to bring up was Thaniri's comment on intelligent scanning. Terran players really need to note the placement of the first 3 pylons in the Protoss base. There really is no guessing with your scan as long as you know where the pylons are; whatever you miss with the reaper, you can intelligently scan so that you catch the last pylon or two. Very smart.
Hell, if you use an SCV scout after Barracks (which you should all be doing and if you aren't I will weep tears of blood and rage), you are easily able to spot the locations of all his in-base Pylons just with your SCV, and if they're close enough together and you're already going to be committed to scanning anyways (looking at you, Habitation Station), you won't even need the Reaper to verify. I mean, in most cases you will, but the Reaper isn't even necessary most of the time to verify Pylon locations, simply to have a less-expensive way to see Blink or Stargate or whatever.
OH, one other point I wanted to bring up was Thaniri's comment on intelligent scanning. Terran players really need to note the placement of the first 3 pylons in the Protoss base. There really is no guessing with your scan as long as you know where the pylons are; whatever you miss with the reaper, you can intelligently scan so that you catch the last pylon or two. Very smart.
Hell, if you use an SCV scout after Barracks (which you should all be doing and if you aren't I will weep tears of blood and rage), you are easily able to spot the locations of all his in-base Pylons just with your SCV, and if they're close enough together and you're already going to be committed to scanning anyways (looking at you, Habitation Station), you won't even need the Reaper to verify. I mean, in most cases you will, but the Reaper isn't even necessary most of the time to verify Pylon locations, simply to have a less-expensive way to see Blink or Stargate or whatever.
Hehe yeah. Most of the time there are two pylons near each other in the main (to prevent any kind of 2rax or some kind of cheese from killing you instantly) and then one hidden near the outskirts of the base somewhere. 90% of the time the twilight council is hidden behind the main mineral line, other times it's off to the side on the "hidden" pylon. Very rarely, players will hide their 2nd or 3rd pylon behind the minerals at their natural expansion. Like vhapter said earlier, robos are almost never hidden; so if you can't find any tech, it's probably either stargate or twilight (or some kind of wonky proxied gate allin lol).
Since the topic of spotting pylons has been mentioned, I think it's worth going over the most common scenarios. If you send an scv scout, there are 3 possibilities: spotting 1 pylon, 2 pylons, or 3 pylons.
1 pylon: If you can only spot a single pylon and 6 probes on gas, it's most likely a proxy stargate, but it could also be a super early proxy twillight. What I've seen Naniwa do, sometimes do it myself, and other players might do is build their second pylon at their natural to spot any ebay blocks... but yeah, this should be pretty rare on the ladder, so that's just one thing to keep in mind to avoid confusion. Basically, if there's no second pylon at his main or natural by the time your scv scout arrives, the protoss sent a probe out after building his gateway in order to proxy his second pylon for some reason.
2 pylons: If the protoss is going msc expand, you should be easily able to locate 2 pylons. The third pylon is only built after the nexus with this build, unless the protoss gets ebay blocked (I hope this bs doesn't get any more popular). That's because you don't need such an early third pylon, since you're cutting gateway units to get an early nexus. So if you see the protoss making anything besides a msc (including the zealot cancel trick, ofc) before his nexus, which in this case should be delayed anyway, and he didn't start his third pylon somewhere in his base by 3:40 at most, then he proxied his third pylon... or he's getting supply blocked.
3 pylons: An early third pylon indicates that the protoss is getting one or more units besides the mothership before the nexus, which could mean fear of an ebay block as explained above, perhaps a slower expansion (not very common, but on the ladder, you know... lol), or ofc a 1 base build.
Spotting the twillight council: Assuming the protoss went msc expand, he will soon build his third pylon. The biggest priorities in this build are obviously the nexus and the msc. After that, the protoss will start wg, his second gas if he hasn't taken it yet, and a third pylon (in no particular order). In this build, the protoss player won't start his fourth pylon before his nexus completes, so the location of the twillight is to be determined by the 3 pylons that were previously spotted... maybe the protoss could proxy a pylon for this purpose, but building a 4th pylon in your base before the twillight council would be pretty unusual.
I believe your biggest priority isn't to spot a twillight council asap, but rather whether there's a robo/forge or not. While some builds can get an early robo, most players will start their robo at 5:00 after their initial stalker. Most players build their robos by their main nexus, so it's a much easier structure to spot. The protoss can also build a forge at about 5:30-5:45 if he decides to delay his gateways That's also a signal that there'll be no blink all-in, though it could be a stargate build. If you see nothing but 1-3 gateways at this point, it's most likely something fishy.
OH, one other point I wanted to bring up was Thaniri's comment on intelligent scanning. Terran players really need to note the placement of the first 3 pylons in the Protoss base. There really is no guessing with your scan as long as you know where the pylons are; whatever you miss with the reaper, you can intelligently scan so that you catch the last pylon or two. Very smart.
Hell, if you use an SCV scout after Barracks (which you should all be doing and if you aren't I will weep tears of blood and rage), you are easily able to spot the locations of all his in-base Pylons just with your SCV, and if they're close enough together and you're already going to be committed to scanning anyways (looking at you, Habitation Station), you won't even need the Reaper to verify. I mean, in most cases you will, but the Reaper isn't even necessary most of the time to verify Pylon locations, simply to have a less-expensive way to see Blink or Stargate or whatever.
