Results vs. Adversaries

Adversary Completed Completion Rate Average Score Score Model Win Rate Deviation p(χ²) Win Model Differential
The Kingdom of Brandenburg-Prussia 9096 60.7% 36.26 5.224 72.36% -0.001 0.8326 -0.507 -0.073
The Kingdom of England 8488 60.57% 35.28 4.186 55.6% -0.072 0.0 -1.464 -0.146
The Kingdom of France (Plantation Colony) 4696 64.79% 40.11 5.727 58.18% 0.001 0.8607 -1.487 0.2932
The Kingdom of Sweden 5242 59.35% 35.0 3.915 72.64% 0.013 0.0422 -0.541 -0.036

Results in Scenarios

Scenario Difficulty Completed Completion Rate Average Score Score Model Win Rate Deviation p(χ²) Win Model Differential
Blitz 0 903 56.94% 25.48 -2.133 67.22% -0.012 0.4312 -0.511 -0.047
Dahan Insurrection 4 1002 65.66% 52.8 10.252 73.95% 0.054 0.0002 -0.357 -0.0089
Guard the Isle's Heart 0 2433 71.58% 43.34 5.205 74.27% 0.006 0.5275 -0.867 -0.5613
Powers Long Forgotten 1 361 53.09% 37.63 0.97 75.35% 0.045 0.0618 -0.016 0.2859
Rituals of Terror 3 299 53.97% 35.49 2.414 54.85% -0.099 0.0003 -0.626 0.0438
Rituals of the Destroying Flame 3 118 39.86% 31.2 0.153 42.37% -0.181 0.0001 -0.814 -0.1075
Second Wave 0 2210 59.04% 28.88 -0.492 71.67% 0.012 0.214 -0.442 0.013
Ward the Shores 2 304 51.88% 34.07 0.85 64.47% 0.019 0.4853 -0.546 0.4269

Results using Layouts

Results using Blight Cards

Blight Card Completed Average Score Score Model Win Rate Deviation p(χ²) Win Model Differential
A Pall Upon the Land 1572 28.53 -1.998 48.79% -0.084 0.0 -2.513 -0.3832
Aid From Lesser Spirits 1363 31.06 -0.885 52.6% -0.071 0.0 -2.403 -0.4104
Back Against the Wall 1368 29.58 -1.628 50.37% -0.07 0.0 -2.5 -0.3568
Disintegrating Ecosystem 1481 26.42 -3.047 46.25% -0.083 0.0 -2.65 -0.357
Downward Spiral 5980 21.64 -3.706 54.5% -0.063 0.0 -2.569 -0.5098
Erosion of Will 1547 27.06 -2.698 46.28% -0.088 0.0 -2.656 -0.3703
Memory Fades to Dust 6223 21.59 -3.788 51.5% -0.063 0.0 -2.687 -0.4277
Promising Farmlands 1577 25.48 -3.465 43.63% -0.112 0.0 -2.756 -0.4682
Tipping Point 1433 26.62 -2.869 46.62% -0.087 0.0 -2.59 -0.361

Results Starting on Boards

Board Completed Completion Rate Average Score Score Model Win Rate Deviation p(χ²) Win Model Differential
A 12711 57.31% 25.69 -0.701 70.45% -0.014 0.0004 0.009 0.0817
B 13734 55.7% 27.63 0.052 72.5% -0.002 0.5121 0.076 0.05
C 14576 59.82% 28.59 0.374 72.36% -0.017 0.0 0.044 -0.194
D 12344 57.92% 26.9 -0.15 72.96% 0.001 0.8386 0.069 0.053
East 771 50.43% 39.75 1.739 63.16% -0.043 0.0105 -0.02 -0.0614
Northeast 729 54.73% 37.57 1.091 61.87% -0.06 0.0005 -0.013 -0.0115
Northwest 963 55.57% 41.65 1.712 69.06% 0.008 0.5966 0.028 -0.0016
West 780 51.28% 34.85 0.058 56.54% -0.097 0.0 -0.166 0.0934

