I would like to remind people how AWESOME Blizzard and David Kim are: - They keep working on the game, even 2 years after the release. How many companies do this? - They gave us a great RTS. Good luck finding a better RTS that is not SC BW.
Meanwhile we bring them nothing. We already paid 50 euros for the game (lolz, that's 3 night's out) and we are still having fun with it!
Apparently, I am one of the precious few on TL who agree with David Kim and his perception of balance.
First off, the stats that they are using, while not terrific, are clearly more involved then just a lump sum of ladder games. It is disappointing that they don't share the details behind the stats, but frankly I don't care that much.
I'm also constantly surprised with how people overstate the problem of imbalance. It's not like I'm looking at the next MLG, or IPL and thinking "well a terran is going to win that tourny." At specific moments in individual games I sometimes think one race has a great advantage over the other, but it isn't a pervasive problem. The late game PvT in particular is such a bizarre point to talk about balance. The late game of every match is burdened by the collective decisions and mistakes of both players so it is never transparent to talk about. Sure, an army of 3/3/1 Chargelots with a handful of stalkers, a couple of archons and 3 well placed storms is fuckin' badass; it is especially so when you got a dozen warpgates to reinforce at the proxy pylon out side the terrans base. But then again, toss can't dodge emps, don't have nearly the drop utility, can't heal, and have units that cost more. Combine that with the fact that a bio ball with Ghosts favours (demands) excellent micro, the protoss army favours positioning (zealots coming at you from all sides is scary) and decision making. Oh, and lets not forgot how much income a player has at this point is entirely dependant on how well expansions were managed (to steal a day9 phrase) in the early game. The late game balance is neither clear nor simple to me.
Finally, I have no problem with one race being advantaged at different points in the game. It adds extra tension to earlier parts of the game and forces players to plan game plans around different times. Games of varied length are very entertaining to me and I think lead to more diverse strategies overall. This was my experience in watching broodwar, and I don't think it'll be any different for SC2.
Overall, I think the game is entertaining and balanced. Whatever discrepancies exist I'm sure the meta-game can cope with them. Plus, HotS will add a bunch of units and fuck everything up again. If you do think it is broken, it won't be broken for that long.
People need to stop saying it is balance in TvP. It's not. Terran win 70% before 20min and Protoss win 70% after 20 mins will makes the matchup looks like 50% but this is not how the game should be.
Marine is Terran strongest unit but it might be Terran's biggest weakness. Bcos of Marine so OP, other Terran unit regrettably have to be weak..
On April 27 2012 18:18 Snowbear wrote: I understand that zergs need better scout options, but what about zvt lategame? Broodlord infestor corruptor is EXTREMELY cost-efficiënt, and I NEVER saw a terran that was NOT ahead beat it.
So what are you saying, Terrans that are behind gamewise late game are suppose to beat a tier3 zerg army?
Probably that terrans who are even or slightly ahead shouldn't just roll over and die if they cannot drop 4 places at once while microing vikings against fungal
Snowbear Korea (South). April 27 2012 19:13. Posts 1771
On April 27 2012 19:11 jeffvip wrote: People need to stop saying it is balance in TvP. It's not. Terran win 70% before 20min and Protoss win 70% after 20 mins will makes the matchup looks like 50% but this is not how the game should be.
If terrans win 70% of their games vs protoss before 20 min, then protoss players would only win 30% after it, not 70%. Early-mid game comes before lategame.
On April 27 2012 19:11 jeffvip wrote: People need to stop saying it is balance in TvP. It's not. Terran win 70% before 20min and Protoss win 70% after 20 mins will makes the matchup looks like 50% but this is not how the game should be.
If terrans win 70% of their games vs protoss before 20 min, then protoss players would only win 30% after it, not 70%. Early-mid game comes before lategame.
Wrong.
Lets take a theoretical sample size of 200 games and say the win rate is 50/50, that means each race wins 100 games. If terran wins 70% of their games before 20 min then terrans win 70 games out of 100 before min 20. If protoss wins 70% of their games after 20 min, that means toss wins 70 games out of 100 after min 20.
For the record he probably used some fictive numbers to prove his point, but a statistical analysis from all the games of MLG Providence 2011, proves his point. The win rates for terran drop dramatically in both TvZ and TvP after a certain point.
Now I wouldn't mind if a race enters any stage of the game with a 5 to 10% advantage, but too much of an advantage skews the risk reward factors into dangerous areas, and it leads to predictability, tedium and frustration, for both players and spectators.
Last edit: 2012-04-27 19:22:21
What am I without a challange? An opportunity to better myself? A chance to shoot for the stars? An struggle for perfection? I am nothing, I live for the challange.
