World Championship match stats

This article contains the breakdown of map and faction stats for the World Championship tournament. The data on the meta is taken from 107 replay files that are available on the related discord server. These cover six matches from the round-robin stage and most replays from the final round.

Some replay files are missing, most notably from the round for 3rd place between Jove and VonAsten. In total I found casts for seven more games on youtube, but only picked out faction, map and commander pick statistics from these. Thus, these stats are based on 114 games in total.

The meta discussions will often compare the pick rates to those from the last big tournament, GSC2, for which the numbers can be found here.

If you have specific questions, please post to the thread, I might be able to pull out the numbers in question.

Index

• General Faction and Map Stats
  
  
  • General Faction Stats
    
  • Map Stats
    
  
  
• Faction Meta
  
  
  • Ostheer
    
  • Soviets
    
  • OKW
    
  • US Forces
    
  • UK Forces

General Faction and Map Stats

This section contains pick and win rates for the different factions and maps. Unfortunately, the replays do not contain any information about who won. So this is something I have to enter into a list manually which is error prone. Let me know if you think something could be off. The brackets can be found here.

There is data available for 114 games, which is significantly less than what we had for the main round of GSC2, where we had data 351 games in total. This means that any calculation of percentages can be considered semi representative at best.

General Faction Stats

This section contains stats on faction picks, wins and matchups. While these stats might give us an idea about the general state of balance in 1v1 to some extend, it should be noted that any the numbers should be taken with a grain of salt, for various reasons:

• Most games are decided by the skill gap between opponents (see below).
  
• We look at a small number of matches.
  
• The games are played on a subset of maps only. Performance on other maps might be very different. Also, starting positions were fixed, so wins or losses might be related to that, despite the valiant efforts to improve the maps prior to the tournament.
  
• Previous tournaments showed that each individual player has certain strengths and weaknesses when it comes to handling different factions.
  

One thing to note is, that most of the games here are from "best-of-5" rounds, while many of the older tournaments were using a "best-of-3" format. This is relevant because very often there is a decisive skill gap between the opponents and the round will be decided by either a 2:0 or 3:0 sweep. In a 2:0 sweep, this means that the better player won both with an axis and allies faction. This changes in a "best of 5", where the winner will have to play twice for either axis or allies.

If we assume, that there is a consensus that a certain faction is strongest, it still will come out with only a minor advantage in "best-of-3" format as in most rounds the faction will have won and lost in most rounds being played by the better and the worse player. In a "best-of-5", however, the stronger faction will come out with a win rate at about 66% if all rounds end in 3:0 sweeps.

Now, in reality things are more complex due to the influence of maps, a players faction preference and the fact that we have five factions in total, but the point stands that stats for "best-of-5" tournaments are more likely to show uneven win percentages than "best-of-3" ones.

You can find a lengthy discussion on how reliable the values are in the later parts of this post, which is based on the discussion of earlier stats for the tournament for 84 games that user eykei posted on reddit (yep, I'm late).





On the axis side, OKW was played significantly more often than OH, unlike in GCS2 where practically both factions were picked at about the same rate. Soviets were always the most played faction on the allied side, being picked more than twice as often as the second most popular faction (USF in GSC2). In WC, however, there were about the same amount of games where the players picked USF compared to the SOV. UKF was generally not big in past 1v1 tournaments, and basically did not happen in WC.

In terms of win percentages we see that SOV and USF overall performed best. Both axis factions ended up with a below 50% win percentage with OH winning barely more than a third of the games; UKF fared even worse.





Looking at specific matchups, we see that OKW performs about even against both SOV and USF. OH on the other hand lost most matches against these two factions. However, there are only very few OH vs USF games, so the win percentage for OH has a relatively large uncertainty to it.

