Do referees book more at the start of the season?
No — and the gap between fouls and cards is the interesting part.
The season's arc
Every round expressed as a percentage of its own season's average round, across 26 seasons and 988 rounds, so the long decline in fouls since 2000-01 cannot masquerade as a within-season shape.
The two ends of the season
| Measure | Round 1 | Peak | Round 38 |
|---|---|---|---|
| Fouls | 106.5% (±3.61) | round 1 | 92.4% (±2.64) |
| Yellow cards | 98.6% (±9.21) | round 6 · 116.4% | 83.2% |
100 = a normal round for that season. Fouls open at their maximum and fall all year, both ends clearing their own margin of error. Cards do not follow: round one is indistinguishable from an average round.
The same thing, as a rate
| Cards per 100 fouls, rounds 1–2 → rounds 5–8 | |
|---|---|
| Change | +1.85 |
| 95% interval | +0.96 to +2.74 |
| Seasons moving the same way | 20 of 26 |
Pairing each season against itself is what makes a difference this size visible. Why is not knowable from this: referees easing in, players not yet cautioned into line, pre-season fitness, a September instruction — all fit the shape equally well.
Round by round
| Matchweek | Fouls | Yellows | Cards / 100 fouls |
|---|---|---|---|
| 1 | 106.5 | 98.6 | 13.2 |
| 2 | 105.5 | 103.7 | 14.2 |
| 3 | 101.4 | 110.2 | 15.6 |
| 4 | 103.6 | 112.0 | 15.6 |
| 5 | 103.9 | 109.0 | 15.1 |
| 6 | 102.3 | 116.4 | 16.5 |
| 10 | 101.9 | 98.8 | 14.0 |
| 15 | 99.9 | 102.0 | 14.7 |
| 20 | 96.3 | 96.7 | 14.5 |
| 25 | 101.0 | 102.6 | 14.6 |
| 30 | 100.4 | 95.9 | 13.7 |
| 34 | 97.5 | 97.2 | 14.4 |
| 36 | 98.6 | 94.4 | 13.7 |
| 37 | 96.6 | 98.2 | 14.7 |
| 38 | 92.4 | 83.2 | 12.9 |
Yellow cards use the public convention, counting a second yellow as the red it became. Red cards are absent: their curve is one dismissal a round bouncing about, not a shape. A round is ten matches by date — real matchweek numbers exist only from 2023-24.
Every figure here is a count, pooled across 988 rounds, comparing one part of the season with another. No referee is compared with any other — see why nobody can be called the strictest referee.