(...) so if you can't find any tech, it's probably either stargate or twilight (or some kind of wonky proxied gate allin lol).
Don't reveal all our secrets! Nobody does gateway all ins anymore, you betrayer of protoss! NOBODY, ok? Good.
On March 24 2014 02:37 vhapter wrote: Since the topic of spotting pylons has been mentioned, I think it's worth going over the most common scenarios. If you send an scv scout, there are 3 possibilities: spotting 1 pylon, 2 pylons, or 3 pylons.
1 pylon: If you can only spot a single pylon and 6 probes on gas, it's most likely a proxy stargate, but it could also be a super early proxy twillight. What I've seen Naniwa do, sometimes do it myself, and other players might do is build their second pylon at their natural to spot any ebay blocks... but yeah, this should be pretty rare on the ladder, so that's just one thing to keep in mind to avoid confusion. Basically, if there's no second pylon at his main or natural by the time your scv scout arrives, the protoss sent a probe out after building his gateway in order to proxy his second pylon for some reason.
2 pylons: If the protoss is going msc expand, you should be easily able to locate 2 pylons. The third pylon is only built after the nexus with this build, unless the protoss gets ebay blocked (I hope this bs doesn't get any more popular). That's because you don't need such an early third pylon, since you're cutting gateway units to get an early nexus. So if you see the protoss making anything besides a msc (including the zealot cancel trick, ofc) before his nexus, which in this case should be delayed anyway, and he didn't start his third pylon somewhere in his base by 3:40 at most, then he proxied his third pylon... or he's getting supply blocked.
3 pylons: An early third pylon indicates that the protoss is getting one or more units besides the mothership before the nexus, which could mean fear of an ebay block as explained above, perhaps a slower expansion (not very common, but on the ladder, you know... lol), or ofc a 1 base build.
Spotting the twillight council: Assuming the protoss went msc expand, he will soon build his third pylon. The biggest priorities in this build are obviously the nexus and the msc. After that, the protoss will start wg, his second gas if he hasn't taken it yet, and a third pylon (in no particular order). In this build, the protoss player won't start his fourth pylon before his nexus completes, so the location of the twillight is to be determined by the 3 pylons that were previously spotted... maybe the protoss could proxy a pylon for this purpose, but building a 4th pylon in your base before the twillight council would be pretty unusual.
I believe your biggest priority isn't to spot a twillight council asap, but rather whether there's a robo/forge or not. While some builds can get an early robo, most players will start their robo at 5:00 after their initial stalker. Most players build their robos by their main nexus, so it's a much easier structure to spot. The protoss can also build a forge at about 5:30-5:45 if he decides to delay his gateways That's also a signal that there'll be no blink all-in, though it could be a stargate build. If you see nothing but 1-3 gateways at this point, it's most likely something fishy.
I assume you're also accounting for gas expenditure and timing with this? It goes without saying that 2 gas done and only one Pylon powering both Gateway and CyberCore is proxy Stargate.
As I am not a Protoss player, what's the logic behind delaying the third Pylon until after 2 gas and the Nexus? Isn't it more efficient and less likely to be E-bay blocked with 1 gas and then it times out better? Or is this a mindgame/middle-of-the-road play to make the follow up Twilight or Stargate tech more difficult to detect off of one initial scout.
On March 24 2014 02:37 vhapter wrote: Since the topic of spotting pylons has been mentioned, I think it's worth going over the most common scenarios. If you send an scv scout, there are 3 possibilities: spotting 1 pylon, 2 pylons, or 3 pylons.
1 pylon: If you can only spot a single pylon and 6 probes on gas, it's most likely a proxy stargate, but it could also be a super early proxy twillight. What I've seen Naniwa do, sometimes do it myself, and other players might do is build their second pylon at their natural to spot any ebay blocks... but yeah, this should be pretty rare on the ladder, so that's just one thing to keep in mind to avoid confusion. Basically, if there's no second pylon at his main or natural by the time your scv scout arrives, the protoss sent a probe out after building his gateway in order to proxy his second pylon for some reason.
2 pylons: If the protoss is going msc expand, you should be easily able to locate 2 pylons. The third pylon is only built after the nexus with this build, unless the protoss gets ebay blocked (I hope this bs doesn't get any more popular). That's because you don't need such an early third pylon, since you're cutting gateway units to get an early nexus. So if you see the protoss making anything besides a msc (including the zealot cancel trick, ofc) before his nexus, which in this case should be delayed anyway, and he didn't start his third pylon somewhere in his base by 3:40 at most, then he proxied his third pylon... or he's getting supply blocked.
3 pylons: An early third pylon indicates that the protoss is getting one or more units besides the mothership before the nexus, which could mean fear of an ebay block as explained above, perhaps a slower expansion (not very common, but on the ladder, you know... lol), or ofc a 1 base build.