Let us check out the faction stats of the four semifinalist. The first number in the tables below indicates the number of games played in a certain matchup, the second number says how many of those the player won:

Luvnest vs. ostheer vs. soviets vs. okw vs. usf vs. ukf ostheer 1/1 2/2 soviets 5/5 6/6 okw 4/3 4/2 usf	Noggano vs. ostheer vs. soviets vs. okw vs. usf vs. ukf ostheer 2/1 soviets 2/2 okw 4/1 3/3 1/1 usf 3/3 5/3
VonAsten vs. ostheer vs. soviets vs. okw vs. usf vs. ukf ostheer soviets okw 5/2 4/4 usf 1/0 6/3	Jove vs. ostheer vs. soviets vs. okw vs. usf vs. ukf ostheer 1/1 1/1 soviets 6/4 okw 2/1 3/1 usf 2/2 3/2

There are notable differences between the players: Luvnest almost exclusively played Soviets on the allied side. VonAsten only played USF and OKW. Noggano mostly picked USF and OKW as well, but also has a few matches as Soviets and OH. Jove has a fairly even split between USF and SOV, but - like the others - prefered OKW in most games.

This again shows that even among the top players it is not as if there is a general consensus on e.g. which allied faction is the strongest and personal preference certainly plays a role. Given that a lot of rounds were purely decided based on the individual player skill, these choices then also have an impact on faction win percentages.

Map Stats

The map pick process was different compared to most previous tournaments. Rather than players picking or vetoing maps individually, the maps to play were actually set in advance.

While Crossing in the Woods was left out, Langreskaya made it back into the map pool. Nexus was used for the first time in a larger tournament.

I did break down the stats into matchup specific values, but the numbers are not overly meaningful due to the even smaller sampling size. Still, some tidbits to be take with loads of salt:

• Langreskaya appears to be a very balanced map. However, it might appear that way because it was mostly played in earlier stages of the tournament, where 3:0 sweeps were more common. Still, it is the only map on which OH had a positive win ratio.
  
• Kholodny Ferma also gave relatively even win rates, potentially with a slight edge for SOV.
  
• Faymonville features reasonable win percentages as well; games on Nexus went similar. However, OH's performance on these maps was lacking.
  
• On Crossroads, USF and SOV did particularly well. OH did very poorly; there was only one OH win on Crossroads out of nine games with Aimstrong defeating CreativeName's USF.
  

Map Played

Crossroads	23
Faymonville Approach	27
Kholodny Ferma (Summer)	24
Nexus	18
Langreskaya	22

Crossroads

Faction Pick Won Won / % ostheer 9 1 11% soviets 7 5 71% okw 14 6 42% usf 15 10 66% ukf 1 1 100%

Matchup Games Axis won Axis won / % ostheer/soviets 2 0 0% ostheer/usf 6 1 16% ostheer/ukf 1 0 0% okw/soviets 5 2 40% okw/usf 9 4 44% okw/ukf 0 0 -

Faymonville Approach

Faction Pick Won Won / % ostheer 9 3 33% soviets 13 6 46% okw 18 10 55% usf 13 7 53% ukf 1 0 0%

Matchup Games Axis won Axis won / % ostheer/soviets 5 1 20% ostheer/usf 3 1 33% ostheer/ukf 1 1 100% okw/soviets 8 6 75% okw/usf 10 4 40% okw/ukf 0 0 -

Kholodny Ferma (Summer)

Faction Pick Won Won / % ostheer 6 3 50% soviets 12 8 66% okw 18 8 44% usf 11 5 45% ukf 1 0 0%

Matchup Games Axis won Axis won / % ostheer/soviets 2 1 50% ostheer/usf 4 2 50% ostheer/ukf 0 0 - okw/soviets 10 3 30% okw/usf 7 4 57% okw/ukf 1 1 100%

Nexus

Faction Pick Won Won / % ostheer 9 3 33% soviets 8 5 62% okw 9 5 55% usf 8 5 62% ukf 2 0 0%

Matchup Games Axis won Axis won / % ostheer/soviets 4 1 25% ostheer/usf 4 1 25% ostheer/ukf 1 1 100% okw/soviets 4 2 50% okw/usf 4 2 50% okw/ukf 1 1 100%

Langreskaya

Faction Pick Won Won / % ostheer 8 5 62% soviets 12 6 50% okw 14 6 42% usf 7 4 57% ukf 3 1 33%

Matchup Games Axis won Axis won / % ostheer/soviets 5 3 60% ostheer/usf 0 0 - ostheer/ukf 3 2 66% okw/soviets 7 3 42% okw/usf 7 3 42% okw/ukf 0 0 -

Back to Index

Faction Meta

The following section contains the meta as extracted from the replay files. Extracting teching choices and build orders (BOs) from replays is tricky because those note only when the command to build a unit was issued, not when it arrived on the battlefield. Likewise, while 'cancel' orders are part of the replay files, but vault (the backend to cohdb) that is used to decode the replays does not define which build was actually canceled, so there is no guarantee that the BOs below are 100% accurate.

