Checking model
Terminal Models
Within-league signed gap · Model A
+0.093
95% CI +0.063 to +0.124 · about +10%
02 / 07 · Value level
Mean residual by decile of predicted value (1 = lowest), log scale
| Row | UEFA | Non-UEFA | Gap |
|---|---|---|---|
| Decile 1 | 0.0047 | 0.0652 | 0.0605 |
| Decile 2 | −0.0249 | 0.1487 | 0.1736 |
| Decile 3 | −0.0950 | 0.0402 | 0.1352 |
| Decile 4 | −0.0174 | 0.0150 | 0.0324 |
| Decile 5 | −0.0237 | 0.0831 | 0.1068 |
| Decile 6 | −0.0535 | 0.0530 | 0.1065 |
| Decile 7 | −0.0242 | 0.1080 | 0.1322 |
| Decile 8 | −0.0325 | 0.1008 | 0.1333 |
| Decile 9 | 0.0769 | 0.1065 | 0.0296 |
| Decile 10 | 0.0057 | 0.0384 | 0.0327 |
Signed gap, log scale, 95% CI
| Row | Estimate | 95% CI low | 95% CI high |
|---|---|---|---|
| Within leagues (primary measure) | 0.0933 | 0.0629 | 0.1240 |
| Within deciles (same predicted value) | 0.0951 | 0.0657 | 0.1247 |
| Within league and decile | 0.1008 | 0.0704 | 0.1309 |
Estimates are pulled towards the middle, and non-UEFA players here are worth more on average, so the comparison is repeated inside each tenth of the model's own predictions. Binning on the prediction, not the actual value, avoids conditioning on the outcome. Non-UEFA players are under-predicted relative to UEFA players in every decile.