Spotting the twillight council: Assuming the protoss went msc expand, he will soon build his third pylon. The biggest priorities in this build are obviously the nexus and the msc. After that, the protoss will start wg, his second gas if he hasn't taken it yet, and a third pylon (in no particular order). In this build, the protoss player won't start his fourth pylon before his nexus completes, so the location of the twillight is to be determined by the 3 pylons that were previously spotted... maybe the protoss could proxy a pylon for this purpose, but building a 4th pylon in your base before the twillight council would be pretty unusual.
I believe your biggest priority isn't to spot a twillight council asap, but rather whether there's a robo/forge or not. While some builds can get an early robo, most players will start their robo at 5:00 after their initial stalker. Most players build their robos by their main nexus, so it's a much easier structure to spot. The protoss can also build a forge at about 5:30-5:45 if he decides to delay his gateways That's also a signal that there'll be no blink all-in, though it could be a stargate build. If you see nothing but 1-3 gateways at this point, it's most likely something fishy.
I assume you're also accounting for gas expenditure and timing with this? It goes without saying that 2 gas done and only one Pylon powering both Gateway and CyberCore is proxy Stargate.
As I am not a Protoss player, what's the logic behind delaying the third Pylon until after 2 gas and the Nexus? Isn't it more efficient and less likely to be E-bay blocked with 1 gas and then it times out better? Or is this a mindgame/middle-of-the-road play to make the follow up Twilight or Stargate tech more difficult to detect off of one initial scout.
To be honest, while it's not common, it's definitely possible to get a proxy twillight instead and go dts. I guess it's also possible to go 4 gate with a warp prism if you have only 3-4 probes on gas. But these things are not likely to happen very often, it's just a possibility you may want to account for.
I'm not sure if I understand your second question, but I'll try to answer:
Single gas mothership core expand allows a player to start a zealot slightly before 3 min, which is better against a blind 3 min mark ebay block. It also allows someone to expand earlier than a double gas msc expand build.
Double gas msc expand with 4 probes on gas provides you with more gas early on (duh), which is useful if you're going stargate or twillight. This delays your expansion for a little while, leaving you more vulnerable to lateish ebay blocks. You can also go 2 gas msc expand into robo too. In other words, double gas is more ambiguous than single gas. But I've seen a game where Parting went gas first (lol) against Flash and proxied a fast stargate off of a single geyser, and I've also seen Dear go proxy stargate with a single gas mothership expand too, but this obviously delays your stargate, stalker, or wg.
Delaying your third pylon allows you to take your Nexus earlier. If you need to let your zealot finish to avoid an ebay block, you can still make your third pylon as late as 3:40, which is after you would normally build your nexus, and still make a stalker + msc without getting supply blocked IF you save your 4th chrono boost (or cut 1 probe in time). That's because you will be 22/26 when your zealot finishes this way instead of 23/26, and still have 16 probes on minerals even after sending a probe down to build your expansion (or 15 on minerals if you're going double gas, but w/e). So basically, if your build is tight enough, there's no reason to get supply blocked becase you delayed your third pylon until you realized you were getting ebay blocked.
It's also possible to do an interesting double gas msc expand that takes a very early nexus. CJherO does that a lot in PvT. He takes both geysers, but favors mineral income over probes on gas (for a while he leaves only 1 probe in each geysers). It's easier to understand if you watch his replay. I've seen him taking a 3:35 nexus in spite of opening double gas. It looks very interesting and allows for a bit more gas than a regular single gas expand by the time you build a tech structure.
I'm not sure if this is exactly what you wanted to know, but I hope I answered your question even if indirectly (lol).
The game JUST finished, so there won't be a vod up yet, but game 3 of Major vs. Stardust, on Heavy Rain, of the Vasacast Invitational 2014, had Stardust going for 2 base blink, with Major defending with a lot of the ideas that have been discussed here so far--reaper scouting, counter attacking with one marauder and 8 (or 9?) marines, killing some probes and sniping the natural nexus, bunkers in pretty much the exact places show in the graphic here, and a widow mine or two. It would be a good one to add to the list of examples. Khaldor cast it, so maybe it will be up on his youtube channel?
The third pylon isn't really needed for MsC expand before the nexus; since your gate will (very likely) be idle for a while - assume no ebay block - you can confortably get your third pylon at 25 supply while making your core and not supply block any probe.
Thank you Teoita and vhapter, that is some very nice concrete information to have. Terrans take note of all that stuff - really good information to have in your back pocket (or a sticky note on your PC screen).