The big red "X" in any build order indicate that no unit was built as the game ended early.

Player and commander names as well as the numbers in the build orders are colored according to their respective win percentages using this scale:

• 0%
  
• 25%
  
• 50%
  
• 75%
  
• 100%
  

As mentioned in the introduction, I did pick out the commander picks and players played from seven VODs. However, all other statistics including the commander loadouts are derived solely from the 107 replay files.

Ostheer Meta

Ostheer meta is available for 40 games. Additionally, I added the game in which Jove won vs. VonAsten on Kholodny Ferma, using Spearhead against USF.

Players

Ostheer Commanders

Everything in the OH section beyond this point will only contain data from the 40 OH games for which a replay is available.

Commanders vs. Faction Breakdown

Commander in Loadout

24	Mechanized Assault Doctrine
19	Spearhead Doctrine
17	German Infantry Doctrine
16	Lightning War Doctrine
12	Jaeger Infantry Doctrine
11	Osttruppen Doctrine
5	Strategic Reserves Doctrine
5	Elite Troops Doctrine
4	German Mechanized Doctrine
4	Blitzkrieg Doctrine
2	Storm Doctrine
1	Festung Armor Doctrine

The once popular "Mechanized Assault" (MA) is back! While never being completely off-meta, the former favourite of many was the second most used commander, closely behind "Spearhead". It did, however, not fare well in terms of wins.

Together with "Lightning War" and "Strategic Reserves", the Tiger doctrines make up about 59% of OH games. Still, this means that the other non-Tiger doctrines make up for 31% of games. 10% of the games ended before the player choose a doctrine.

The former go-to doctrine, "Mobile Defense" not only was never actually picked, but didn't even make it into any loadout.

Teching

The table below summarizes most teching choices.



The numbers are relatively comparable to those we got for GCS2. There are slightly more games in which T4 was built and T1 was skipped, but the differences might very be within the range of uncertainty.

Back to Index

Units Built

The following tables show the amount of units built during the tournaments divided in buildings/mines, infantry and vehicles. The column 'total' denotes the total number of times this unit was build; 'games' is the number of games it was build in at least once. Blue fonts indicate doctrinal units.

Buildings

Infantry

While grens obviously are still the most built OH infantry unit, there are some noteable differences when comparing the stats to those for GSC2:

• Panzergrendiers were way more common this time. WCs saw about one squad per game, whereas in GSC2 had one squad in four to five games.
  
• Assault grenadiers were used in significant numbers.
  

Vehicles

During GCS2 the Panzer IV was the most commonly build OH vehicle. But this time around it was the 222. Tigers did appear more often, relatively speaking, as did StuG Gs (1:2 vs. 1:6 for GCS2). This is probably related to the players stalling for a Tiger more often than they did in GCS2, where Tiger doctrines were not commonly used. Similarily, only one Panther was build.

StuG-III Es where build in only one out of the eight Mechanized Assault games, so this unit is clearly not the selling point of the doctrine.

Build Orders

Back to Index

Soviet Meta

Soviet meta is available for 50 games. For players and commander picks I added two games of Jove beating VonAstens OKW using Shock Rifle and Armored Assault Tactics.

Players

Soviet Commanders

Again, any numbers in the SOV section below this point exclude the two Jove vs. VonAsten games.

Commanders vs. Faction Breakdown

Commander in Loadout

30	Armored Assault Tactics
29	Guard Motor Coordination Tactics
28	Urban Defense Tactics
15	Guard Rifle Combined Arms Tactic
14	Shock Rifle Frontline Tactics
10	Airborne Troops Tactics
6	Counterattack Tactics
4	Partisan Tactics
3	Mechanized Support Tactics
3	Lend Lease Tactics
3	Defensive Tactics
2	NKVD Rifle Disruption Tactics
1	Soviet Combined Arms Army
1	Soviet Shock Army
1	Advanced Warfare Tactics

The most commonly used commanders for SOV feature several of the all time classics like Armored Assault, Guard Motor Coordination (GMC) and Shock Rifle Frontline. A newcomer in the top three is Urban Defense, which in the past saw only some niche use. The new Airborne Troops made it into several loadouts but was not used much.

