MODEL PREDICTION · 2026-07-20

J. De Jong vs J. Choinskiprediction

Result pending
JONGWIN PROBABILITYCHOINSKI
54%
model prob.
@1.80
odds · 56% impl.
H2H 1–1 JongRest 4d vs 6d🎾Serve 62%📈Form 7/10
THE MODEL'S REASONING

Ranking: #73 vs #75 (better ranked)

Recent form: 5/10 in recent matches

Calibrated model probability (~65% out-of-sample accuracy). Not a guarantee: the model ≈ the market on average, so the odds already capture almost all the edge. 18+ · gamble responsibly.
@1.85
fair odds
−2.9%
expected value
HOW EACH FACTOR MATTERS
Level (Elo/ranking)▸ Jong●●
Signals mixed: Choinski's Elo is higher (1910 vs 1879), but De Jong's ranking (#73 vs #75) and baseline strength (47% vs 30%) push the model to 54%.
Serve/return▸ Choinski●●
Choinski serves better (66% vs 62%) with identical return numbers (39% each), giving him the edge in holding serve.
Head-to-head▸ Choinski
Series tied 1-1, but Choinski won the most recent meeting in 2026, De Jong's win dates back to a 2024 Challenger.
Form= Even
Both arrive on a one-match losing streak with similar recent records (last10: WWWWWLWLWL vs WLWLWWWWWL).
Rest▸ Jong●●
Choinski has played 6 matches in the last 14 days versus De Jong's 2, raising fatigue risk despite an extra rest day (6 vs 4).
LEVEL AND MODEL

The underlying ratings tell slightly different stories. Choinski holds a higher Elo (1910 vs 1879), which typically signals a stronger overall level, but De Jong's ranking edge (#73 vs #75) and a much higher baseline strength score (47% vs 30%) push the composite model toward him. That gap in baseline strength is the largest single number in the data set, and it's the main reason the model lands at 54% for De Jong despite Choinski's Elo advantage.

In short, the numbers don't align uniformly behind one player — De Jong's edge is more about baseline model strength than raw rating superiority, which keeps this a close, competitive matchup on paper.

SERVE VS RETURN

Choinski's serve numbers are the sharpest data point favoring him: he wins 66% of service points compared to De Jong's 62%, a four-point gap that matters over best-of-three sets where holds compound quickly. Since both players return at an identical 39%, this serve differential is not offset elsewhere — it's a clean edge for Choinski in the点 mechanism of who controls more service games.

This means Choinski is likely to face fewer break points per set, putting pressure on De Jong to find breaks against a better server, a tougher ask given the return numbers are level.

FATIGUE AND SCHEDULE

Workload is where De Jong's profile improves. Choinski has played 6 matches in the last 14 days, compared to just 2 for De Jong. That kind of volume can erode serve power and movement over a three-set match, especially against an opponent who has had lighter recent duty.

De Jong's extra rest before this match (though only two days more: 4 vs 6 days since last match) is less significant than the difference in total matches played — the accumulated load on Choinski is the more relevant fatigue signal here.

HISTORY AND FORM

The two have split their previous meetings 1-1, with Choinski taking the more recent encounter in 2026 and De Jong's win coming from a 2024 Challenger event — different tiers, limited signal for today's matchup. Recent form is essentially a wash: both players enter on one-match losing streaks with comparable overall records over their last ten matches.

Neither H2H nor current form provides a decisive lean, so this factor should not be weighted heavily against the serve and workload signals already discussed.

VALUE READ

The model gives De Jong a 54% chance to win, but the market prices him closer to 56% implied probability at 1.79 odds. That gap produces a negative expected value of -3.4%, meaning the price does not offer a backable edge according to this model — the market is, if anything, slightly more confident in De Jong than the model itself.

This is a case where being the 'favorite' does not translate into value. With serve metrics favoring Choinski and workload favoring De Jong, the match reads as close to a coin flip, and the negative EV suggests there's no statistical edge to act on at the current price.

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 →