• Log InLog In
  • Register
Liquid`
Team Liquid Liquipedia
EDT 09:34
CEST 15:34
KST 22:34
  • Home
  • Forum
  • Calendar
  • Streams
  • Liquipedia
  • Features
  • Store
  • EPT
  • TL+
  • StarCraft 2
  • Brood War
  • Smash
  • Heroes
  • Counter-Strike
  • Overwatch
  • Liquibet
  • Fantasy StarCraft
  • TLPD
  • StarCraft 2
  • Brood War
  • Blogs
Forum Sidebar
Events/Features
News
Featured News
TL.net Map Contest #21: Voting10[ASL20] Ro4 Preview: Descent11Team TLMC #5: Winners Announced!3[ASL20] Ro8 Preview Pt2: Holding On9Maestros of the Game: Live Finals Preview (RO4)5
Community News
Chinese SC2 server to reopen; live all-star event in Hangzhou21Weekly Cups (Oct 13-19): Clem Goes for Four3BSL Team A vs Koreans - Sat-Sun 16:00 CET7Weekly Cups (Oct 6-12): Four star herO85.0.15 Patch Balance Hotfix (2025-10-8)81
StarCraft 2
General
RotterdaM "Serral is the GOAT, and it's not close" Chinese SC2 server to reopen; live all-star event in Hangzhou The New Patch Killed Mech! Weekly Cups (Oct 13-19): Clem Goes for Four 5.0.15 Patch Balance Hotfix (2025-10-8)
Tourneys
Merivale 8 Open - LAN - Stellar Fest Tenacious Turtle Tussle RSL Season 3 Qualifier Links and Dates $1,200 WardiTV October (Oct 21st-31st) SC2's Safe House 2 - October 18 & 19
Strategy
Custom Maps
Map Editor closed ?
External Content
Mutation # 496 Endless Infection Mutation # 495 Rest In Peace Mutation # 494 Unstable Environment Mutation # 493 Quick Killers
Brood War
General
Is there anyway to get a private coach? BW General Discussion OGN to release AI-upscaled StarLeague from Feb 24 BGH Auto Balance -> http://bghmmr.eu/ SnOw's Awful Building Placements vs barracks
Tourneys
ASL final tickets help [Megathread] Daily Proleagues [ASL20] Semifinal B 300$ 3D!Community Brood War Super Cup #4
Strategy
TvP Upgrades Relatively freeroll strategies [I] Funny Protoss Builds/Strategies [I] TvP Marine Usage
Other Games
General Games
Stormgate/Frost Giant Megathread Path of Exile Nintendo Switch Thread Dawn of War IV ZeroSpace Megathread
Dota 2
Official 'what is Dota anymore' discussion LiquidDota to reintegrate into TL.net
League of Legends
Heroes of the Storm
Simple Questions, Simple Answers Heroes of the Storm 2.0
Hearthstone
Deck construction bug Heroes of StarCraft mini-set
TL Mafia
TL Mafia Community Thread SPIRED by.ASL Mafia {211640}
Community
General
Things Aren’t Peaceful in Palestine Russo-Ukrainian War Thread US Politics Mega-thread YouTube Thread The Chess Thread
Fan Clubs
The herO Fan Club!
Media & Entertainment
Korean Music Discussion Anime Discussion Thread Series you have seen recently... [Manga] One Piece Movie Discussion!
Sports
2024 - 2026 Football Thread TeamLiquid Health and Fitness Initiative For 2023 MLB/Baseball 2023 Formula 1 Discussion NBA General Discussion
World Cup 2022
Tech Support
SC2 Client Relocalization [Change SC2 Language] Linksys AE2500 USB WIFI keeps disconnecting Computer Build, Upgrade & Buying Resource Thread
TL Community
The Automated Ban List Recent Gifted Posts
Blogs
The Benefits Of Limited Comm…
TrAiDoS
Sabrina was soooo lame on S…
Peanutsc
Our Last Hope in th…
KrillinFromwales
Certified Crazy
Hildegard
Customize Sidebar...

