Website Feedback
Closed Threads
IRC Chat
irc.quakenet.org #teamliquid
IRC Web ClientTeamSpeak 3 (100 users) | |
|
 Daigomi South Africa. January 21 2008 05:36. Posts 4302 | Profile Blog # |
Maps in the Balance
by Daigomi
TeamLiquid: Final Edits
Through the last decade of professional Starcraft there has been a plethora of battles fought between all races, and each and every race has gone through a period of suffering- a period of drought where wins were few and far between.
During these times of drought there is inevitably an outcry by those hit hardest by it, and the most common thing heard at such a period is the phrase "imbalanced". The idea of "imbalance" (or "imba") rests on two pillars: The pillar of strategy, and the one of map design.
Much has been said about strategy, and today it is accepted that the races are balanced the way they are. Thus, whenever any match-up is seen as imba, it is usually blamed on the map rather than the races.
What is a map imbalance, and what makes a map imbalanced? The first way in which a map can be seen as imbalanced is if there are, for a specific race, no feasible strategies available that give that race an equal chance of winning, or on the opposite end of the spectrum, if there are strategies available to a race that gives them too great a chance of winning. The second way in which a map can be seen as imbalanced is if a certain race has too few, or too many, feasible strategies available to them on a given map.
With this in mind I will look at the map as a mixture of strategy and geography, and as it is impossible to analyze every geographical feature of a map it may prove to be easier to look at a map in terms of the strategies used on the map and to see why these strategies were effective on the map.
The maps that will be analyzed are the maps from the current Proleague season, and I will look at the features of each map that makes the map either balanced or imbalanced.
Baekmagoji(TvP: 8-16, TvZ 11-3, PvZ 3-7)![[image loading]](http://www.teamliquid.net/tlpd/images/maps/160_Baekmagoji.jpg)
Baekmagoji is not the easiest map to start off with as it is a map created with the goal of deviating from standard map traits. The map has proven to be very imbalanced, but not favorable to a specific race; it can rather be seen as a complicated game of rock/paper/scissors, where each race is strong against one race, but weak against the other.
The geography of the map is straightforward with two starting points connected by three possible routes, including a short direct route. The first, and most notable, difference between Baekmagoji and more standard maps is its placement of resources: The starting base has 18 resource spots, and two vespene geysers. This deviation from standard design greatly affects the strategies available on the map, as standard strategies are based on starting points with 8 resource spots and only one vespene geyser. The next difference in map design is the placement of the natural expansions, in that there are no natural expansions that are easy to secure. The expansions are placed so that the multiple routes allow a player to bypass the expansions and run straight into the main base.
The first imbalanced matchup on Baekmagoji is TvP, with Terran managing to win only a third of the games played between the races on the map. The biggest reason for this imbalance is slightly obscure: The corridor split in front of the chokes, combined with the small corridor, makes it very hard for Terran to launch a decent tank push. There is no room for Terran to push in, so he has to push from one side, which allows a Protoss force to flank him, or launch a counterattack against his base very easily.
As far as a tank push usually gets
Added to this problem is the double gas given to Protoss at the start of the game, allowing Protoss to fast tech to templars and arbiters, and because of the bad geography for Terran, a timing push is not as effective as it should be.
No PvT is complete without an few arbiters
The second imbalanced matchup on Baekmagoji is TvZ, with Zerg only winning three of its first fourteen games. The reason for this was made clear in only the second game played on the map which was between Darkelf and Calm. Darkelf used the extra gas and resources at the starting location to go for a metal build, with which he then continued to slaughter group after group of Zerg's forces.
The reason the metal build is so effective on the map is thanks to a combination of factors. The first factor that influences the balance on Baekmagoji is the ineffectiveness of the natural counter for metal: Mutalisks. Early mutalisk harass is ineffective as Terran only plays from one base on Baekmagoji, making it easier to defend the mineral line. Secondly, Terran gets his metal much faster and much more cheaply because he does not spend resources on a second command center, and he can get his second gas before his second command center would usually be finished. Finally, the direct route between the bases is a short route for ground forces to travel on, which makes it possible for Terran to go from a defensive position to an offensive position in a very short period of time. All these factors make a metal build very effective on Baekmagoji.
