Beijing · R16 · Hard · 28 Sep 2024
Roman Safiullin v Jannik Sinner
79–91 in points · Jannik Sinner won
Why it went the way it did
Jannik Sinner won 8.4 more points than Roman Safiullin relative to what the serve alone predicted. Roman Safiullin's biggest leaks: return execution −2.5, slice shot selection −2.4, touch execution −1.5.
Each shot moves its hitter's chance of winning the point. Those moves are split into direction choice, shot selection and execution, and summed here. Read both columns together: one player's gain is the other's loss.
| Shot | Roman Safiullin | Jannik Sinner | ||||||
|---|---|---|---|---|---|---|---|---|
| Dir | Sel | Exec | Total | Dir | Sel | Exec | Total | |
| Serve 1st | +0.2 | ±0.0 | −1.0 | −0.8 | ±0.0 | +0.1 | −1.8 | −1.6 |
| Serve 2nd | +0.1 | +0.1 | +0.6 | +0.8 | ±0.0 | ±0.0 | −0.4 | −0.5 |
| Return | ±0.0 | +0.4 | −2.5 | −2.0 | +0.1 | +1.2 | +1.7 | +3.0 |
| Forehand | −0.1 | +2.2 | −0.7 | +1.4 | +0.2 | +2.2 | −1.4 | +1.0 |
| Backhand | +0.4 | +0.4 | −1.2 | −0.5 | +0.3 | +0.8 | +2.4 | +3.5 |
| Slice | −0.1 | −2.4 | −0.5 | −3.0 | ±0.0 | −0.8 | −0.1 | −0.9 |
| Net | +0.3 | +0.1 | +0.6 | +1.1 | +0.5 | ±0.0 | −0.6 | −0.2 |
| Touch | ±0.0 | −0.4 | −1.5 | −2.0 | +0.1 | ±0.0 | −0.4 | −0.4 |
| Other | ±0.0 | ±0.0 | ±0.0 | ±0.0 | ±0.0 | −0.3 | −0.3 | −0.6 |
| All shots | +0.9 | +0.2 | −6.1 | −5.0 | +1.1 | +3.1 | −0.8 | +3.4 |
Set by set
| Set | Roman Safiullin | Jannik Sinner | Net, Roman | |
|---|---|---|---|---|
| 1 | −0.7 | −2.9 | +2.2 | |
| 2 | −4.6 | +4.8 | −9.4 | |
| 3 | +0.2 | +1.4 | −1.2 |
Adaptation
First-serve direction mix set by set, with the share of points each direction won. The scores measure whether the server moved toward the directions that had been winning earlier in this match, set to set and every two to three service games, in points per 100 first serves. Read them as a habit, not a gain: across the tour, a direction's edge in early sets barely predicts the later ones. Whether the returner responded can't be told apart from noise within one match.
Roman Safiullin serving
| Set | Wide | Body | T |
|---|---|---|---|
| 1 | 53% 56% | 9% 33% | 38% 77% |
| 2 | 51% 33% | 3% 100% | 46% 63% |
| 3 | 63% 73% | – | 38% 44% |
Set to set −0.79 per 100 first serves in sets 2+
Within sets −0.11 per 100 first serves, every two to three service games
Jannik Sinner serving
| Set | Wide | Body | T |
|---|---|---|---|
| 1 | 48% 58% | 12% 67% | 40% 40% |
| 2 | 53% 70% | 5% 100% | 42% 100% |
| 3 | 58% 68% | 6% 50% | 36% 67% |
Set to set +0.16 per 100 first serves in sets 2+
Within sets +0.20 per 100 first serves, every two to three service games
Counterfactual: the best direction every time
If each shot had gone in the direction that is usually worth most for that stroke and position, this is how many more points each player would expect to win. It's an upper bound: the best direction isn't always on, and the chart can't see the ball's pace or position.
Roman Safiullin
6.4 points left on the table across 313 shots 2.06 per 100 shots
- Rally, shots 5–8: drive to your forehand side hit FH through the middle ×10; FH crosscourt rates higher · 0.51 pts
- Serve +1: short return to your backhand side hit FH drop shot inside-out ×2; FH drop shot inside-in rates higher · 0.27 pts
- Serve +1: deep return to your middle hit FH down the line ×6; FH crosscourt rates higher · 0.20 pts
- Return of a 2nd serve, T, ad court hit FH through the middle ×2; FH inside-out rates higher · 0.18 pts
Jannik Sinner
6.0 points left on the table across 319 shots 1.87 per 100 shots
- Long rally, 9+: drive to your backhand side hit BH through the middle ×8; BH crosscourt rates higher · 0.33 pts
- Serve +1: deep return to your middle hit FH through the middle ×4; FH crosscourt rates higher · 0.29 pts
- Rally, shots 5–8: drive to your backhand side hit BH through the middle ×7; BH crosscourt rates higher · 0.27 pts
- Return of a 1st serve, wide, ad court hit BH down the line ×3; BH crosscourt rates higher · 0.27 pts