Spirit Island Digital

Detailed Statistics:
Snake (4 Boards)

Games Completed	Blight Flip Rate	Average Score	Win Rate
27	62.96%	32.33	66.67%

Results using Spirits

Spirit	Completed	Blight Flipped	Average Score	Win Rate
A Spread of Rampant Green	7	28.57%	39.14	100.0%
Bringer of Dreams And Nightmares	6	50.0%	23.33	50.0%
Devouring Teeth Lurk Underfoot	3	33.33%	28.67	66.67%
Eyes Watch From the Trees	8	75.0%	28.63	62.5%
Fathomless Mud of the Swamp	1	100.0%	50.0	100.0%
Heart of the Wildfire	11	81.82%	28.55	45.45%
Keeper of the Forbidden Wilds	13	69.23%	28.15	46.15%
Lightning's Swift Strike	7	71.43%	32.43	42.86%
Lightning's Swift Strike — Wind	2	100.0%	28.0	100.0%
Lure of the Deep Wilderness	1	100.0%	20.0	0.0%
Many Minds Move as One	2	50.0%	25.0	100.0%
Ocean's Hungry Grasp	5	20.0%	28.2	100.0%
Rising Heat of Stone and Sand	3	66.67%	38.33	66.67%
River Surges in Sunlight	4	25.0%	34.75	100.0%
Serpent Slumbering Beneath the Island	7	57.14%	40.57	85.71%
Shadows Flicker Like Flame	2	100.0%	49.5	100.0%
Sharp Fangs Behind the Leaves	3	66.67%	40.33	100.0%
Stone's Unyielding Defiance	1	100.0%	20.0	0.0%
Sun-Bright Whirlwind	3	33.33%	44.0	100.0%
Thunderspeaker	12	66.67%	32.92	50.0%
Vengeance as a Burning Plague	1	100.0%	27.0	100.0%
Vital Strength of the Earth	4	75.0%	37.75	50.0%
Vital Strength of the Earth — Resilience	2	100.0%	28.0	100.0%

Results with Adversaries

Adversary	Level	Difficulty	Completed	Blight Flipped	Average Score	Win Rate
The Kingdom of Brandenburg-Prussia	0	1	2	0.0%	28.0	100.0%
The Kingdom of Brandenburg-Prussia	2	4	1	0.0%	62.0	100.0%
The Kingdom of Brandenburg-Prussia	3	6	3	33.33%	50.0	100.0%
The Kingdom of Brandenburg-Prussia	4	7	1	100.0%	20.0	0.0%
The Kingdom of Brandenburg-Prussia	6	10	2	100.0%	26.0	0.0%
The Kingdom of England	3	6	4	75.0%	32.5	50.0%
The Kingdom of France (Plantation Colony)	1	3	2	100.0%	28.0	100.0%
The Kingdom of France (Plantation Colony)	4	8	5	100.0%	28.0	20.0%
The Kingdom of Sweden	4	6	2	50.0%	47.0	100.0%

Results using Scenarios

Scenario	Difficulty	Completed	Blight Flipped	Average Score	Win Rate	Win Model	Differential	Model (Diff≥10)
Dahan Insurrection	4	1	0.0%	62.0	100.0%

Results using Blight Cards

Blight Card	Completed	Average Score	Win Rate
A Pall Upon the Land	1	21.0	0.0%
Aid From Lesser Spirits	1	50.0	100.0%
Back Against the Wall	2	25.0	50.0%
Downward Spiral	2	48.0	100.0%
Erosion of Will	2	32.5	50.0%
Memory Fades to Dust	3	29.67	66.67%
Promising Farmlands	2	26.0	0.0%
Strong Earth Shatters Slowly	2	28.0	100.0%
Tipping Point	1	19.0	0.0%
Untended Land Crumbles	1	20.0	0.0%

Results Using Boards

Board	Completed	Blight Flipped	Average Score	Win Rate
A	21	52.38%	34.57	76.19%
B	24	66.67%	30.96	62.5%
C	15	60.0%	34.13	66.67%
D	25	64.0%	33.4	64.0%
E	3	100.0%	25.33	66.67%
F	4	50.0%	26.0	100.0%
G	11	72.73%	32.55	63.64%
H	5	60.0%	27.6	40.0%

Results Using Options

Option	Completed	Blight Flipped	Average Score	Win Rate
Not Using Events or Tokens	1	0.0%	62.0	100.0%
Start of Game Growth	1	100.0%	27.0	100.0%
Unlimited Undo	8	25.0%	31.75	87.5%
Using Tokens but Not Events	2	0.0%	28.0	100.0%

Terror Level

Terror Level	Completed	Average Score	Win Rate
I	2	22.56	0.0%
II	10	30.98	30.0%
III	9	34.53	100.0%
IV	6	37.77	100.0%

Results

Result	Completed
Win: Invaders destroyed (Terror III)	9
Win: Invaders destroyed (Terror II)	2
Win: Terror victory (Terror IV)	6
Lose: Adversary condition	2
Win: Scenario condition	1
Lose: Blight pool ran out	6
Lose: Spirit destroyed	1

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.

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.

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.

Model (Diff ≥10)

Same as Win Model, but instead counts all games of difficulty 10 or higher, for all time.

Layouts

Diagrams of the layouts are on the main page. Data collected before v.2.7.1 (May 2, 2025) use the end game state, even if the island layout changed due to powers like Cast Down Into the Briny Deep. Data collected from v.2.7.1 onward uses the layout as it was at setup.