And no TvZ is complete without some metal
The last matchup on Baekmagoji is PvZ, a matchup that hasn't truly been tested yet, but which has given Zerg a good winning percentage up to date. The first advantage offered to Zerg in this matchup is the tiny natural expansion inside the main base which has creep covering it. An early expansion to this spot gives Zerg a small initial resource boost.
A further advantage on Baekmagoji for Zerg is the corridors leading up to the choke in front of the main bases- this setup is ideal for Zerg to start an early contain in, allowing Zerg to expand comfortably. The extra gas also allows Zerg to research lurkers quickly. Finally, the closeness of the 12' and 1' expansions, and similarly 5' and 6' expansions, allows Zerg to easily take two expansions early in the game.
A few lurkers can contain Protoss long enough to take over the map
Games to watch:
Blue Storm(TvP: 16-17, TvZ: 14-29, PvZ: 14-24)![[image loading]](http://www.teamliquid.net/tlpd/images/maps/159_Blue%20Storm.jpg)
Another new map that has been added this season, Blue Storm, has gained a quick following- being played in all three major leagues. The map has shown a decent balance in the TvP matchup, but Zerg has come to dominate against both Terran and Protoss. The most striking feature of the map is the great chasm in the middle of the map, dividing the map into two sections. The map has two main routes between the starting locations: A direct route through the chasm making use of a minimum choke at each player's natural, and a winding route that medium and large size units can also use.
While it is not immediately clear why the map favors Zerg in the ZvT matchup, there is one trait that stands out in ZvT games on the map: Map control. The chasm in the middle of the map allows Zerg to safely take the expansions on his side of the map, while fending off any Terran troops trying to cross the divide.
The chasm can easily be defended early in the game...
By making use of a combination of mutalisks and zerglings early in the game, and switching over to a combination of defilers, zerglings, and lurkers, the Zerg player can defend the middle chasm with relative ease.
A further feature is the ledges on the map, especially the ledges behind the natural expansions, which make it possible for Zerg to gain an early advantage, which in turn allows him to take map control before too much pressure is applied to him. These ledges allow Zerg to mutalisk harass with relative ease. In recent games Terran has attempted to counter this disadvantage by going for vultures followed by wraiths, and in some games this counter has worked wonderfully.
A final advantage the map offers to the Zerg player lies in the routing between the higher ground expansions, and the direct route between the two bases. These routes allow zerglings to either bypass or flank small Terran forces, making them more threatening early in the game, which once again helps the Zerg player to control the map.
Zerg uses the minimum choke to do some damage
In the ZvP matchup the Zerg advantage is less pronounced, but there are certain small advantages that add up to a noticeable advantage. The first advantage Zerg has is the position of the mineral only expansion which is close to the natural, and relatively safe. Countering a Protoss FE by taking the mineral only can be done without too great a threat.
A further advantage that the map offers to the Zerg player is the maneuverability offered by the multiple routes. Unlike in the Terran matchup though, these routes offer Zerg the greatest advantage during the late-game where it splits up the Protoss forces. The Protoss player needs to choose between defending one expansion and risk being caught out of position, or spreading his forces to thin and being overrun. The routes, combined with the large area in the main base which is ideal for drops, makes the map very difficult for Protoss to control. Thus maneuverability is one Zerg's greatest advantages on the map against Protoss.
Four bases destroyed in under three minutes
A final advantage Zerg has on the map against both Protoss and Terran has to do with the pathing issues caused by the minimum choke between the bases. While this is not strictly an advantage, and the players can easily overcome it, it adds an extra strain on the Protoss and Terran player's game, which could lead to potential mistakes. The Zerg player overcomes the pathing issues by building most of its hatcheries at the mineral only expansion.
815 shows us how to macro on Blue Storm
Games to watch:
Katrina(TvP: 5-14, TvZ: 12-11, PvZ: 22-11)![[image loading]](http://www.teamliquid.net/tlpd/images/maps/162_Katrina.jpg)
Katrina has one of the worst records of any map currently in use. Terran has managed to win only 25% of its games against Protoss on the map, while Zerg has won exactly a third of its games. The problems with the balancing can be seen almost immediately by looking at the layout of the map. In TvP, Terran's greatest problem is its inability to get a fast push going. With the multiple routes, any Terran tank push can be flanked, or outright avoided. This inability for Terran to pressure effectively allows Protoss to channel the safe gas at the natural into either fast arbiters or fast carriers.