The biggest problem of all is first, that you cannot reliably scout the type of allin (or pressure) that you face. Reliably meaning, that you can scout it 100% if you take all the steps necessary. What are the options of scouting as a terran vs Protoss early in the game:
- Units like Reapers: Can scout almost their entire base if it gets in, but let's face it: it can be shutdown and cannot always give you the information you need
- Scan: As also mentioned earlier, if you are lucky the scan can reveal an important tech building from the Protoss, however, this is nowhere near realiable. Multiple Scans might do the job, yet this would equal a big investment from the terran, and still doesn't guarantee you safety. Yes you will get good and important information, certain tech buildings can still be built outside of the base. The only reliable things that a Scan can really reveal, is whether or not the Toss expanded for instance, gas timings and harvester count (which is still very nice, and can help you figuring out what is coming, you still might be in the dark about the kind of pressure though)
- Marines/scvs: Will allow you do scout the map for pylons or proxy tech
Let's say you scout a certain amount of gas having been mined, but you are still in the dark about the tech the toss player has chosen. People have said that the gas might tell you what you are up against yes, but thats only relevant, if you assume the other person is playing perfectly. Jumping to conclusions from gas count, might be useful in the very highest level of play, but I would say even low GM doesn't macro consistent enough to really make assumptions, given that the different builds from Protoss (especially allin builds) only differ by a very small margin. This is a key point in my opinion. Being forced to look at all the small and vague details to figure out the type of Protoss play, but then this might only be due to a macro mistake of your opponent.
Once the blink stalker allin is in progress, and even if you can hold it for now, as mentioned ealier, I also always have the problem of knowing when exactly Protoss transitions. Again I think Terran lacks reliable scouting. Still the most valuable option to go for, is scan the third base and see if a Nexus is in Progress or not. Then again, the toss might be going for a hidden expansion, in which case you would need some kind of unit outside of your base (which can be tracked down earlier by stalkers) unless you want to use more scans. Counting Stalkers as proposed before isn't reliable, a good player might not show the true Stalker count on purpose, and hide them elsewhere. It happened to me before, that you think the protoss is stopping the aggression and expanding, due to lack of units. Once you salvage your bunkers and move out, you then get surprised by hidden units.
Last, even if you hold against blink stalker, as mentioned earlier you can only really scout if the Protoss is getting a third, but not the tech he chooses to follow up his blink play. Even if you survive, there are many options that can catch you offguard, like a DT followup. This results in extra caution the Terran player has to take, and hence in more defensive investment (or in the worst case a completely wrong unit mix). Because of this, even if you hold the Blink Stalker allin, you might be behind in the macro game that follows.
If you do mistakes in any of those stages (pre- allin, during the allin, and transitioning out of it), you can lose the game, even if you do good in the others. That's why on non-pro level especially, I personally think that builds like the Blink stalker allin are so strong and frustrating to play against.
So this ultimately becomes a question of: When do we know that Protoss has stopped committing to their pressure? That's one thing the OP is trying to find out with this thread.
The most obvious one (but difficult to check without a Reaper) is a 3rd Nexus coming down.
After that, counting Stalkers. If the count isn't getting any bigger (blink stalkers almost never split up) then start a 3rd CC maybe?
After that, what's toss's transitions? Some can use the Blink Stalkers to snipe turrets and follow up with a DT drop. Others can go Oracle, some can go Collo, others will do Templar. Identifying all of those without wasting too many scans is also key.
And finally, we have to remember Terran will almost have certainly taken some damage from the pressure, especially on a map like Heavy Rain or Daedalus Point, where blinking into the main, then blinking out and pressuring the nat with a timewarp on the ramp is very very dangerous for the Terran player.
Trying to use a lot of scans to check for all the above when you're already economically behind is less than ideal.
I think way too much of the above hinges on just not losing the Reaper, so maybe be extremely conservative with it? That's all I can come with atm.
The two main ones you'd have to worry about is DTs and HTs, it's just a better use of the tech you've already invested in. Going Colossi is like buying a house in Europe, then marrying someone who's stuck in South America because of Visa issues or something when you knew about it in advance, it's just not very intelligent. Templar Archives takes 50 seconds to make, and it takes ~80 seconds for Storm to finish (with 3 Chronoboosts). So that's 2 minutes and 10 seconds for your next tech to kick in. To go for Colossi, you need 65 seconds for the Robo, 65 for the Support Bay, then each Colossus takes 50 seconds with 3 Chronoboosts. For just one Colossus (spending Chronos perfectly), you need 3 minutes. Range takes an additional minute and 20 seconds over that without Chronoboost, and nobody Chronoboosts Colossus Range. So High Templar (which is the far better tech path anyways in my opinion) takes literally half the time to transition into. A Stargate takes 60 seconds to make, it takes 34 seconds to get out an Oracle with 2 Chronoboosts, and it'll take 30-50 seconds to get to the enemy mineral line from your base, depending on the map (whereas the other two don't have to attack to be ahead). So again, that's just over 2 minutes, like the High Templar, and probably around the time that Terran would hit your front if they break out. So on one end, you can do tons of damage to an undefended base, on the other you have nothing to deal with the bio at your front. The Oracle play sort of dies too if they threw up preemptive Turrets to defend a DT play. So overall, Oracles also seem like a bad way to go, even though it's the fastest tech you can go for so far. The last transition, DTs, takes 100 seconds for the Dark Shrine, 5 seconds to Warp in a DT, and ~20 seconds to walk to the base from a Proxy Pylon. So overall, again just over 2 minutes. A DT transition will delay the Terran push, if not outright kill them, by forcing them to put up Turrets, and forcing Scans all the way to your base as you send in 1 DT at a time to slowly pick off units. I wouldn't call it the most efficient way to spend your gas (especially not compared to HTs), but it really pulls a Terran down when they're already feeling behind.