Results Using Boards

Board Completed Completion Rate Average Score Score Model Win Rate Deviation p(χ²) Win Model Differential
A 21977 54.59% 28.35 -0.084 73.19% 0.013 0.0 -0.023 0.0499
B 22428 54.77% 29.0 0.062 74.16% 0.014 0.0 -0.032 -0.0583
C 23762 56.61% 29.84 0.649 74.48% 0.005 0.1005 0.088 -0.1497
D 21461 55.04% 28.88 0.117 74.74% 0.019 0.0 0.005 -0.0104
East 1924 50.03% 41.8 1.626 65.02% -0.025 0.0212 0.028 -0.0135
Northeast 1722 52.07% 41.96 1.589 66.03% -0.019 0.0968 -0.046 -0.0442
Northwest 1668 52.27% 41.56 1.149 67.03% -0.012 0.2786 0.166 0.1366
West 1864 49.87% 40.49 0.902 62.39% -0.039 0.0004 -0.266 -0.0069

Results Using Options

Option Completed Completion Rate Average Score Score Model Win Rate Deviation p(χ²) Win Model Differential
No Blight Card 7569 45.62% 15.93 -3.998 56.24% -0.127 0.0 -1.308 -0.2388
Not Using Events or Tokens 23358 51.49% 21.35 -2.864 69.86% -0.017 0.0 -0.008 0.2098
Skip Initial Invader Actions 2604 68.24% 36.35 3.594 94.2% 0.095 0.0 1.452 0.6052
Start of Game Growth 3283 66.77% 36.44 3.473 90.5% 0.071 0.0 0.633 0.3484
Unlimited Undo 7089 60.9% 31.68 1.942 83.95% 0.057 0.0 0.438 -0.0023
Using Tokens but Not Events 5764 61.66% 30.4 0.264 81.71% 0.051 0.0 0.354 0.3107

Results With Allies

Terror Level

Terror Level Completed Average Score Win Rate
I 8799 22.24 35.57%
II 30506 29.69 72.5%
III 14268 32.43 85.85%
IV 3035 36.13 99.28%

Power Progression

Score Model: -4.722 Win Model: -1.052
Power Progression? Completed Completion Rate Average Score Win Rate
Standard 49192 60.75% 30.0 74.53%
Power Progression 7416 42.01% 14.09 51.86%

Results

Result Completed
Win: Scenario condition 937
Lose: Scenario condition 490
Lose: Adversary condition 2569
Win: Invaders destroyed 36427
Lose: Invader deck ran out 205
Win: Terror victory 2985
Win: Sacrifice victory 160
Lose: Spirit destroyed 464
Lose: Blight pool ran out 12371

Number of Players

Score Model: 0.37 Win Model: -0.08
Num. Completed Completion Rate Average Score Win Rate Dev. p(χ²)
1 31576 62.09% 24.95 68.87% -0.022 0.0
2 15054 53.17% 29.93 73.0% 0.002 0.6057
3 4790 52.75% 32.56 77.14% 0.012 0.0432
4 5188 49.97% 35.8 78.62% 0.024 0.0

Online Multiplayer

Score Model: -1.242 Win Model: 0.091
Multiplayer? Completed Completion Rate Average Score Win Rate
Single-Player 52703 56.36% 27.92 71.21%
Multiplayer 3905 76.28% 27.78 76.26%

Notes

Score Model

Ridge regression coefficient using all independent variables as predictors, and the score as the outcome. This corresponds 1:1 with score, e.g. a coefficient of +1 means a predicted score 1 higher than average. Ridge regression is used instead of standard linear regression because of high multicollinearity. Exception: For Difficulty, the Score Model is a separate model, using only difficulty as a predictor.

Deviation

The difference between the actual win rate of this combination of elements and the win rates of each of those elements on their own, averaged. A positive value means that this combination results in a win more often than would be expected.

p(χ²)

This is the significance value of a χ² test comparing the observed data and expected value. In most cases, this expected value is the mean of the win rates of each element. For team position, the expected value is an equal likelihood of winning in all positions. Lower p-values (no greater than 0.01) mean a greater likelihood that the difference is meaningful and not due to chance.

Win Model

Logistic regression coefficient using all independent variables as predictors, and whether the game was won or not as the outcome. Higher values indicate greater likelihood of winning when present in the game. For continuous variables (e.g. number of players), this number is per unit. model.
Exception: For Difficulty, the Win Model is a separate model, using only difficulty as a predictor.

Differential

The difference between the logistic regression coefficient for this item using this Spirit, as compared to using this item in general. A negative number means that this Spirit underperforms using this item, and a positive number means that it overperforms.

Layouts

Diagrams of the layouts are on the main page. Layouts for completed games are shown as the same as what they were when the game was started, despite powers like Cast Down Into the Briny Deep.

Last updated: Sep 08, 2023, 09:55 AM PDT