Protoss sets off with his carriers ten minutes into the game
Once carriers are built, the multiple ridges and the distance between bases allows Protoss to safely, and effectively, micro his carriers, while the distance between bases, as well as the size of the main/natural base, makes recall very hard to stop, and allows Protoss to outmaneuver Terran comfortably.
Constant recalls allow Protoss to dominate the map
In the PvZ matchup there are fewer clear advantages to Protoss. Perhaps the biggest advantage Protoss has on Katrina is access to a fairly close third gas, which allows Protoss to amass a corsair fleet (often along with a reaver mass) quickly.
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/KTCorsairs1.png) Corsairs and Reavers go hand in hand on Katrina
The multiple routes around the third expansion are also a problem for Zerg, and a double FE can be exploited by attacking on two fronts, or by circumventing the second expansion. Drops are also dangerous against Zerg, as the distances between the bases make it hard for Zerg to defend all his expansions (while a few cannons and a stormer can easily defend a Protoss expansion). A final problem with resource gathering for Zerg is the exposed mineral only expansions, making them difficult to take and control at any point during the game.
As the game progresses to the late game, a big problem for Zerg is the positioning of all the mineral lines in the middle of the map. Except for the two resource lines per main, every single mineral line can be stormed from an obscured location by a well placed templar. It is practically impossible to protect against all the storm spots on Katrina, which makes the game very hard for Zerg once the middle expansions, as well as the mineral only expansions, become necessary.
A final hurdle Zerg faces on the map is the distance between the starting locations, especially if the players start opposite each other on the map. The distance between the starting locations enables Protoss to FE without any threat, which in turn limits the strategies that the Zerg player has to his disposal. The ramp also makes cannons more effective in defending against early rushes, which forces Zerg into a macro based game where expansions are difficult to protect.
Opposite starting locations and cannons nullifies early pressure
Games to watch:
Un'goro Crater(TvP: 4-0, TvZ: 12-12, ZvP: 2-2)
The last map too look at is Un'goro Crater, a map with only one bad matchup, and the easiest map to find the problem with. The layout of the map is similar to that of any standard macro map: Four main bases, with one easily attainable natural expansion per base. The middle of the map is an open field through which armies can travel to all of the possible bases. While this is a fairly standard setup for a map, Un'goro Crater has one geographical feature that differs from all other standard maps: The center of the map can be built on.
In most matchups this is not an issue, but in TvP this minor difference causes major problems for the Protoss player. Allowing someone to build in the center of the map allows Terran to build turrets in the middle of the map, which allows Terran to start a slow push backed up with Turrets. The addition of turrets to a slow push prevents Protoss from making use of strategies such as Zealot bombs, or Reaver drops, to diminish the Terran player's tank number. The only way for Protoss to stop a tank push is by taking it head on- a strategy that is usually not very effective.
The progression of a slow push with turrets
A further problem for Protoss is distance between the bases, which is uncomfortably short. Terran can begin a tank push in his own base, and reach the Protoss base in a matter of moments, giving the opponent little time to react. The short routes also make it impractical for Protoss to go for any fast Carrier or fast Arbiter builds, as a timing push can move from the Terran player's base to the Protoss player's base in no time. Even if carriers are built though, the map is not suited to carriers as the open center, and lack of cliffs, makes the carriers easy to target and hard to micro.
Turrets can be build from the one base to the other
A final problem Protoss has on the map concerns the ramps into the bases. A few well placed tanks, mines, and perhaps supply depots, can make it near impossible for Protoss to leave or enter a base. This is even worse at the corner expansions where the small ramps, and narrow routes to the ramps, make it easy to set up a contain.
Games to watch:
-----
After examining the four new maps used in the Proleague this season, the interplay between strategy and map balance becomes clear, and it can be seen that no race is simply imbalanced- a race is only imbalanced in the context of the maps that the race plays on.