On April 27 2012 19:11 jeffvip wrote: People need to stop saying it is balance in TvP. It's not. Terran win 70% before 20min and Protoss win 70% after 20 mins will makes the matchup looks like 50% but this is not how the game should be.
If terrans win 70% of their games vs protoss before 20 min, then protoss players would only win 30% after it, not 70%. Early-mid game comes before lategame.
Pls don't post some shallow reply. I believe u understand what I meant.
Marine is Terran strongest unit but it might be Terran's biggest weakness. Bcos of Marine so OP, other Terran unit regrettably have to be weak..
On April 27 2012 19:11 jeffvip wrote: People need to stop saying it is balance in TvP. It's not. Terran win 70% before 20min and Protoss win 70% after 20 mins will makes the matchup looks like 50% but this is not how the game should be.
If terrans win 70% of their games vs protoss before 20 min, then protoss players would only win 30% after it, not 70%. Early-mid game comes before lategame.
he obviously meant from the games, that last under/over 20 minutes terran wins x%.
Well, clearly, the nature of the stats matter. When I mentioned "stats" in that previous post, I only meant winrates, and I see winrates as valid because they are the purest statistic available to see who's actually winning the most in the big picture. Certain types of stats are misleading because they filter out the nitty-gritty information on which race is actually winning overall. As an extreme example, if I took the number of Terrans that won the GSL, compared it to the number of Protosses that won the GSL, and uses that "stat" to whine about Terran being overpowered, that would be ridiculous because it fails to take winrates in the big picture into account. Similarly, the advancement rate you bring up fails to clearly take into account winrates, and it also fails to take into the account the fact that the majority of players at the start of Code S this season were Terran. If we had a Code S with 31 Terrans and 1 Protoss and the Protoss made it to the finals, the Protoss advancement rate would be 100%, but that doesn't really mean much.
Let me be absolutely clear by stating my belief that, in the end, all that matters is who wins. If you're winning a majority of the time, you have no business whining about balance, because in the end, despite your complaining, you're still able to win a majority of the time. I also assume that top players of all races are more or less nearly equal in "skill" (putting similar amounts of practice in on the team house schedules, etc.), because if you don't make that assumption, then sure, stats are meaningless in discussing balance. Therefore, winrate stats are extremely pertinent to judging balance.[/QUOTE]
I agree, it does not matter how you win nor how you lose, what matters is how many times you win over a preiod of time.
I also think using GSL code S, ro16 or ro8 as your primary argument is too simple. A players does not advance in GSL only because he plays a certain race but also who he plays against, daily form and dumb luck. The sample size for a single GSL season is too small to draw conclusions important enough to balance the game.
Statistics is an awesome tool to mine useful data over longer periods of time. I do think however that lower league statiscts are irrelevant due to battle.net matchmaking system always will adjust players winning rates towards 50% meaning the good statistics are a result of the matchmaking system and not the balance, thus, the statistics should only be applied in the top layers of leagues where the matchmaking system can't effectively adjust a players winrate towards 50%
Seems to be an announcement with the implication that they will now focus most of their resources on HotS and shifting some away from maintaining and monitoring WoL.
✉ Tweets @sc2channel ⌦ Blog: http://www.teamliquid.net/blog/JOJO ⌫ "Arbiterssss... build more arbiterssss." Click 'Profile' for awesome shiro art!
On April 27 2012 18:25 Laserist wrote: One 'balance' word at context then full of terran whine. How could you forget GomTvT days.... Jeez Please stop whining and practice.
That proves my point. The graph you showed us points out that there is no point of whine discussions. Unless someone can say that he/she outmacroed, outmicroed, have better multitasking and better in every aspect and stilll loose the game, there is no point of whining. Even if someone would have said that kind of thing, it would be highly distorted by subjectivity.
TLDR, no whine just practice
To be fair. Those are tournament statistics and doesn't show anything that is going on in the ladder. Plus the data samples that the korea has is too small to draw any kind of conclusion from it. Then there is that there might be foreign players affecting the korean statistics and koreans affecting international statistics. It doesn't also show how the games were won either by each race. Most likely terran got their wins by all-inning versus protoss players rather than let the game draw out into the late-game. So it isn't that great of an indicator to show how the game is balanced.
If this is true, it would mean that a) your point of lategame imbalance is true b) that the Terran standard strategy should be allin and every Terran who doesn't is a freaking cheeser and only through Terrans being bad (= not playing standard = allin) the MUs looks balanced.
I prefer to think of people having a clue what they play, but maybe I'm wrong and Terrans really don't play their standard strategies and instead try to cheese around with long games.