Website Feedback

Closed Threads



Active: 1761 users

Full Map Analysis

Blogs > tzenes
Post a Reply
tzenes
Profile Blog Joined March 2010
Canada64 Posts
Last Edited: 2011-03-11 17:40:12
March 11 2011 08:11 GMT
#1
I had originally hoped to collect more data on the Blizzard maps before releasing this, but given that they seem to be phasing those maps out of major tournaments I figured I'd share my findings so that they could be useful for as long as still possible.

The original question I wanted to answer is as follows:

Many tournaments allow losing players to pick the next map, so which map should I pick?

To solve this question I grabbed my graphing calculator (and a spreadsheet) and began to input data. I needed a fairly large source of data to pull from, but I also wanted to avoid any inaccuracies caused by much better players stomping on worse players. While those samples are statistically significant, they indicate the need for a much larger data set. Ultimately, I chose to go with TLPD's data on a per map basis. Some of my more preliminary work can be found here: http://gaming.stackexchange.com/questions/15526/what-are-the-terrans-best-and-worst-1v1-maps/15550#15550

You can find this data here: http://www.teamliquid.net/tlpd/tlpd_select.php

My spreadsheet is here: https://spreadsheets.google.com/ccc?key=tsdNUMXQu6hIurnQigHkGTQ&authkey=CMX1050H#gid=0

Almost immediately you'll notice that no map is considered "Balanced." As a threshold for Balanced/Imbalanced I decided to require a statistical 95% confidence. Many scientific journals require P < .1 or P < .05 (90% and 95% respectively), so I feel this is a good number. As you can see, there is no map which has a P value greater than .95, thus there is no map which I am 95% or greater confident that it is balanced.

Based on this, the remainder of the data falls into two categories: Maps which may or may not be balanced, and maps which are imbalanced.

A closer look at the data will also reveal the following fact: Zerg's lose a lot. Based on my sample data ZvP is 47.6% win rate and ZvT is 44.1% win rate. Initially I assumed that a Balanced map would have a 50% win rate for all races, but the result is that all maps came up imbalanced against Zerg. As a result I had to recalculate my data based on the observed Zerg win rate. When we take that in to account we see there are a number of maps where Zerg have an unusually high win rate (though not necessarily better than 50%). One could argue that means all maps are imbalanced against Zerg, only some less so, but that sort of argument is irrelevant in answering my original question. I only seek to identify the maps which are most likely to provide an advantage from the pool.

With out further ado here is your tl;dr; the following maps are most likely Imbalanced:

1) TvZ Xel'Naga (favoring Terran) at 99.97% confidence
2) ZvT Shakuras (favoring Zerg) at 99.96% confidence
3) ZvP Blistering Sands (favoring Zerg) at 99.92% confidence
4) PvT Xel'Naga (favoring Protoss) 99.8% confidence
5) ZvT Blistering Sands (favoring Zerg) 98.6% confidence

Probably the most shocking map on this list is Xel'Naga which favor's Terran against Zerg and Protoss against Terran (PvZ is not significantly different from standard). There isn't a lot of literature that points to Zerg doing poorly on Xel'Naga, but is it the most likely map to be imbalanced against them.

Another interesting map is Blistering Sands. It seems in both match ups it favors Zerg. This is strange as Blistering Sands is frequently quoted as a map with a fairly short rush distance, but it turns out that distance is still fairly long (~40 seconds), and an easily defensible natural seem to put this map more in favor of the Zerg. Another worthwhile point to notice is the ease of Overlord scouting. The horse shoe shaped main, leaves many angles for an overlord to squeeze in.

A map which didn't make the list (but deserves honorable mention) is Metalopolis. While still not a great map for Zerg the confidences were 9.85% and 12.1% (for PvZ and PvT). Though I can't say with statistical certainty that this is a bad map for Zerg, gut instinct tells me that those numbers are very low. I've written about Metal before (http://www.teamliquid.net/forum/viewmessage.php?topic_id=179183) though my original measurements assumed that Zerg should win 50% of the time (instead of the 45% of the time they win on average across all maps).