Then there's the cost of those 4 transitions. The Colossus transition is EXTREMELY expensive. You need 200/100 for the Robo, 200/200 for the Support Bay, 300/200 for each Colossi (600/400 before Range is done), and 200/200 for Colossus Range. The total cost for 2 Colossi with Range 4 minutes and 20 seconds after you start the transition is 1400/900. For Templar, it's 150/200, 100/300 for 2 Templar, 200/200 for Storm, and 200/200 for Zealot Legs which totals to 650/900. So you have enough money left over to get 7 more Zealots. Then Oracles take 150/150 each with 150/150 for the Stargate, so 300/300. DTs take 150/150 for the Dark Shrine and 125 for each DT (let's say 2), which is 400/400 total. Obviously, the Colossi and HTs are a long term investment, as you'll be using those for the rest of the game. The Oracles are very unlikely to be used frequently for the rest of the game, and the DTs are useful now and then, but not nearly as critical as HTs and Colossi. Off of 4 gasses, 900 gas takes just over 2 minutes to get. You can imagine how difficult it would be to get these critical tech units out with only 2 gasses running.Though, with only 2 gasses, they'd basically be all in anyways.
So, in summary, the transition is most likely to be High Templar. The second most likely (though very unlikely) are DTs, then Stargate, then Robo. The only scenario where I see the Protoss going Robo is if they went 1 base Blink, expanded cause they did a ton of pressure, then teched to Colossi, simply because you absolutely need 4+ gasses to get Templars promptly.
However, knowing all this, one hole seems to be the fact that rushing Templars means they won't have much room to fit in a Robo with Obsevers. So, if you hold the pressure then counter immediately, having a few Widow Mines would be uncontested except by Storms (and you'd need 2 Storms). So, I guess having well-positioned Widow Mines to defend Blink would be useful later on because the chances of them having detection to deal with them in your counterpush with those Widow Mines would be difficult unless they delayed their Templar Tech a bit. Maybe it's possible, but you would need to cut something. So yay for the Widow Mine buff! A glimmer of hope towards punishing 2 base Blink pressure and their transitions!
So this ultimately becomes a question of: When do we know that Protoss has stopped committing to their pressure? That's one thing the OP is trying to find out with this thread.
The most obvious one (but difficult to check without a Reaper) is a 3rd Nexus coming down.
After that, counting Stalkers. If the count isn't getting any bigger (blink stalkers almost never split up) then start a 3rd CC maybe?
After that, what's toss's transitions? Some can use the Blink Stalkers to snipe turrets and follow up with a DT drop. Others can go Oracle, some can go Collo, others will do Templar. Identifying all of those without wasting too many scans is also key.
And finally, we have to remember Terran will almost have certainly taken some damage from the pressure, especially on a map like Heavy Rain or Daedalus Point, where blinking into the main, then blinking out and pressuring the nat with a timewarp on the ramp is very very dangerous for the Terran player.
Trying to use a lot of scans to check for all the above when you're already economically behind is less than ideal.
I think way too much of the above hinges on just not losing the Reaper, so maybe be extremely conservative with it? That's all I can come with atm.
The two main ones you'd have to worry about is DTs and HTs, it's just a better use of the tech you've already invested in. Going Colossi is like buying a house in Europe, then marrying someone who's stuck in South America because of Visa issues or something when you knew about it in advance, it's just not very intelligent. Templar Archives takes 50 seconds to make, and it takes ~80 seconds for Storm to finish (with 3 Chronoboosts). So that's 2 minutes and 10 seconds for your next tech to kick in. To go for Colossi, you need 65 seconds for the Robo, 65 for the Support Bay, then each Colossus takes 50 seconds with 3 Chronoboosts. For just one Colossus (spending Chronos perfectly), you need 3 minutes. Range takes an additional minute and 20 seconds over that without Chronoboost, and nobody Chronoboosts Colossus Range. So High Templar (which is the far better tech path anyways in my opinion) takes literally half the time to transition into. A Stargate takes 60 seconds to make, it takes 34 seconds to get out an Oracle with 2 Chronoboosts, and it'll take 30-50 seconds to get to the enemy mineral line from your base, depending on the map (whereas the other two don't have to attack to be ahead). So again, that's just over 2 minutes, like the High Templar, and probably around the time that Terran would hit your front if they break out. So on one end, you can do tons of damage to an undefended base, on the other you have nothing to deal with the bio at your front. The Oracle play sort of dies too if they threw up preemptive Turrets to defend a DT play. So overall, Oracles also seem like a bad way to go, even though it's the fastest tech you can go for so far. The last transition, DTs, takes 100 seconds for the Dark Shrine, 5 seconds to Warp in a DT, and ~20 seconds to walk to the base from a Proxy Pylon. So overall, again just over 2 minutes. A DT transition will delay the Terran push, if not outright kill them, by forcing them to put up Turrets, and forcing Scans all the way to your base as you send in 1 DT at a time to slowly pick off units. I wouldn't call it the most efficient way to spend your gas (especially not compared to HTs), but it really pulls a Terran down when they're already feeling behind.