In GCS2 the most picked commander was Guard Rifle Combined, which still saw some use but was not as prevalent as back then. Another popular commander - Tank Hunter - didn't even make it into any of the loadouts.

Teching

The table below summarizes most teching choices.


Stats for GCS2 indicated about a 50:50 split between T1 and T2 start. In this tournament, T2 starts were way more common. Likewise, T4 was build more commonly, likely related to the concentration on heavier tanks.

Back to Index

Units Built

The following tables show the amount of units built during the tournaments divided in buildings/mines, infantry and vehicles. The column 'total' denotes the total number of times this unit was build; 'games' is the number of games it was build in at least once. Blue fonts indicate doctrinal units.

Buildings

Infantry

Given that there were more T2 starts instead of T1, it probably isn't surprising that more conscripts were build, both in terms of games as well as in relation to penals. That said, the difference is actually not that big when compared to GCS2.

A more significant difference is, that more games saw Zis-3s (roughly 75% vs. a little over 50% of GCS2 games), again probably due to players wanting an AT option while stalling for a heavy tank.

Guards saw a little less usage while shock troops were used a lot whereas the latter practically were not used at all in GSC2. The M-42 also was called in a significant number of times.

Vehicles

As always, the T-70 remains the most commonly build vehicle for the Soviets. T-34/76 were somewhat more popular now than in GSC2, while the build rate for T-34/85 was basically unchanged. A bigger shift happened in terms of "heavy" tanks: Relative to the total number of games, IS-2 tanks were used twice as often in WC, and a couple of KV-2 and KV-8 were built. GCS2 on the other hand saw more KV-1-action. Similarly, SU-85 were slightly more common now, likely to counter the opposing big cats.

Build Orders

Back to Index

Oberkommando West Meta

OKW meta is available for 67 games. Additionally, I have picked out at least the commander picks and player stats for six more games with OKW, specifically:

• Joves OKW with Grand Offensive Doctrine (GOD) losing to Jesulins USF using Mechanized on Nexus.
  
• Brosras OKW with the same Doctrine losing to VonIvans USF with Mechanized on Crossroads.
  
• VonAstens OKW (GOD) losing to Joves SOV with Shock Rifle on Kholodny Ferma.
  
• VonAstens OKW (GOD, what else?) winning over Joves USF (Heavy Cav) on Faymonville.
  
• Joves OKW (GOD) losing to VonAstens USF (Heavy Cav) on Faymonville.
  
• And finally VonAstens OKW (guess which) losing to Joves Soviets with Armored Assault on Crossroads.
  

Players

OKW Commanders

From this point on the statistics exclude the beforementioned six games for which I could not find a replay.

Commanders vs. Faction Breakdown

Commander in Loadout

59	Grand Offensive Doctrine
51	Luftwaffe Ground Forces Doctrine
21	Breakthrough Doctrine
19	Special Operations Doctrine
18	Overwatch Doctrine
14	Feuersturm Doctrine
13	Elite Armored Doctrine
6	Scavange Doctrine

GOD was by far the most used commander for OKW in this tournament. The pick rate is even slightly higher than Special Operations was in GCS2. The latter made it into a couple of loadouts but was never picked. Like at GSC2, Luftwaffe Ground Forces was the second most picked, although the results in terms of win-rates are not great.

Teching

The table below summarizes most teching choices.

The green percentages are based on all games. The blue percentages for the upgrades are computed based on only when the required tech was built.


During GSC2 almost half of the players started with Battlegroup vs. Mechanized. This changed significantly now with Mechanized being clearly in the lead. T4 was built only slighlty more often than during GCS2 (75% vs. 71%) and Panzer authorization was basically always unlocked.

Back to Index

Units Built

The following tables show the amount of units built during the tournaments divided in buildings/mines, infantry and vehicles. The column 'total' denotes the total number of times this unit was build; 'games' is the number of games it was build in at least once. Blue fonts indicate doctrinal units.

