Checking model
Terminal Models
Within-league signed gap · Model A
+0.093
95% CI +0.063 to +0.124 · about +10%
05 / 07 · Tier errors
Non-UEFA minus UEFA, percentage points, 95% CI
| Row | Estimate | 95% CI low | 95% CI high |
|---|---|---|---|
| Wrong tier (error-rate gap (ER1a)) | 0.0248 | 0.0007 | 0.0488 |
| Placed below true tier (under-tier gap) | 0.0330 | 0.0131 | 0.0543 |
| Placed above true tier (over-tier gap) | −0.0083 | −0.0240 | 0.0082 |
Non-UEFA minus UEFA, percentage points
| Row | Estimate | 95% CI low | 95% CI high |
|---|---|---|---|
| Emerging (617 non-UEFA; detectable gap ≈ 3.8 pp) | −0.0129 | −0.0420 | 0.0139 |
| Established (539 non-UEFA; detectable gap ≈ 6.2 pp) | 0.0203 | −0.0233 | 0.0611 |
| Elite (246 non-UEFA; detectable gap ≈ 10.3 pp) | −0.0354 | −0.1126 | 0.0374 |
| World Class (94 non-UEFA; detectable gap ≈ 16.4 pp) | −0.0523 | −0.1702 | 0.0577 |
The extra tier errors for non-UEFA players are mostly placements below their actual tier, consistent with the value under-prediction. Per-tier comparisons rest on small groups (94 non-UEFA World Class, 246 Elite): intervals wide enough to fit no gap or a sizeable one, so they are not read as evidence either way.