Karen Khachanov vs Alex Molčan prediction time: the board has Khachanov priced at -476 to win Friday, October 2 in Beijing, a number that normally ends the conversation before it starts. It doesn't here. The model is marking this one a PASS, which means no official action on either side, and the reason has nothing to do with doubting that Khachanov is the better player on paper.
It has to do with what his own numbers are doing underneath that price, and what Molčan has been doing to get to this match in the first place.
The case for Khachanov
Khachanov's last match was about as clean as a straight-sets win gets: a 6-3, 6-3 dismissal of Felix Auger-Aliassime in 72 minutes on September 30, Beijing's R32. He landed 73.3% of his first serves and won 90.9% of the points behind them, broke three times in four chances, and never faced a break point himself. That's a dominant service performance, and it sits on top of a hard-court hold rate the model has him at 87% for, a real baseline built on more than one afternoon.
He's also rested. His previous singles match before Beijing was all the way back on September 11, so he walked into this event with recovery time most players at this stage of a season don't get.
Why the model isn't chasing the number
Here's the catch: that 73.3% first-serve rate against Auger-Aliassime is a long way above where Khachanov has actually been living. Over his trailing three months, his first-in has run 55.4%, with 76.4% of those first serves won. One strong serving night doesn't overwrite a three-month pattern, it just means he hasn't shown the model yet whether Beijing is a new level or a good day.
The second serve is the bigger tell. Khachanov won only 50% of his second-serve points against Auger-Aliassime, which sits below the roughly 54% to 55% he's posted across his twelve-month and six-month windows. If that number slides back toward his recent baseline instead of holding at Beijing-opener levels, second serves turn into coin flips, and Molčan is built to make him pay for exactly that.
Molčan isn't walking in cold
Molčan's road to this match has been unusual. He retired from a Hagen Challenger match on August 4, then didn't play again for roughly four weeks before a stretch of Davis Cup and tour matches brought him back. Since then he's gone 4-3, and the last three of those wins all came on hard courts in Beijing qualifying and the main draw, over Vít Kopřiva, Marco Trungelliti, and Tomáš Macháč.
Across that three-match Beijing run, his point-weighted serve numbers are 72.4% first-in, 71.3% first-serve points won, and 62.3% second-serve points won, with just 2 double faults across 221 service points. That second-serve number is above the 50% Khachanov just posted in his own opener, and it's the clearest sign that Molčan's return games and service games are both holding up on this surface right now, not just surviving on it.
The workload is real, too. Molčan has played five matches and twelve sets in the last fourteen days, roughly nine hours on court, and he played as recently as October 1, a day closer to this match than Khachanov. That cuts both ways: it's fatigue, but it's also rhythm, and his qualifying matches showed real vulnerability on serve, with his own service games broken down repeatedly even in wins. Neither side of that ledger is settled.
The verdict
Put it together and this reads as a legitimately contestable match wearing a lopsided number. Khachanov's path to winning it clean is straightforward: hold his first-serve percentage near what he showed against Auger-Aliassime, lean on his own second-serve return game to pressure Molčan's service games, and close it out before the match turns into extended baseline rallies. Molčan's path is to keep doing what's worked in Beijing already, protect his second serve at the level he's shown this week, and make Khachanov's first-serve number the whole story.
The model is marking this a PASS. That's not a knock on Khachanov, and it's not a hidden lean toward Molčan either. It just means this isn't a bet we're making at the number the market is offering, and we're not going to pretend otherwise by dressing it up as anything more than what it is.
- Event · Beijing · ATP (2026) · Friday, October 2, 2026 · 1:00 AM ET
- Market · Karen Khachanov -476 — 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.

