Season script report

Most Predictable WTA Players

Every match, our model publishes a win probability, and prices its own uncertainty. This board compares each player's actual results with that script: straight-sets players land where the model expected, wild cards keep tearing the script up. A season-level read with sample sizes and confidence intervals, never a career label.

The metronome · 2025
Elise Mertens
14 upsets where the model expected 19.1 · 36-20 in 56 matches · model called it right 75.0% of the time · still within noise for this sample
The wild card · 2025
Jelena Ostapenko
21 upsets where the model expected only 13.2 · 17-19 in 36 matches · model called it right 41.7% of the time

Script vs chaos: the two ends of 2025

Bars show extra upsets: how many more (or fewer) times the model's favorite fell in this player's matches than the model itself expected. Verdicts in the table are assigned only when the difference is statistically meaningful for that sample. Everyone else stays "On serve". How to read this →

Full board · 98 players with ≥20 matches (2448 matches, source: out-of-sample backtest)

# Player Verdict Upsets Expected Extra Model hit Confidence W-L M
1 Elise Mertens On serve 14 19.1 -5.1 ±6.8 75.0% 65.9% 36-20 56
2 Jessica Pegula On serve 18 23.2 -5.2 ±7.5 74.3% 66.8% 51-19 70
3 Maya Joint On serve 13 17.1 -4.1 ±6.3 72.3% 63.7% 27-20 47
4 Ann Li On serve 16 20.2 -4.2 ±6.9 70.4% 62.5% 31-23 54
5 Kamilla Rakhimova On serve 6 8.8 -2.8 ±4.6 77.8% 67.4% 8-19 27
6 Aryna Sabalenka On serve 12 16.1 -4.1 ±6.7 82.9% 77.1% 59-11 70
7 Camila Osorio On serve 10 13.4 -3.4 ±5.6 73.0% 63.9% 20-17 37
8 Victoria Azarenka On serve 5 7.4 -2.4 ±4.2 76.2% 64.7% 9-12 21
9 Madison Keys On serve 12 15.5 -3.5 ±6.3 76.0% 69.0% 37-13 50
10 Caty McNally On serve 5 7.3 -2.3 ±4.1 76.2% 65.1% 10-11 21
11 Viktoriya Tomova On serve 7 9.7 -2.7 ±4.8 75.0% 65.5% 9-19 28
12 Caroline Dolehide On serve 9 11.8 -2.8 ±5.3 71.0% 62.0% 12-19 31
13 Elena Rybakina On serve 15 18.6 -3.6 ±7.0 77.3% 71.8% 48-18 66
14 Cristina Bucsa On serve 10 12.7 -2.7 ±5.4 70.6% 62.8% 15-19 34
15 Marie Bouzkova On serve 12 14.8 -2.8 ±6.0 72.7% 66.3% 27-17 44
16 Marta Kostyuk On serve 14 16.8 -2.8 ±6.2 68.2% 61.8% 25-19 44
17 Linda Noskova On serve 20 23.1 -3.1 ±7.4 67.7% 62.7% 37-25 62
18 Iga Swiatek On serve 13 15.9 -2.9 ±6.7 81.7% 77.7% 57-14 71
19 Belinda Bencic On serve 15 17.7 -2.7 ±6.6 71.2% 65.9% 36-16 52
20 Yulia Putintseva On serve 13 15.1 -2.1 ±5.9 68.3% 63.2% 15-26 41
21 Jaqueline Cristian On serve 15 17.0 -2.0 ±6.3 68.1% 63.9% 26-21 47
22 Ekaterina Alexandrova On serve 24 26.2 -2.2 ±7.9 66.7% 63.7% 48-24 72
23 Veronika Kudermetova On serve 17 18.7 -1.7 ±6.6 66.0% 62.5% 27-23 50
24 Jasmine Paolini On serve 17 18.8 -1.8 ±6.8 70.7% 67.6% 41-17 58
25 Maria Sakkari On serve 15 16.4 -1.4 ±6.2 66.7% 63.5% 20-25 45
26 McCartney Kessler On serve 20 21.5 -1.5 ±7.0 63.0% 60.1% 31-23 54
27 Emma Raducanu On serve 16 17.2 -1.2 ±6.4 66.7% 64.2% 27-21 48
28 Leylah Fernandez On serve 18 19.0 -1.0 ±6.7 66.0% 64.1% 30-23 53
29 Sorana Cirstea On serve 15 15.9 -0.9 ±6.2 67.4% 65.3% 28-18 46
30 Viktorija Golubic On serve 7 7.6 -0.6 ±4.4 70.8% 68.2% 10-14 24
31 Anna Blinkova On serve 15 15.8 -0.8 ±6.1 65.1% 63.3% 21-22 43
32 Xinyu Wang On serve 14 14.5 -0.5 ±5.8 64.1% 62.8% 18-21 39
33 Renata Zarazua On serve 8 8.3 -0.3 ±4.6 70.4% 69.1% 10-17 27
34 Polina Kudermetova On serve 11 11.4 -0.4 ±5.2 65.6% 64.4% 11-21 32
35 Elena-Gabriela Ruse On serve 8 8.3 -0.3 ±4.4 65.2% 64.1% 11-12 23
36 Elina Avanesyan On serve 10 10.3 -0.3 ±4.9 63.0% 62.0% 12-15 27
37 Hailey Baptiste On serve 13 13.3 -0.3 ±5.7 66.7% 65.9% 20-19 39
38 Anastasia Potapova On serve 14 14.2 -0.2 ±5.8 64.1% 63.5% 23-16 39
39 Naomi Osaka On serve 16 16.2 -0.2 ±6.2 64.4% 64.0% 30-15 45
40 Diana Shnaider On serve 19 19.2 -0.2 ±6.7 64.2% 63.8% 28-25 53
