Average PF by draft slot
Eight observations per slot. Pick 8 has the lowest historical average, but the overall slot effect is not statistically meaningful.
| Slot | Avg PF | Avg rank |
|---|
| 1 | 1,554.19 | 5.25 |
| 2 | 1,507.09 | 6.50 |
| 3 | 1,593.02 | 3.88 |
| 4 | 1,484.75 | 7.00 |
| 5 | 1,517.97 | 6.25 |
| 6 | 1,591.63 | 4.75 |
| 7 | 1,586.81 | 4.50 |
| 8 | 1,471.51 | 7.00 |
| 9 | 1,569.99 | 4.62 |
| 10 | 1,556.42 | 5.25 |
A categorical comparison of PF across the 10 draft slots is not significant (ANOVA p = 0.548). The weak slot pattern should not be treated as evidence that a particular slot is inherently good or bad.
PF by move activity
Manager-seasons split into four equal groups by number of transactions.
| Activity | Avg moves | Avg PF | Avg PF rank |
|---|
| Low | 7.85 | 1,507.40 | 6.35 |
| Mid-low | 18.40 | 1,549.35 | 5.25 |
| Mid-high | 29.75 | 1,546.17 | 5.45 |
| High | 43.20 | 1,570.43 | 4.95 |
The highest-activity quartile averaged about 63 more PF than the lowest-activity quartile. After normalizing within each season, the moves-to-PF relationship remains weak (Spearman ρ ≈ +0.08). When managers make more moves than their own usual level, the relationship with PF is somewhat stronger (ρ ≈ +0.18), but still not strong enough to establish a reliable effect.