Then there's the cost of those 4 transitions. The Colossus transition is EXTREMELY expensive. You need 200/100 for the Robo, 200/200 for the Support Bay, 300/200 for each Colossi (600/400 before Range is done), and 200/200 for Colossus Range. The total cost for 2 Colossi with Range 4 minutes and 20 seconds after you start the transition is 1400/900. For Templar, it's 150/200, 100/300 for 2 Templar, 200/200 for Storm, and 200/200 for Zealot Legs which totals to 650/900. So you have enough money left over to get 7 more Zealots. Then Oracles take 150/150 each with 150/150 for the Stargate, so 300/300. DTs take 150/150 for the Dark Shrine and 125 for each DT (let's say 2), which is 400/400 total. Obviously, the Colossi and HTs are a long term investment, as you'll be using those for the rest of the game. The Oracles are very unlikely to be used frequently for the rest of the game, and the DTs are useful now and then, but not nearly as critical as HTs and Colossi. Off of 4 gasses, 900 gas takes just over 2 minutes to get. You can imagine how difficult it would be to get these critical tech units out with only 2 gasses running.Though, with only 2 gasses, they'd basically be all in anyways.
So, in summary, the transition is most likely to be High Templar. The second most likely (though very unlikely) are DTs, then Stargate, then Robo. The only scenario where I see the Protoss going Robo is if they went 1 base Blink, expanded cause they did a ton of pressure, then teched to Colossi, simply because you absolutely need 4+ gasses to get Templars promptly.
However, knowing all this, one hole seems to be the fact that rushing Templars means they won't have much room to fit in a Robo with Obsevers. So, if you hold the pressure then counter immediately, having a few Widow Mines would be uncontested except by Storms (and you'd need 2 Storms). So, I guess having well-positioned Widow Mines to defend Blink would be useful later on because the chances of them having detection to deal with them in your counterpush with those Widow Mines would be difficult unless they delayed their Templar Tech a bit. Maybe it's possible, but you would need to cut something. So yay for the Widow Mine buff! A glimmer of hope towards punishing 2 base Blink pressure and their transitions!
Collosi might be an expensive transition, but it's also a very common one because Blink Stalkers synergize very well with Collosi and Protoss typically has 6-12+ Blink Stalkers left over when transitioning out of 2 base blink. While it's true that HT and/or Chargelot Archon (Occassionally with DT rather than HT) is the most common follow up to 2 base blink, Collosi is done very often too. I've never even seen an Oracle follow up (any pro game links / ect?).
Just a quick suggestion, why not remake the lost reaper to check what the Protoss follow up is(tech/3rd/upgrades)? The reaper, even if shot down eventually, can help identify where to scan to figure out the follow up rather than blind scanning.
Unless the production time bottlenecks your barracks too much, but the information gleaned (if you spot an archive or no 3rd base etc) would be essential? Scans only come after energy accrual and you only get them at certain time intervals, the reaper can guide your scanning both by what it sees and what it doesn't see
On topic of transitions: the only two that are reasonable are HT (and chargelot/archon variants; the specific details change from player to player and game to game) and Colossi. Both are reasonable, and both can be done easily. The Colossus transition being slower isn't an issue because a) many blink>colo builds get a robo somewhat fast (as in, while the pressure is going on) and b) your stalkers buy time for the tech to kick in.
Blink > oracle isn't a thing, and blink > dt is either a desperate measure when you are behind, or something you do when blink wins you the game but the terran wont leave, so you kind of tell him to gtfo.
On March 24 2014 17:21 MrInocence wrote: Just a quick suggestion, why not remake the lost reaper to check what the Protoss follow up is(tech/3rd/upgrades)? The reaper, even if shot down eventually, can help identify where to scan to figure out the follow up rather than blind scanning.
Unless the production time bottlenecks your barracks too much, but the information gleaned (if you spot an archive or no 3rd base etc) would be essential? Scans only come after energy accrual and you only get them at certain time intervals, the reaper can guide your scanning both by what it sees and what it doesn't see
The traditional reasoning for not building that follow-up Reaper is that Terran is trying to ration their gas a lot when defending Blink. You need a lot of minerals to build the appropriate amount of Bunkers, plus your gas is pretty much all sunk into getting Stim, Marauders, and Medivacs. I would much rather have to Scan twice than have only one Medivac for an additional minute of game time. Even when not playing versus Blink, if your build is tight and gearing for the 10-11 minute window with bio+Medivac pressure, you should never really have the opening to get an additional Reaper without adversely affecting either Medivacs or upgrades. Plus, they build almost twice as slowly as Marines, meaning for each Reaper you build, it's two less Marines in your army plus at least 1 less Marauder or Medivac.
On March 24 2014 17:36 Teoita wrote: On topic of transitions: the only two that are reasonable are HT (and chargelot/archon variants; the specific details change from player to player and game to game) and Colossi. Both are reasonable, and both can be done easily. The Colossus transition being slower isn't an issue because a) many blink>colo builds get a robo somewhat fast (as in, while the pressure is going on) and b) your stalkers buy time for the tech to kick in.