41 Clara Tauson On serve 19 19.2 -0.2 ±6.9 67.2% 66.9% 36-22 58
42 Dayana Yastremska On serve 19 18.8 +0.2 ±6.6 61.2% 61.6% 28-21 49
43 Coco Gauff On serve 17 16.7 +0.3 ±6.5 69.6% 70.2% 42-14 56
44 Ons Jabeur On serve 10 9.8 +0.2 ±4.8 63.0% 63.9% 14-13 27
45 Eva Lys On serve 13 12.6 +0.4 ±5.6 66.7% 67.7% 21-18 39
46 Elina Svitolina On serve 17 16.6 +0.4 ±6.3 63.8% 64.8% 33-14 47
47 Barbora Krejcikova On serve 9 8.6 +0.4 ±4.5 60.9% 62.5% 14-9 23
48 Magda Linette On serve 18 17.4 +0.6 ±6.3 59.1% 60.5% 20-24 44
49 Qinwen Zheng On serve 9 8.5 +0.5 ±4.8 71.9% 73.3% 20-12 32
50 Amanda Anisimova On serve 21 20.3 +0.7 ±7.0 63.8% 65.0% 43-15 58
51 Katerina Siniakova On serve 15 14.3 +0.7 ±5.8 60.5% 62.3% 17-21 38
52 Suzan Lamens On serve 16 15.2 +0.8 ±6.0 61.9% 63.8% 19-23 42
53 Emma Navarro On serve 21 20.0 +1.0 ±6.8 61.1% 62.9% 30-24 54
54 Ajla Tomljanovic On serve 14 13.2 +0.8 ±5.7 64.1% 66.2% 17-22 39
55 Daria Kasatkina On serve 16 15.0 +1.0 ±6.0 61.9% 64.2% 21-21 42
56 Ashlyn Krueger On serve 17 15.9 +1.1 ±6.1 60.5% 63.1% 21-22 43
57 Bernarda Pera On serve 9 8.2 +0.8 ±4.4 59.1% 62.9% 7-15 22
58 Olga Danilovic On serve 13 11.9 +1.1 ±5.3 60.6% 63.9% 16-17 33
59 Elisabetta Cocciaretto On serve 14 12.8 +1.2 ±5.5 60.0% 63.4% 15-20 35
60 Katie Boulter On serve 12 10.9 +1.1 ±5.0 58.6% 62.4% 11-18 29
61 Danielle Collins On serve 11 9.9 +1.1 ±4.9 62.1% 65.9% 15-14 29
62 Iva Jovic On serve 9 8.0 +1.0 ±4.4 62.5% 66.6% 14-10 24
63 Magdalena Frech On serve 16 14.6 +1.4 ±5.9 60.0% 63.4% 15-25 40
64 Sofia Kenin On serve 22 20.4 +1.6 ±6.8 57.7% 60.8% 27-25 52
65 Paula Badosa On serve 12 10.7 +1.3 ±5.0 60.0% 64.2% 18-12 30
66 Victoria Mboko On serve 11 9.7 +1.3 ±4.9 63.3% 67.6% 20-10 30
67 Yue Yuan On serve 11 9.5 +1.5 ±4.7 59.3% 64.8% 9-18 27
68 Lulu Sun On serve 14 12.3 +1.7 ±5.4 58.8% 63.9% 13-21 34
69 Lucia Bronzetti On serve 16 14.0 +2.0 ±5.7 57.9% 63.2% 15-23 38
70 Marketa Vondrousova On serve 12 10.2 +1.8 ±4.9 57.1% 63.5% 18-10 28
71 Kimberly Birrell On serve 14 12.1 +1.9 ±5.3 56.2% 62.3% 14-18 32
72 Jessica Bouzas Maneiro On serve 19 16.7 +2.3 ±6.2 57.8% 62.9% 24-21 45
73 Beatriz Haddad Maia On serve 16 13.9 +2.1 ±5.7 57.9% 63.5% 16-22 38
74 Ella Seidel On serve 9 7.4 +1.6 ±4.2 55.0% 63.0% 12-8 20
75 Katie Volynets On serve 13 11.0 +2.0 ±5.1 59.4% 65.6% 11-21 32
76 Yuliia Starodubtseva On serve 11 9.2 +1.8 ±4.7 57.7% 64.7% 9-17 26
77 Anastasia Pavlyuchenkova On serve 11 9.1 +1.9 ±4.6 54.2% 62.0% 11-13 24
78 Mirra Andreeva On serve 19 15.7 +3.3 ±6.4 66.1% 71.9% 40-16 56
79 Alexandra Eala On serve 9 6.8 +2.2 ±4.1 57.1% 67.6% 9-12 21
80 Anna Kalinskaya On serve 21 17.6 +3.4 ±6.3 53.3% 61.0% 25-20 45
81 Karolina Muchova On serve 17 13.7 +3.3 ±5.8 56.4% 64.9% 24-15 39
82 Rebecca Sramkova On serve 20 16.3 +3.7 ±6.2 55.6% 63.9% 20-25 45
83 Anastasija Sevastova On serve 10 7.4 +2.6 ±4.1 50.0% 63.2% 9-11 20
84 Tatjana Maria On serve 15 11.6 +3.4 ±5.2 54.5% 64.9% 14-19 33
85 Varvara Gracheva On serve 15 11.4 +3.6 ±5.1 50.0% 62.0% 12-18 30
86 Emiliana Arango On serve 12 8.7 +3.3 ±4.5 53.8% 66.7% 12-14 26
87 Peyton Stearns Wild card 17 12.8 +4.2 ±5.5 52.8% 64.6% 16-20 36
88 Anna Bondar Wild card 13 9.3 +3.7 ±4.7 48.0% 62.7% 10-15 25
89 Moyuka Uchijima Wild card 17 12.3 +4.7 ±5.5 54.1% 66.8% 12-25 37
90 Sonay Kartal Wild card 15 10.4 +4.6 ±5.0 48.3% 64.2% 15-14 29
91 Anhelina Kalinina Wild card 12 7.8 +4.2 ±4.3 42.9% 62.7% 9-12 21
92 Lois Boisson Wild card 11 6.9 +4.1 ±4.1 45.0% 65.7% 14-6 20
93 Liudmila Samsonova Wild card 26 19.1 +6.9 ±6.7 50.9% 64.0% 30-23 53
94 Alycia Parks Wild card 21 14.7 +6.3 ±5.8 44.7% 61.3% 15-23 38
95 Laura Siegemund Wild card 14 8.9 +5.1 ±4.6 48.1% 67.2% 13-14 27
96 Zeynep Sonmez Wild card 14 8.7 +5.3 ±4.5 41.7% 63.7% 11-13 24
97 Donna Vekic Wild card 20 12.6 +7.4 ±5.5 42.9% 63.9% 14-21 35
98 Jelena Ostapenko Wild card 21 13.2 +7.8 ±5.6 41.7% 63.4% 17-19 36