With that in mind we can see why Terran has been less successful this season. in both the Starleagues and the Proleague. With both Katrina and Blue Storm being played in the ODT qualifiers, and Blue Storm played in the MSL Survivor, Terran has had a much harder time than usual in qualifying.
In the proleague the only players that have been successful were those willing to adapt, and the season saw an upsurge in younger or less experienced players doing well. The top four terran players of the season were: Sea, HiYa, go.go, and Darkelf.
This season has shown that Starcraft is not a game with fixed strategies. It is a game where each race has unique units with unique capabilities that need to be used to its maximum efficiency on each map. Medics and marines are not effective against Zerg on Blue Storm, so it is up to Terran to find strategies that are effective.
Of course there are imbalanced maps. It is easy to think of a map which has no viable strategies for a certain race, but it is also easy to fall into the pitfall of thinking that a map in which the standard strategy does not work, is an imbalanced map.
As long as there are new maps to play on, the best players will be those who push themselves both physically and mentally, and as long as there are players giving their all, there will be fans worldwide keeping this ten year old game alive.Last edit: 2008-01-21 14:15:04 |
|
|
| SpiritoftheTunA United States. January 21 2008 12:50. Posts 5840 | Profile Blog # | |
| | :3 || “So shall it be in the end of the world: the angels shall come forth, and sever the wicked from among the righteous, and shall cast them into the furnace of fire: there shall be the weeping and the gnashing of teeth.” Matthew 13:49-50 | |
|
|
| Titusmaster6 United States. January 21 2008 12:58. Posts 3354 | Profile Blog # |
Nice write up overall, it's definitely true that "imbalanced maps" can create exciting, unorthodox strategies and different executions.
ZvP on Katrina is the bitch, I lost like 5 games in a row. But looking at the replays and reflecting on my play, I found playing on Katrina to be the most exciting even though I lost. The map just makes you think differently and it's definitely one of my fav maps. |
| | Shorts down shorts up, BOOM, just like that. |
|
|
| CDRdude United States. January 21 2008 13:10. Posts 5413 | Profile Blog # |
| Fantasy II and Persona were taken out of the OSL for balance reasons. While I can clearly see why Persona was taken out, could you explain Fantasy II? I see that the TvZ percentage is pretty bad, but it isn't obvious why. |
| | This is my signature. There are many others like it, but this one is mine. | |
|
|
| Felagund Philippines. January 21 2008 13:20. Posts 504 | Profile Blog # |
Thanks for explaining Horse Many Minerals--I mean, Baekmagoji. That's a pretty weird map.
Excellent FE.  |
| | TL CJ Entusman #5 "now she is unarmed, u shall go gather ur army, siege ur tanks and her choke and send some vessels to spot her lurkers, at this time she may have defilers so, if u spot some, unsiege and bring fbs" -Ki_Do |
|
|
 jkillashark United States. January 21 2008 13:30. Posts 5245 | Profile Blog # |
| This was a flipping good read. |
| | Do your best, God will do the rest. 고린도전서 10:31. |
|
|
 mikeymoo Canada. January 21 2008 13:58. Posts 7064 | Profile Blog # |
Good read. It's nice to go into the intricacies of each map, because I play very rarely nowadays.
Quick fix: Caption should read "no TvZ is complete without metal"
 |
| | o_x | Ow. | 1003 ESPORTS dollars | If you have any questions about bans please PM Kennigit | |
|
|
| Carefree United States. January 21 2008 14:04. Posts 1456 | Profile Blog # |
| Nice article. |
| | DebOnAire - 「 Bisu[Shield] 」 |
|
|
HonestTea  January 21 2008 14:26. Posts 5000 | Profile Blog # |
Excellent read.
Each article, and each writer, brings a new wrinkle into the game.
This detailed analysis and discussion about maps and strategy is yet another way to make our watching experience thicker.
Mani talked about the maps in terms of memories and history, and Daigomi gives us the strategic and gameplay side.
