Our Iva Jovic vs Harriet Dart prediction for Friday, October 2, 2026 starts with a number that looks like a typo: Jovic is -1667 on the moneyline heading into their 11:00 PM ET match in Beijing. That is about as lopsided as tennis pricing gets, and the model's own read of the two games actually agrees the gap is real. But this is a PASS on the official card. No official action is going out on this match, and that call stands on its own regardless of how good the matchup looks on paper.
Here is why the numbers point one way while the card stays closed.
Why the model leans Jovic
Jovic walked out of Guadalajara with a WTA 500 title as the No. 2 seed without dropping a set, and two of those four wins were the kind that hold up: a quarterfinal romp over Frech where she won 61% of total points and saved all eight break points she faced, and a final win over Stearns where she won 56.5% of points and faced zero break points. That sits on top of a 12-month hard-court record of 72% service games held and 39% return games won against mostly mid-tier opposition, good for 26-12 over that stretch. Before Guadalajara she'd also come back from 2-4 down in the third set to beat Eala at the US Open.
Dart's hard-court record over the same long window is 61% holds and 44% return games won, built mostly against weaker fields, and her three wins to reach this match in Beijing came against a qualifier, a wildcard ranked outside the top 1,000, and a No. 1297 debutant playing her first WTA main-draw match. Against that debutant, Dart won just 53.3% of total points across three sets. None of that moves the needle the way Jovic's level does.
The number that actually decides this one
The model isn't just leaning on form, it's leaning on a specific matchup inside the match: Dart's second serve against Jovic's return. Dart wins only 41% of points on her second serve and has been handing out double faults at a 6.2% rate, while Jovic's return has eaten second serves at a 57% rate this season. Dart's return numbers look stronger on the surface, but against servers rated like Jovic she's converting only 33% of return games, not the 44% her overall hard-court number suggests. Run it forward and the model projects Jovic holding serve at roughly a 70% clip in this match against roughly 53% for Dart. That's the core of the edge, not just a seed beating a lower-ranked player.
The case for caution
The model's own risk notes flag two real issues with Jovic coming in. She hasn't played a match since the Guadalajara final on September 19, so she's carrying 14 days of rest plus the travel from Mexico to China into this one, and her first serve has been dipping to 58-59% over the past three months and the past month specifically. That dip showed up in the box scores: Sonmez broke her four times in the Guadalajara second round, and she faced 12 break points against Bucsa in the semifinal. Dart's best tool is exactly the kind of weapon that preys on a shaky second serve, she's winning 60% of her second-serve return points, and if Jovic's delivery is off early this could stretch into a third set the way Dart's US Open comeback over Stearns did.
Dart also arrives with rhythm but not rest. She's played three matches in four days in Beijing, about 348 minutes of tennis, 183 of them in a three-set, three-hour win over that debutant two days before this one. Fresh legs against a tired arm is part of why the price sits where it does, and part of why the model isn't willing to call the gap wider than its own numbers support.
The verdict
This is a PASS. The model's read of the two games favors Jovic, and nothing above changes that read, but we are not putting out a bet on this match. Treat it as an evaluation, not a play, and look for the next card update.
- Event · Beijing · WTA (2026) · Friday, October 2, 2026 · 11:00 PM ET
- Market · Iva Jovic -1667 — recorded when the pick was posted
- Model · T Money, Botcappers' AI Tennis model — how it is built and what it does not know
- Record · every pick is timestamped before the event and graded in public, win or lose
- Statistics · the model's own match/game database, current to the posting time above. Figures a reader cannot open here are internal to that database, not published third-party data.