Blink > oracle isn't a thing, and blink > dt is either a desperate measure when you are behind, or something you do when blink wins you the game but the terran wont leave, so you kind of tell him to gtfo.
Agreed, I've seen both HT and colossi work just fine. And it's not even hard to fit a robo + observers/immortals in there when going HT either (you just can't take a stupid early 3rd). The only other real threat is DTs, which, like you said, are fairly uncommon.
That said, I don't think Terran should be worrying so much about what TECH the Protoss is going for afterwards because you can't really have ghosts or vikings out that early anyway and your initial counterattack push will force the Protoss to reveal his AoE. Terrans should be focusing on 3 things concerning followups: 1) when the Protoss takes his natural gases, 2) if the Protoss takes a 3rd, and 3) if the Protoss is going single or double upgrades.
1) When Protoss takes natural gases
Knowing the timing of the natural gases allows you to gauge how heavy your opponent is committing to blink. Generally the natural gases signal that the Protoss will not be warping in anymore units and will instead be putting all resources toward teching and/or economy. Timings can range to before the stalker attack (meaning a very light commitment to pressure) to when the Protoss is convinced that his pressure has failed and is forced to transition.
2) If the Protoss takes a 3rd
Protoss players can't easily take a 3rd, contrary to popular belief. Most of the time, taking an early third behind blink pressure forces them to sacrifice something else in order to get it. For a HT transition, for instance, it's impossible to get an early 3rd, double upgrades, a robo for observers, and fast HT all at the same time. Again, whatever the transition is, knowing if there is a 3rd base is far more valuable information as it actually affects your decision making directly.
3) If the Protoss is going single or double upgrades
In a way, this is a finer point, but it still deserves mentioning. Whenever Protoss goes double upgrades, the Terran is on a bit of a timer to do damage or suffer an insanely powerful 3-3 attack. I'm not sure exactly what the proper response is, but I feel like Terrans playing against double upgrades need to either do some massive counter damage with drops or a frontal assault or they need to focus on tech and perhaps grab a faster 4th.
Anyways, these are just some of my thoughts concerning scouting the Protoss followup. I don't think scouting for HT vs. colossi really matters that much compared to a lot of other things.
Colossi take much longer to get. If I recall corretly, it takes 3:20 just to get a single colossus out without any chrono boosts if you don't have a robo, and 4:25 or something for thermal lance to finish. Since you already have a twillight council up, you can make hts in just 50 seconds and morph them into archons if you need some defense, It takes 2:40 to get a templar archives up and research storm. As already mentioned, you can also add a robo to get immortals if you're going ht.
So whenever you choose to transition, teching to storm is definitely faster than going colossi without having a robo. I believe this is why storm is a more common transition at pro level - it's just more accessible. Getting a robo during the push doesn't make it necessarily "faster" than the templar route either, because you could also make a templar archives during the push anyway. But it does make the transition smoother and safer.
Also going templar is generally preferred right now (which makes me happy); blink pressure transitioning so easily into templar simply makes the macro variations so much scarier.
Hi, I'm only a gold terran so I can't possibly grasp everything you guys do (especially the startpost, I find it most difficult to not getting raped by focusfire on the bunker I am not repairing at that moment).
I just have one little suggestion on my mind and I'll try to commit this to the discussion: Actually I remember something day9 has said very often and that is "I like when players make small adjustments". If there is a problem with scouting if the protoss takes a third or not after the pressure all-in, why don't we just make the small adjustment, take out 1 marine/reaper/scv and copy the protoss. We hide this little guy somewhere near the potential third on the map or maybe to make this nearly effortless just let it sit at one of the potential thirds.
This actually allows us to have vision over one of the possible locations and also to scout a potential third and I'm sure you guys know the timings for that third after blink right?
On March 24 2014 08:09 AoWAraGorn wrote: The biggest problem of all is first, that you cannot reliably scout the type of allin (or pressure) that you face. Reliably meaning, that you can scout it 100% if you take all the steps necessary. What are the options of scouting as a terran vs Protoss early in the game:
- Units like Reapers: Can scout almost their entire base if it gets in, but let's face it: it can be shutdown and cannot always give you the information you need
- Scan: As also mentioned earlier, if you are lucky the scan can reveal an important tech building from the Protoss, however, this is nowhere near realiable. Multiple Scans might do the job, yet this would equal a big investment from the terran, and still doesn't guarantee you safety. Yes you will get good and important information, certain tech buildings can still be built outside of the base. The only reliable things that a Scan can really reveal, is whether or not the Toss expanded for instance, gas timings and harvester count (which is still very nice, and can help you figuring out what is coming, you still might be in the dark about the kind of pressure though)
- Marines/scvs: Will allow you do scout the map for pylons or proxy tech
Let's say you scout a certain amount of gas having been mined, but you are still in the dark about the tech the toss player has chosen. People have said that the gas might tell you what you are up against yes, but thats only relevant, if you assume the other person is playing perfectly. Jumping to conclusions from gas count, might be useful in the very highest level of play, but I would say even low GM doesn't macro consistent enough to really make assumptions, given that the different builds from Protoss (especially allin builds) only differ by a very small margin. This is a key point in my opinion. Being forced to look at all the small and vague details to figure out the type of Protoss play, but then this might only be due to a macro mistake of your opponent.