Really enjoyed it, I can see you must have worked hours and hours on it. Good writing, too - simple and easy to understand for goons like me.Last edit: 2008-01-21 14:35:27 |
| | returns upon momentous occasions. |
|
|
| ilovejonn Canada. January 21 2008 15:03. Posts 2510 | Profile Blog # |
| Nice analysis on the maps indeed. |
| | Snowflakes in January, Heart warm like February, I wouldn't ordinarily.. |
|
|
 MoNKeYSpanKeR United States. January 21 2008 16:01. Posts 2869 | Profile Blog # |
| nOice, don't blame the race blame the map :D |
| | <3's Mani and Seraphim, thx for the second chance. TSL Name: TSL-mSLeGenD |
|
|
 Waxangel United States. January 21 2008 16:13. Posts 19118 | Profile Blog # |
awesome analysis
was very nice for me, since I don't follow progaming that closely anymore and don't get to know the maps so well.
In general, I think that claiming imbalance for any matchup with a sample size less than 20 is a bit premature. While some maps will stay imbalanced till the end, other maps will pull an Arkanoid and end up reversing itself. |
| | Hey HP can you redo everything youve ever done because i have a small complaint? |
|
|
| FoieGras Canada. January 21 2008 16:54. Posts 270 | Profile Blog # | |
|
|
| samachking Bahrain. January 21 2008 18:05. Posts 4949 | Profile Blog # |
Nice Write up
Another factor I guess to factor in map balance, is to see the number of mirrors on it, if its a Proleague map.
Look at Geometry for example |
| | "And then Earthlings discovered tools. Suddenly agreeing with friends could be a form of suicide or worse. But agreements went on, not for the sake of common sense, or decency, or self preservation, but for friendliness." |
|
|
| besiger Croatia. January 21 2008 18:11. Posts 2440 | Profile Blog # | |
| | A weak will coupled with delusions of grandeur |
|
|
 semioldguy United States. January 21 2008 19:19. Posts 7265 | Profile Blog # |
| sometimes it just takes a while for a player to realize a viable or successful strategy on a map for a given matchup. This was a great read Daigomi, of the newer members, you are one of the best! |
| | I'm such a good lover because I practice a lot on my own. |
|
|
| dybydx Canada. January 21 2008 22:00. Posts 1764 | Profile Blog # |
i propose a standardized method to measure which map is more imba than the other.
let x = min(TvP, PvT)
y = min(TvZ , ZvT)
z = min(PvZ, ZvP)
q = min(x , y , z)
balance meter = x * y * z * 8
the lowest score is 0, and the highest score is 1. lower score = imba
we can also calculate the "hotness of a specific race"
hot meter for terran = [sum over all map of] chance of t win in a map^2 / number of maps x 2
lowest score is 0, highest score is 1. 1 being red hot.
hence, according to my imba meter, (using latest statistics right now on TLPD map info)
Baekmagoji = 0.189841968
Blue Storm = 0.48798288 Most balanced
Katrina = 0.328012272
Un'Goro Crater = 0 Most imba
This means Un'Goro Crater (it the TvP imba continues) has to be removed. It would make sense too i mean if the toss player has no chance of winning this map, it would be very unfair to play a T player in the finals. (assuming other maps are balanced)
|
| |
|
| P_King Russian Federation. January 21 2008 23:44. Posts 363 | Profile # | |
| |
|
 GrandInquisitor New York City. January 22 2008 00:40. Posts 7910 | Profile Blog # |
| Many of the Blue Storm PvT games are proxies by P, I think? |
| | What fun is it being cool if you can’t wear a sombrero? |
|
|
| foeffa Belgium. January 22 2008 00:54. Posts 2056 | Profile Blog # | |
| | 觀過斯知仁矣. Ma Jaeyoon, always #1. Entusman #17. Go HoNtresh! |
|
|
| 1 2 3 Next All | |
|
|
|
| |
|
Sidebar Settings...
|
Store
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/MapsintheBalance_Cover.jpg)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/BMTankPush1.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/BMArbiterAll_S.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/BMMetalAll_S.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/BMContain1.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/BSChasm1.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/BSChasm2.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/BSMinChoke1.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/BSManueverAll_S.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/BSZergHatcheriesAll_S.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/KTCarriers2.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/KTArbiters1.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/KTStartingLocations1.png)
![[image loading]](http://www.teamliquid.net/tlpd/images/maps/161_Un'Goro%20Crater.jpg)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/UCSlowPush2_s.png)
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/Daigomi/UCSlowPush1.png)
