Published: 13 August 2026 · Game / balance / content: 1.0.0 / 1.0.0 / 1.0.0
+9.34 mean-score points
10.5× as many elite careers
What we tested
Each paired seed used the same position and the same alternating branch through five career events. Only the eight-pick draft policy changed:
- Fixed-order baseline: take pace, shooting, passing, dribbling, defending, physical, skill and weak foot in that order, regardless of the displayed ratings.
- Greedy best-available: on every scout card, take the highest rating among attributes not already locked.
The 10,000 seeds were split evenly across forward, midfielder, defender and goalkeeper—2,500 paired seeds per position. That produced 20,000 simulated careers in total.
Results
| Draft policy | Mean score | Elite (95+) | Elite rate |
|---|---|---|---|
| Fixed order | 76.31 | 125 | 1.25% |
| Greedy rating | 85.65 | 1,314 | 13.14% |
The greedy policy gained 9.34 mean-score points and 11.89 percentage points in elite rate. The elite count was 10.512 times the baseline.
In balance 1.0, reading the largest number on each card is substantially stronger than the fixed-order baseline. This is evidence for a balance review—not proof that every human player will discover or use the same policy.
Reproduce it
Clone the site repository at the matching version, install dependencies and run npm run test:engine. The test asserts content budgets, deterministic snapshots, safe challenge URLs and the
10,000 paired-seed thresholds before printing the aggregates.
Limits
- This is a deterministic engine diagnostic, not a study of observed player behavior.
- The fixed-order policy is a transparent baseline, not random drafting and not an optimal counter-strategy.
- Results describe balance version 1.0.0 and should not be generalized to later versions without rerunning the experiment.
How to cite
coperogame.games (2026), “10,000 Career Runs: Football Strategy & Balance Experiment,” balance version 1.0.0.