Once the blink stalker allin is in progress, and even if you can hold it for now, as mentioned ealier, I also always have the problem of knowing when exactly Protoss transitions. Again I think Terran lacks reliable scouting. Still the most valuable option to go for, is scan the third base and see if a Nexus is in Progress or not. Then again, the toss might be going for a hidden expansion, in which case you would need some kind of unit outside of your base (which can be tracked down earlier by stalkers) unless you want to use more scans. Counting Stalkers as proposed before isn't reliable, a good player might not show the true Stalker count on purpose, and hide them elsewhere. It happened to me before, that you think the protoss is stopping the aggression and expanding, due to lack of units. Once you salvage your bunkers and move out, you then get surprised by hidden units.
Last, even if you hold against blink stalker, as mentioned earlier you can only really scout if the Protoss is getting a third, but not the tech he chooses to follow up his blink play. Even if you survive, there are many options that can catch you offguard, like a DT followup. This results in extra caution the Terran player has to take, and hence in more defensive investment (or in the worst case a completely wrong unit mix). Because of this, even if you hold the Blink Stalker allin, you might be behind in the macro game that follows.
If you do mistakes in any of those stages (pre- allin, during the allin, and transitioning out of it), you can lose the game, even if you do good in the others. That's why on non-pro level especially, I personally think that builds like the Blink stalker allin are so strong and frustrating to play against.
This. How to defend Blink all-in or pressure is trivial and Terran will come out ahead. How to gain the scouting information in particular how much Protoss commits and when and how Protoss transitions, on the other hand, is difficult thanks to the nature of fog of war.
Well, I've already suggested leaving an scv or something to spot a third behind this push at the very least. Having an extra reaper out also makes scouting any transitions easier. If the protoss has nothing at home though, even an scv or marine should be enough to see if he's teching. If you do see something defending his base, then even though you can't see what kind of tech path he chose, you know he's teching.
You guys are making it sound much harder than it actually is. It's definitely harder to scout the push coming than it is to send an scv or marine to look for tech structure when there should be literally no units at home (if it's a real all in).
could you add a location for where we terrans should place our production into the attack path pictures? that would be awesome! im never sure where i should be putting my rax eg on Polar night.
I haven't played very much or watched many of the pro-matches. But from my limited amount of experience with it I think the best way to deal with it is just playing with superior reaction, micro and great multi-tasking drops to follow it up. (there are several maps like Heavy Rain that seem to be unsolvable though) Theoretically, the best way would be through revolutionary building placement and bunker placement but I don't see it ever coming to fruition before blizzard changes the map pool enough to negate the play significantly.
It is true that Terrans have placed less significance in building placement than Protoss players for around a year. If building placement was more developed, I bet blink builds won't be as big of a problem as it is today. (Like first few barracks, supply depot, and ebays closely grouped around the mineral line with bunker placements close to the CC and covering significant amount of the vulnerable space) Instead of building to keep zerglings from getting superior surface area, build to expose minimal area to stalker fire and maximize maneuverability of unstimmed bio and scvs to get into repair positions.
In practice, I am still trying to defend oracle builds and scv pulling and hope that the opponent is makes a mistake in his micro.
The extra reaper and marine are hard to afford, but the scv idea has potential for sure. It's quite tough execution wise because during that phase of the game the Protoss will have his stalkers on the map, clearing towers and securing proxy pylon spots at least, but it's definitely doable.
Ive been following this thread for a few days... and I see on many tourney the 2base blink...
Im now currently plat (was 8x masters on eu, 3x masters on cn and 2x masters on kr). When im laddering... I rarely see any perfect blink openings. Its probably I always hijack their gas at the 11th scv. 3-4 sec before the protoss takes their gas when you arrive at their base.
If they manage to get their second gas... I hijack their expansion (a bunker if i dont wanna expand fast or a ebay).
Of course this will mess up your own build order... but if you delay their BO with more then a miniute your own BO will be only delay about 30 sec... I dont spend my time try to counter the BO.. I just rather mess their timings up :D
Has anyone tried to go for a fast 4 base expand by 5 minutes? So you don't make any units for a while but then by the time the push comes at around 20 mins you have maybe 50 cannons everywhere. I'm only in bronze but I can see this being extended to the pro level
I'll grant you that some of the super-tough Blink maps (looking at you Heavy Rain) are gone, but it's not like all the new maps shut down Blink a la Habitation Station (which is a really bad Blink map). For example, Overgrowth has two points of possible entry but both are defendable with relatively easy Bunker placement and you only have to bounce between the two points as opposed to cover an entire massive cliff. The other maps are not nearly as easy to execute on as on Heavy Rain, but because Blink-friendly also means Reaper-friendly, I expect to see people still going 2-base Blink and trying very hard to hide it from your scouts.
The thing that kind of sucks about the new map pool is that blink maps are also good maps for reaper scouting, so it becomes more difficult to scout as well. But I guess if there's one less all in to scout then it's a fair tradeoff.