Serve projections: who we read best, and worst

Before every match we project how many aces each player will serve. The bar is how much closer that projection lands than the naive forecast anyone can build without a model, which is the average of the player's own last 5 matches. Positive means we add something on that player. Ranking by raw error would just rank the tour by serve volume, so the comparison is always against each player's own baseline.

We read it our projection beats the player's own last-5 average, and the whole 95% interval agrees Level the gap is inside the noise for this sample, where most players belong Baseline wins the last-5 average reads this player better than we do

Full board below, 110 players with at least 12 matches in 2025 where both our projection and the last-5 baseline could be scored (source: out-of-sample backtest). A verdict is printed only when the 95% interval of the per-match comparison stays on one side of zero. Everyone else is level with the baseline for this sample, which is where most players belong.

# Player Verdict Edge Our miss Baseline miss Actual avg Projected avg M

How this is measured, and what it does not claim

An upset is a match where our model's favorite loses. The model does not expect zero upsets: a 55/45 call is expected to go wrong 45 times out of 100. Adding those probabilities across a player's schedule gives their expected upsets, tailored to the exact opponents they faced. The board compares that number with the upsets that actually happened: fewer than expected earns Straight sets, more than expected earns Wild card, and anything within the statistical noise for that sample stays On serve, which with 60-plus matches is where most players genuinely belong.

Two independent model vintages agree on who broke script within a season, but a wild-card season does not predict a wild-card next season. That is why this is a season report, not a career trait, and why sample sizes and 95% intervals are always shown. The full reasoning, including the proper-scoring-rule version of this metric (the Brier delta) that backs the verdicts, is in the explainer.

The serve board answers a different question with a different metric, and the two never mix. Upsets are about who wins, and are scored against the model's own expectation. Aces and double faults are counts, and are scored against the forecast anyone could make without a model, the average of that player's last five matches. A player can be perfectly on script and still be the one whose serve we read worst.

Recent seasons use our as-published daily record (the same reconciled predictions behind the performance page); earlier seasons use a strict out-of-sample backtest of the current model. One source per season, never mixed. Probabilities and calibration are public on the model transparency page.