Because no map had a P value greater than .95, I can't say with statistical certainty that any map is Balanced, however, I can list the maps which are more likely balanced:

1) PvT on Blistering Sands (86.5% confident)
2) ZvP on Shakuras (82.5% confident)
3) ZvP on Lost Temple (74.0% confident)
4) PvT on Metal (73.2% confident)


Edit: here is the data for patch 1.2 only: https://spreadsheets.google.com/ccc?key=0Am57dGT4LkjRdHVWcDMxLXFqSlFVOVJfS3hwazluZHc&hl=en&authkey=CJCVoeUL


jalstar
Profile Blog Joined September 2009
United States8198 Posts
March 11 2011 08:31 GMT
#2
I tried to do this a while ago but you need to specify date ranges. Start with 10/14 to get only patch 1.2 data and then go even later for some maps.

TvZ on Steppes was 39-35 from 1.2 release to the end of 2010, and 19-7 in 2011. So clearly trends change very quickly, making this analysis very difficult.
jalstar
Profile Blog Joined September 2009
United States8198 Posts
March 11 2011 08:36 GMT
#3
http://www.teamliquid.net/tlpd/details.php?section=sc2-international&type=maps&id=222&part=games&matchup=all&league=any&from_year=2010&from_month=12&from_day=3&to_year=2011&to_month=3&to_day=3&action=Update#tblt-2883-1-1-DESC

Also there are hundreds of games with no map data. If you look at the last 3 months you see a pretty clear T > P > Z = T on the unknown data at least, I'm guessing it doesn't change much if you add known maps.
tzenes
Profile Blog Joined March 2010
Canada64 Posts
March 11 2011 17:40 GMT
#4
On March 11 2011 17:31 jalstar wrote:
I tried to do this a while ago but you need to specify date ranges. Start with 10/14 to get only patch 1.2 data and then go even later for some maps.

TvZ on Steppes was 39-35 from 1.2 release to the end of 2010, and 19-7 in 2011. So clearly trends change very quickly, making this analysis very difficult.


So I produced a similar sheet restricted to 1.2:

https://spreadsheets.google.com/ccc?key=0Am57dGT4LkjRdHVWcDMxLXFqSlFVOVJfS3hwazluZHc&hl=en&authkey=CJCVoeUL

Ultimately we see little difference in the data (though I am not confident that Metal is Balanced in PvT).

The top 4 unbalanced maps remain unbalanced.

With less data 1 map does fall out of that category and you can no longer be sure if it is unbalanced (though the probability that it is remains high).

Ultimately less data gives me less power to say things, but it doesn't not significantly change the balance picture
nosliw
Profile Blog Joined December 2008
United States2716 Posts
March 11 2011 17:45 GMT
#5
awesome stats work guys!
Treehead
Profile Blog Joined November 2010
999 Posts
March 11 2011 21:24 GMT
#6
This definitely answers the question you set out to ask: assuming your opponent does something "normal", which map is the easiest to beat him on?

I want to point out, for people who skipped straight to the analysis that this doesn't answer another question: which maps inherently favor one race over another?

Why? Your numbers reflect that what certain races are doing currently are not effective on the maps you've listed. Your numbers do not reflect a wide variety of strategies, however, as most of the time people use one form or another of "standard" strat. Maybe zealot/carrier is amazing on Xel'Naga Caverns. Maybe it isn't - you never really see that type of play from competitive gamers, so we don't know.

Once upon a time, a strategy would be good because the only people who could make it to the top used that strategy, in a process similar to natural selection. With how publicized strategies are these days, everyone does what the pros do. Therefore, lots of things which might work are never tried.