Buildings

Infantry

Apart from Panzerfüsiliers being way more common compared with GSC2, the rest of the infantry units appear roughly at the same rate. Panzer authorization - which was unlocked in all but one games that saw T4 - was actually intended make Obers more accessible. This might have worked out to some degree: This time Obers were build roughly every third game whereas in GSC2 it is closer to at least one Ober every fourth game.

Vehicles

While the Panzer IV was the most commonly build OKW vehicle in GSC2, now the Luchs and Puma top the chart. Instead of the Command Panther we now find the Tiger in the top 5.

Build Orders

Back to Index

USF Meta

USF meta is available for 49 games. Additionally, I found five more games as VoDs from which I took faction, map and commander pick stats:

• Jesulin (Mechanized) winning against Joves OKW (GOD) on Nexus;
  
• VonIvan (Mechanized) winning against Brosras OKW (GOD) on Crossroads;
  
• VonAsten with Heavy Cav losing to Joves OH (Spearhead) on Kholodny Ferma;
  
• Jove (Heavy Cav) losing to VonAstens OKW (GOD) on Faymonville;
  
• VonAsten (Heavy Cav) winning over Joves OKW (GOD) on Faymonville;
  

Players

USF Commanders

From this point on, the five VoD matches are not taken into account.

Commanders vs. Faction Breakdown

Commander in Loadout

48	Heavy Cavalry Company
39	Mechanized Company
26	Recon Support Company
10	Infantry Company
10	Airborne Company
7	Tactical Support Company
3	Urban Assault Company
2	None picked
1	Rifle Company
1	Armor Company

The most commonly used commander was Heavy Cavalry Company like in GSC2. However, Mechanized Company was a close second this time, achieving the same win rate.

Teching

The table below summarizes most teching choices.



Like in GSC2, most players went lieutenant. In fact, going by the replay files, VonIvan was the only player going for a captain first. In most games captain tech was unlocked at some point. However, while the mechanized platoon was almost always unlocked, that was not the case for the mechanized company.

Back to Index

Units Built

The following tables show the amount of units built during the tournaments divided in buildings/mines, infantry and vehicles. The column 'total' denotes the total number of times this unit was build; 'games' is the number of games it was build in at least once. Blue fonts indicate doctrinal units.

Buildings

Infantry

Compared with GSC2 there are not too many changes when it comes to infantry. Obviously, Cavalry Riflemen are a new addition to USF's rooster and actually turned out to be used relatively frequently. Mortars were slightly less prevalent this time around (1 in 5 games vs. 1 in 3 games during GSC2). Also, additional REs were build only in about 20% of matches, while this happened in more than 50% of the matches in GSC2.

Vehicles

After the Ambulance, the Sherman remains the second most built vehicle. The Stuart had a stronger showing with at least one appearing in about 70% of matches, while it was only close to 30% in GSC2. Also, the M20 was build about twice as often now, relatively speaking. In both tournaments Pershings and Jacksons were brought in in about 30% of games.

Build Orders

Back to Index

UKF Meta

And, for completenesses sake, there is UKF meta for 8 games.

Players

UKF Commanders

Commanders vs. Faction Breakdown

Commander in Loadout

7	Royal Artillery Regiment
5	Lend Lease Assault Regiment
4	Royal Engineering Regiment
3	Vanguard Operations Regiment
2	Tactical Support Regiment
2	Mobile Assault Regiment
1	Special Weapons Regiment

Teching

The table below summarizes most teching choices.

Green percentages are computed based on the total number of games, blue numbers are percentages of games in which the prerequisite was available.

Back to Index

Units Built

The following tables show the amount of units built during the tournaments divided in buildings/mines, infantry and vehicles. The column 'total' denotes the total number of times this unit was build; 'games' is the number of games it was build in at least once. Blue fonts indicate doctrinal units.

Buildings

Infantry

Vehicles

Build Orders

Back to Index



Special thanks to RagnarTheGamer and SneakEye for helping me out with with updating my script for the new unit and ability IDs. Also thanks to Whiteflash for providing the minimap for Nexus.