Agricola Statistics Update: games ending Dec '22 - Jul '23
Posted: 13 July 2023, 23:15
Thanks again to BGA (especially lordalx) and Lookout for allowing this analysis that will benefit the community! I'm just reusing the format from the previous thread here.
November 2021-December 2021: https://boardgamearena.com/forum/viewtopic.php?t=23574
April 2022-September 2022: https://boardgamearena.com/forum/viewtopic.php?t=28051
Unless otherwise stated, all data is for games ended between Dec 30 2022 and Jul 13 2023, where all players end with 200+ Elo, no additional spaces, Normal and Competitive modes combined, and with both CD and "New" cards enabled. "n" denotes how many tables, unless it's a category for a specific group of players, in which case I'll specify players.
First is analysis of Seating Order (starting position):
2-player (n=3181)
2-player, Additional Action Spaces (AAS) (n=1996)
3-player (n=267)
4-player (n=8944)
4-player, AAS (n=4152)
Looking at how Expert (500+ Elo) players are affected by starting position in 4-player games:
4-player with banlist, Experts (n=6351 players)
4-player, AAS, Experts (n=3178 players)
In many cases the disparities between seating positions are smaller than in the first data set. That could be attributed to many possible reasons, including improved community skill level or random fluctuations. Last seat in 4p games still seems a little rougher than the other seats, and last seat in 3p seems to be the biggest disadvantage by a large margin.
Key terms for the card statistics in following posts:
PWR is a measure of how often a card is on a 1st place tableau. ADP is the average draft position. APR is the average played round. WtdPWR is an experimental variant of PWR where instead of counting 1st places as 1 credit, I gave 0.6 credits to 1st place, 0.3 credits to 2nd place, and 0.1 credits to 3rd place. Elo/Play is a new calculation, the average Elo won or lost across all times the card was played.
The latter 2 stats are attempts to evaluate which cards are possibly strong because of how often they help a player get 2nd or 3rd place, or more generally win Elo - as opposed to regular PWR which only cares about getting 1st place or not.
November 2021-December 2021: https://boardgamearena.com/forum/viewtopic.php?t=23574
April 2022-September 2022: https://boardgamearena.com/forum/viewtopic.php?t=28051
Unless otherwise stated, all data is for games ended between Dec 30 2022 and Jul 13 2023, where all players end with 200+ Elo, no additional spaces, Normal and Competitive modes combined, and with both CD and "New" cards enabled. "n" denotes how many tables, unless it's a category for a specific group of players, in which case I'll specify players.
First is analysis of Seating Order (starting position):
2-player (n=3181)
Code: Select all
order elo_start score end_1st end_2nd avg_rank
1 370.28 42.38 49.8% 50.2% 1.50
2 368.59 42.30 51.0% 49.0% 1.49
Code: Select all
order elo_start score end_1st end_2nd avg_rank
1 342.56 47.65 50.6% 49.4% 1.49
2 344.23 47.69 49.9% 50.1% 1.50
Code: Select all
order elo_start score end_1st end_2nd end_3rd avg_rank
1 314.39 42.87 37.5% 31.5% 31.1% 1.94
2 305.32 42.69 35.6% 33.0% 31.5% 1.96
3 308.80 41.64 27.0% 36.3% 36.7% 2.10
Code: Select all
order elo_start score end_1st end_2nd end_3rd end_4th avg_rank
1 393.21 43.44 26.3% 25.2% 24.8% 23.8% 2.46
2 393.14 43.39 26.9% 24.9% 24.3% 24.0% 2.45
3 392.02 43.20 24.3% 25.3% 25.4% 25.0% 2.51
4 392.51 42.78 23.3% 24.8% 25.3% 26.6% 2.55
Code: Select all
order elo_start score end_1st end_2nd end_3rd end_4th avg_rank
1 393.29 47.74 26.6% 25.3% 24.3% 23.8% 2.45
2 393.15 47.55 26.2% 25.7% 24.4% 23.7% 2.46
3 391.36 47.23 24.3% 25.0% 24.9% 25.7% 2.52
4 394.61 46.96 23.7% 24.2% 26.1% 26.1% 2.55
4-player with banlist, Experts (n=6351 players)
Code: Select all
order elo_start score end_1st end_2nd end_3rd end_4th avg_rank
1 566.47 48.01 48.3% 28.9% 15.7% 7.1% 1.82
2 565.05 48.08 50.2% 27.1% 14.9% 7.8% 1.80
3 563.35 47.94 46.8% 30.6% 15.1% 7.5% 1.83
4 564.98 47.35 44.9% 28.7% 17.0% 9.3% 1.91
Code: Select all
order elo_start score end_1st end_2nd end_3rd end_4th avg_rank
1 574.41 52.78 48.8% 27.5% 16.1% 7.6% 1.82
2 577.19 52.79 45.3% 30.0% 15.9% 8.8% 1.88
3 575.78 52.40 45.2% 29.5% 16.8% 8.5% 1.89
4 578.43 52.15 44.2% 29.2% 18.2% 8.4% 1.91
Key terms for the card statistics in following posts:
PWR is a measure of how often a card is on a 1st place tableau. ADP is the average draft position. APR is the average played round. WtdPWR is an experimental variant of PWR where instead of counting 1st places as 1 credit, I gave 0.6 credits to 1st place, 0.3 credits to 2nd place, and 0.1 credits to 3rd place. Elo/Play is a new calculation, the average Elo won or lost across all times the card was played.
The latter 2 stats are attempts to evaluate which cards are possibly strong because of how often they help a player get 2nd or 3rd place, or more generally win Elo - as opposed to regular PWR which only cares about getting 1st place or not.