I wasn't implying it is true or not. Ladder is of course much more different from tournaments. Since people in tournaments play to win rather than for fun. There is also that foreigners tend to hate people who do cheeses and/or all-ins, which leads them to not to focus to play the game that way. You can really see the difference if you watch games in korean ladder where all the races tend to do a lot of all-ins and cheeses across the leagues. While at foreign ladder people are much more likely to play macro games.
I think it's stupid to complain about imbalance if you're not on pro-level because if you are not at that level you are not using using your race's full potential and the problem you have in a certain MU can easily be fixed with better macro and micro.
Brace your're selves. Grammar, nazis are Coming
HolyArrow United States. April 27 2012 19:27. Posts 6306
Well, clearly, the nature of the stats matter. When I mentioned "stats" in that previous post, I only meant winrates, and I see winrates as valid because they are the purest statistic available to see who's actually winning the most in the big picture. Certain types of stats are misleading because they filter out the nitty-gritty information on which race is actually winning overall. As an extreme example, if I took the number of Terrans that won the GSL, compared it to the number of Protosses that won the GSL, and uses that "stat" to whine about Terran being overpowered, that would be ridiculous because it fails to take winrates in the big picture into account. Similarly, the advancement rate you bring up fails to clearly take into account winrates, and it also fails to take into the account the fact that the majority of players at the start of Code S this season were Terran. If we had a Code S with 31 Terrans and 1 Protoss and the Protoss made it to the finals, the Protoss advancement rate would be 100%, but that doesn't really mean much.
Let me be absolutely clear by stating my belief that, in the end, all that matters is who wins. If you're winning a majority of the time, you have no business whining about balance, because in the end, despite your complaining, you're still able to win a majority of the time. I also assume that top players of all races are more or less nearly equal in "skill" (putting similar amounts of practice in on the team house schedules, etc.), because if you don't make that assumption, then sure, stats are meaningless in discussing balance. Therefore, winrate stats are extremely pertinent to judging balance.
I agree, it does not matter how you win nor how you lose, what matters is how many times you win over a preiod of time.
I also think using GSL code S, ro16 or ro8 as your primary argument is too simple. A players does not advance in GSL only because he plays a certain race but also who he plays against, daily form and dumb luck. The sample size for a single GSL season is too small to draw conclusions important enough to balance the game.
Statistics is an awesome tool to mine useful data over longer periods of time. I do think however that lower league statiscts are irrelevant due to battle.net matchmaking system always will adjust players winning rates towards 50% meaning the good statistics are a result of the matchmaking system and not the balance, thus, the statistics should only be applied in the top layers of leagues where the matchmaking system can't effectively adjust a players winrate towards 50%
I completely agree with everything you said, and I guess I was unclear if any part of my previous post implied otherwise.
The sample size in the GSL is indeed extremely small, but since people want to try using GSL occurrences to complain about lategame TvP, I felt it was necessary to bring up GSL stats to at least show that Terrans have won a slight majority of TvPs in the GSL overall. And if we actually wanted to look at a substantial sample size, such as the one used in the tournament winrate graphs released monthly, we'd see that things are still slightly Terran favored in TvP - or at least, they were in March (haven't seen April's stats yet). The bottom line is that I find it irksome that Terrans whine about TvP balance in the midst of stats like this: http://i.imgur.com/VCQcQh.png. Much larger sample size than the GSL, and only includes high-level tournaments. If TvP is imbalanced lategame, then something must be REALLY imbalanced at some other part of the game in favor of Terran for there to be stats like that.
On April 27 2012 19:23 M7Jagger wrote: I think it's stupid to complain about imbalance if you're not on pro-level because if you are not at that level you are not using using your race's full potential and the problem you have in a certain MU can easily be fixed with better macro and micro.
I think most people are complaining about balance they see at the highest level, like what they see from DH, MLG, GSL etc, while many of us play the game, not all play at GM level or higher. I would never dare comment on my experiences of ladder balance but I can comment on observations I make while following GSL for example.
What am I without a challange? An opportunity to better myself? A chance to shoot for the stars? An struggle for perfection? I am nothing, I live for the challange.
On April 27 2012 13:09 avilo wrote: I think at this point I may have to start collecting and casting TvP replays from pro players where the Terran wins the big battles and is up 30-70 supply and protoss still is in the game and ends up winning from hitting psi storms. It's depressing they still do not address TvP lategame after all this time (TvZ lategame isn't much better after the ghost nerf).
It's also disappointing that a balance designer is equating winrate statistics with how the game is currently being played/the metagame. Statistics have NOTHING, absolutely nothing to do with the current metagame.
What about the converse? The part where terran is wrecked early on but survives on its super efficient units, super healing bunkers, and mules?
I think the bitching about TvP late game is mainly that terrans don't like Toss having the ability to apply sustainable and persistent straight pressure on.
