WBP 2v2 tourney stats

For completeness, here are the stats of the WBP 2v2 tourney. I had a few issues finding the correct games due to battlefy and the apparently somewhat chaotic turn of events, but I should have almost all games and certainly all of the games involving the better known players.

If you want to compare the stats below to previous ones, well, the closest I can offer are those for the NMC 2v2 tournament.

In total I found data for 58 matches. The longest match of the tourney was the second game of the - later nullified - semi finals between "The lost Legion" and the "Turkuaz" with 58 minutes in which Desert Eagle also produced the largest total damage of the tourney with 62468 (playing soviets).

In terms of team setups we had:




Stats for individual factions:



It has to be said though, that apparently all BO3 rounds ended with a clear 2:0, so the minor difference between axis and allied win rates are due to what happened in the earlier BO1 rounds.

Map stats:


Happy holidays!

So USF was theleast played faction yet had the highest winrate.

while

Whermacht was the most played faction with the lowest winrate? Correct me if I am wrong.


Was this because of the callin meta(pershing, callie, m10)? I haven't watched any of the games.

Double Prokw OpieOp

So USF was theleast played faction yet had the highest winrate.

while

Whermacht was the most played faction with the lowest winrate? Correct me if I am wrong.


Was this because of the callin meta(pershing, callie, m10)? I haven't watched any of the games.

Well, due to there only being two axis factions, it is no surprise that one of them ended up being the most played faction.

I should have noted that while I didn't include confidence intervals, they ended up being 5% or more for all factions so that the confidence interval in all cases covered the 50% mark (with the exception of OH, which was slightly below that).

But yeah, in particular USF+x was somewhat more successful than SOV+UKF.

That said, this is just data for 58 games so this at best is one data point and not an authoritative statement of the 2v2 faction balance. In comparison, the WPC series or ESL stats included more than 3 and 8 times as many games, respectively; and that was for 1v1 without effects of faction synergies and so on. If the same trend shows up consistently in upcoming tourneys, the numbers might be considered more reliable .

I should have noted that while I didn't include confidence intervals, they ended up being 5% or more for all factions so that the confidence interval in all cases covered the 50% mark (with the exception of OH, which was slightly below that).

But yeah, in particular USF+x was somewhat more successful than SOV+UKF.

That said, this is just data for 58 games so this at best is one data point and not an authoritative statement of the 2v2 faction balance. In comparison, the WPC series or ESL stats included more than 3 and 8 times as many games, respectively; and that was for 1v1 without effects of faction synergies and so on. If the same trend shows up consistently in upcoming tourneys, the numbers might be considered more reliable .

I see. I was actually surprised to see the win ratio. Show Spoiler

Thank you for your hard work, keep us informed with your excellent analytical data

although the sample is quite small I think it is interesting that double Soviet or double Ostheer seem to be doing bad indicating that faction have good synergy...

It has to be said though, that apparently all BO3 rounds ended with a clear 2:0, so the minor difference between axis and allied win rates are due to what happened in the earlier BO1 rounds.

Therefore one cannot really use this data to say that allies are OP or not. Nonetheless seeing Noggano and Mate struggling against Igor and Arthur, who played terribly, was a clear indicator in my opinion....

although the sample is quite small I think it is interesting that double Soviet or double Ostheer seem to be doing bad indicating that faction have good synergy...

They do and they don't.

Double Soviets or Ostheer have huge strategic implications that the opposing team can use to their advantage, especially if they're mixed teams.

Remember, factions were designed for years to just win against other factions at different stages of the game.

Kubels and Sturms do a whole lot more to support Ostheer trying to establish a footprint. It also puts the load off of flimsy Grens for holding the line against allied infantry.

Riflemen support maxims far better than conscripts ever can, and have more options and abilities than any Soviet infantry, stock or no, could hope for.

But the EFA factions DID have a semblance of balance AND teamgame scaling before WFA and Brits.

although the sample is quite small I think it is interesting that double Soviet or double Ostheer seem to be doing bad indicating that faction have good synergy...

Well, the "double x" counts are pretty low, so we're on shaky ground here.

If we add the results for this tourney and the NMC cup (yes, different patches and different commanders, but the basic faction design wasn't changed), we get this:



so the win percentages for single faction teams are lower than for mixed factions (given the low sample counts this has to be taken with a truckload of salt; USF/USF has only two games, the difference between OH/OH and OKW/OKW on one hand and OH/OKW on the other isn't significant, same for UKF/UKF and SOV/UKF; the only faction were this is close to significant SOV/SOV).

If we compute winrates for all mixed teams, we get about 53% while single factions teams end up at 37%. Again, no rock solid proof, but some hint that single faction teams might be weaker.

...also skimming through the matches it seems that the better known players play mixed faction teams and their higher skill level will to some extend improve win rates for those teams.

So, lower winrates might partially be also due to better players perceiving single faction teams as UP (which in itself I'd consider relevant).

If e.g. Jove and Alastor would have played SOV/SOV instead of SOV/UKF, that winrate probably would have been significantly higher. But then again, maybe they wouldn't have won the tourney.

So there is a bit of a chicken and egg problem: Are successful players successful because they play a certain faction, or are certain factions successful because they are used by skilled players? And my guess is that the answer is: A bit of both...

Good work man ! Thanks.

What did you use to collect data? Did you just scour coh2.org for replays?

What did you use to collect data? Did you just scour coh2.org for replays?

There are basically no replays for this tourney...

I download the match history (you know, the same data that is used to generate the list in "Recent Matches" tabs of the leaderboards) for each involved player. When there is a tourney, I have a script that downloads the data every 1-2 hours.

For that I need a list of steam IDs the rest is fairly automated by now.

In this case, however, it was kind of tricky, because battlefy basically only listed the teamnames (duh!), no IDs. At least I knew the IDs for the "usual suspects" and with those I started a bit of an iterative process, like, query the few IDs I knew, read the data and search for the IDs of their opponent, then search for their games and so on.

Finally, when I have (at least most of) the data, I have to go through what I got and the brackets, because the data often contain matches not related to the tourney, or rematches (like, a player might have had a disconnect in the first few minutes and the players decided to redo the match). Also, this step often turns out to be a bit tricky because brackets are often not accurate (like, they might say 2:0 but the advancing team actually won by default and no games were played...).