MODEL PREDICTION · 2026-07-27

A. Potapova vs V. Williamsprediction

Result pending
POTAPOVAWIN PROBABILITYWILLIAMS
87%
model prob.
@1.15
odds · 87% impl.
🎾Serve 56%📈Form 7/10
THE MODEL'S REASONING

Ranking: #28 vs #471 (better ranked)

Recent form: 7/10 in recent matches

WATCH FOR

!Returning from a long layoff (28d) — possible rustiness

Calibrated model probability (~64% out-of-sample accuracy, validated specifically on WTA). Not a guarantee: the model ≈ the market on average, so the odds already capture almost all the edge. 18+ · gamble responsibly.
@1.15
fair odds
+0.4%
expected value
HOW EACH FACTOR MATTERS
Level (Elo/ranking)▸ Potapova●●●
Potapova is ranked #28 vs #471, and the baseline model gives her 61% vs 8% — a wide gap in overall level.
Form▸ Potapova●●
7-3 in her last 10 with wins over Gauff (Elo 1962) and Muchova (Elo 1952), though she is on a 1-match losing streak.
Rest▸ Williams●●
28 days off court with 0 matches in the last 14 days — a long layoff that can produce rustiness for Potapova despite her favorite status.
Serve/return▸ Potapova
Potapova holds serve at 56% and wins 36% of return points, a solid all-around profile; no comparable numbers exist for Williams.
RANKING GAP

The clearest signal in this match is the level differential: Potapova sits at #28 in the world, while the model's own reference puts Williams far outside the top tiers at #471. That gap is echoed in the baseline probabilities, 61% for Potapova against just 8% for Williams before any adjustments — a spread wide enough that most of the model's confidence stems from raw level rather than situational factors.

This kind of gap typically translates into a lopsided expected points differential across a match, since the better-ranked player is expected to control rallies and service games more consistently, even without surface or head-to-head data to refine it further.

FORM AND MOMENTUM

Potapova's recent form is a genuine positive: 7 wins in her last 10 matches, including notable results over Coco Gauff (Elo 1962) and Karolina Muchova (Elo 1952) — both meaningfully higher-rated opponents. Beating players of that caliber suggests she can raise her level against strong competition, which matters here given the ranking gap already favors her.

That said, she arrives on a 1-match losing streak, a small but real caveat. No form data exists for Williams, so the model cannot weigh her recent trajectory, leaving Potapova's form edge somewhat one-sided by necessity rather than by direct comparison.

LAYOFF RISK

The main risk flag here isn't about the opponent but about Potapova herself: she has been out for 28 days with zero matches played in the last two weeks. Layoffs of this length can affect timing and rhythm, particularly on return games where reflexes and reads develop through match play.

This is a context risk, not a probability adjustment — the model does not know how it will manifest, only that it exists. It's worth keeping in mind as a source of variance around an otherwise strong favorite profile.

SERVE-RETURN PROFILE

Potapova's own service and return numbers — 56% of service points won and 36% of return points won — depict a player competent on both sides of the ball, without a glaring hole in her game plan. However, since no equivalent serve or return percentages exist for Williams, this factor stands alone rather than as a direct comparison, limiting how much weight it can carry analytically.

VALUE READ

The model rates Potapova's win probability at 87%, essentially in line with the market's implied 88% at odds of 1.14. The resulting expected value is -0.4%, meaning this is not a value bet — the market has already priced in the same level gap and form edge that the model identifies.

Potapova is clearly the more probable winner on the numbers, but being the favorite is not the same as offering value. At these odds, backing her offers no edge over what the market already reflects, and the layoff risk adds a layer of uncertainty that isn't fully captured in either number.

Impact and analysis from real match data (Elo, form, head-to-head, rest, surface vs baseline, weather, altitude). The model ≈ the market on average; the odds already capture almost all the edge. 18+ · gamble responsibly.

Analyze today's matches →