I'm just saying, TvP should be balanced - but this doesn't mean MMM v. Gateway/Colossi must be balanced on Shakuras Plateau, just because this happens to be how most people play the matchup right now.
Please log in or register to reply.
Live Events Refresh
WardiTV Invitational
11:00
Group A
WardiTV918
TKL 196
IndyStarCraft 123
Rex93
LiquipediaDiscussion
[ Submit Event ]
Live Streams
Refresh
StarCraft 2
Harstem 291
TKL 196
LamboSC2 159
IndyStarCraft 123
Rex 93
StarCraft: Brood War
Britney 38106
GuemChi 4027
EffOrt 1161
Shuttle 1132
Flash 1114
Larva 527
Mind 442
Hyun 362
Light 322
Mini 245
[ Show more ]
Leta 199
Rush 186
Pusan 94
firebathero 79
Sea.KH 76
JYJ65
Mong 64
ToSsGirL 51
Shinee 47
Sharp 41
zelot 40
Backho 29
Icarus 29
JulyZerg 22
ajuk12(nOOB) 15
Terrorterran 13
Noble 12
Dota 2
Gorgc4747
qojqva1907
Dendi629
420jenkins307
XcaliburYe167
Counter-Strike
olofmeister2523
allub268
oskar148
Other Games
singsing2220
B2W.Neo427
crisheroes387
Lowko343
KnowMe332
Sick236
Happy231
Liquid`LucifroN212
Fuzer 118
Trikslyr29
Chillindude25
Organizations
Counter-Strike
PGL282
StarCraft 2
Blizzard YouTube
StarCraft: Brood War
BSLTrovo
sctven
[ Show 12 non-featured ]
StarCraft 2
• StrangeGG 56
• AfreecaTV YouTube
• intothetv
• Kozan
• IndyKCrew
• LaughNgamezSOOP
• Migwel
• sooper7s
StarCraft: Brood War
• BSLYoutube
• STPLYoutube
• ZZZeroYoutube
Dota 2
• C_a_k_e 1345
Upcoming Events
Online Event
2h 26m
RSL Revival
12h 26m
RSL Revival
20h 26m
WardiTV Invitational
21h 26m
OSC
1d 1h
SKillous vs goblin
Spirit vs GgMaChine
ByuN vs MaxPax
Afreeca Starleague
1d 18h
Snow vs Soma
Sparkling Tuna Cup
1d 20h
WardiTV Invitational
1d 22h
CrankTV Team League
1d 23h
BASILISK vs Streamerzone
Team Liquid vs Shopify Rebellion
Team Vitality vs Team Falcon
RSL Revival
2 days
[ Show More ]
Wardi Open
2 days
CrankTV Team League
2 days
Replay Cast
3 days
WardiTV Invitational
3 days
CrankTV Team League
3 days
Replay Cast
4 days
CrankTV Team League
4 days
Replay Cast
5 days
The PondCast
5 days
CrankTV Team League
5 days
Replay Cast
6 days
WardiTV Invitational
6 days
CrankTV Team League
6 days
Liquipedia Results

Completed

Acropolis #4 - TS2
WardiTV TLMC #15
HCC Europe

Ongoing

BSL 21 Points
ASL Season 20
CSL 2025 AUTUMN (S18)
C-Race Season 1
IPSL Winter 2025-26
EC S1
Thunderpick World Champ.
CS Asia Championships 2025
ESL Pro League S22
StarSeries Fall 2025
FISSURE Playground #2
BLAST Open Fall 2025
BLAST Open Fall Qual
Esports World Cup 2025
BLAST Bounty Fall 2025

Upcoming

SC4ALL: Brood War
BSL Season 21
BSL 21 Team A
BSL 21 Non-Korean Championship
RSL Offline Finals
RSL Revival: Season 3
Stellar Fest
SC4ALL: StarCraft II
CranK Gathers Season 2: SC II Pro Teams
eXTREMESLAND 2025
ESL Impact League Season 8
SL Budapest Major 2025
BLAST Rivals Fall 2025
IEM Chengdu 2025
PGL Masters Bucharest 2025
TLPD

1. ByuN
2. TY
3. Dark
4. Solar
5. Stats
6. Nerchio
7. sOs
8. soO
9. INnoVation
10. Elazer
1. Rain
2. Flash
3. EffOrt
4. Last
5. Bisu
6. Soulkey
7. Mini
8. Sharp
Sidebar Settings...

Advertising | Privacy Policy | Terms Of Use | Contact Us

Original banner artwork: Jim Warren
The contents of this webpage are copyright © 2025 TLnet. All Rights Reserved.