The biggest imbalance in TvP lategame is the mass colossus switch. Typically terrans won't have enough vikings to be able to deal with the switch and just get killed. Obviously, prematurely building vikings is a bad idea because they're useless against zealot/storm/stalker/archon and you really wanna be powering everything into bio. Even when scouting the colossus switch, one starport with a reactor isn't good enough to counter a double robo so more starports are required to defend it. Then if the next battle is about even, and you are left with a bunch of vikings (and he doesn't have colossus) he can go straight back into templar/zealot. It's this lategame flexibility which Terran struggles to deal with. And I say this as a protoss who has been exploiting this fact for months now.
I suspect that the winrate on cloud kingdom is so heavily in protoss's favour because CK tends to produce long games, and Terrans tend to do worse vs protoss in long games.
I really dont understand how Korhal is so heavily protoss favoured in ZvP...
I don't think I've ever seen PartinG win a PvT with an end-game Collosus switch. Second of all, for a Collosus switch late-game to work, you have to drop 2-3 Robo's down. If Terran fails to scout that and fails to keep tabs on the Protoss army, they deserve to lose. There is plenty of time to react to get vikings out. I disagree a lot with your gesture of Collosus being the difference-maker late-game. I think it has a lot more to do with the mass amounts of HT Protoss can afford to spread around the map and soak up energy when they get on 4+ bases late-game. Collosus are seeming pretty terrible in PvT as of late.
Last edit: 2012-04-27 19:31:23
"Who hired this awful fountain gunner? He can't hit shit." - Yiss
Finally some bloody thought on the scouting of early game Z vs T the stupid marine kills the overlord and then the fricken helions kill the lings... darkness ensues.
Plexa did raise a good point though, while indeed many more games lately have been ended by good use of storms, a lot of games have also come down to fast and effective switches between Colossus and HT.
In the last game of Oz vs Maru in the GSL RO16, Maru lost after Oz switched into Colossus off double robo and Maru didn't have the resources or time to counter effectively.
I'd also like to add that this is issue also exists in TvZ late game, tech switches from BL to Ultralisk, and back again hurt Terran massively due to a lack of a flexible and strong late game army that can deal with both compositions.
Last edit: 2012-04-27 19:37:21
What am I without a challange? An opportunity to better myself? A chance to shoot for the stars? An struggle for perfection? I am nothing, I live for the challange.
On April 27 2012 18:42 aintthatfunny wrote: Lol at the amount of terran whine
Watch this:
No offense but I'll believe and trust in what he says when he is in Code S and beating nearly every protoss he encounters, until then it is just theory, and it doesn't always sync with reality.
I don't know what stats and statistics David Kim studies, but they seem radically different from what we've seen on monthly win rate reports we get from TLPD. I'm really worried because Blizzard seems disconnected from the situation at the highest level of play.
Instead of focusing on GSL, GSTL, KSL level results and stats they seem to insist on balancing it out using ladder stats, which makes absolutely no sense given that a game on Master's level is still drastically different from a game of players at the highest level.
He also fails to take into account or even address the issue viability and strength of certain strategies or races at certain stages of the game.
I don't have much faith for their balancing or even data collection method at this point.
He doesn't have to be in code s to be true. BUT I want to see the style he describes be practically proven. First: It is very hard to play. Second: It's really hard to get to your fourth base in TvP. I've seen Creatorprime crushing the best terrans in the world with a 3-3 coloss/archon timing. The Problem is, the lategame composition he describes is very expensive for terran. espcially the ghosts are. ghosts cost 150 rescources per supply, vikings cost 112,5. Colossus costs 83, Templar/archon 100. Upgrades are cheaper too for protoss and protoss needs fewer upgrades until the very late game, since terran also has to upgrade ship weapons. The second weakpoint of his theory is that he assumes, that the terran hits every emp. 2 or three storms are enough to simply kill a mostly marine/ghost based army. In theory, you can argue, this should be possible. But the reality just shows us, that NO terran in the world is capable to prevent storms from getting off in the lategame. And this is the point of being too theoretical. You need strategies that are executable in real situations. Even to get to the point where you can afford 7-8 macro OCs, which are about 4k minerals, is really hard. Because the Protoss is willing to trade with you, once he gets his composition together, because his is cheaper, and therefor he reaches his one sooner. And even if he is in the attackers disadvantage, he will trade ok with you (if you are out in the open) or just contain and massexpand (if you are setting up a too strong defence on 3bases). It's an interesting view at the game, but way too theoretical. And the big point here is, it is way too hard to execute this style as terran, since one mistake usually leads to instant defeat.
Store
WCS
SC2Links Mobile App …
M7Jagger Sweden. April 27 2012 19:23. Posts 237
