Paula Badosa is a massive favorite against Justina Mikulskyte in Sao Paulo on Wednesday, September 16, 2026, at 4:30 PM ET, and the market has her at -2500. Any Paula Badosa vs Justina Mikulskyte prediction has to start with that number: it says the book sees almost no path for Mikulskyte at all.
The model likes Badosa's side of this matchup on the merits. It is not backing the bet. This one grades PASS, a side we favor on paper but aren't putting official money behind.
Why Badosa is the right side on paper
Badosa's 12-month hard-court record sits at a rough 6-9 (40%), the kind of number that would normally scare a handicapper off. But that window is stale: it was built against tougher opposition and dragged down by a January-March slump that no longer reflects where she's playing. Zoom out to 24 months on hard courts and she's 25-17 (about 60%), a truer read on her current level.
She also won a WTA 125 title in Bastad, reached the final in Iasi, and made the quarterfinals in Hamburg, going 11-2 after Wimbledon. That stretch was on clay, so treat it as a form and durability signal rather than proof for a hard court in Sao Paulo. The hard-court evidence that matters more directly came right after: at the US Open, she beat Daria Kasatkina 6-1 6-4, putting 65.9% of first serves in against a 56% baseline over the prior 12 months, and winning 64.3% of second-serve points against a 41.8% baseline. Both numbers sit well above where she'd been serving. She followed that with a competitive three-set loss to Coco Gauff, 4-6 7-6(5). That's the same surface as Sao Paulo, and it's the freshest data point she has.
The leak the model can't ignore
Set against that is Badosa's own weak spot: a 65% hold rate with double faults running at 9.5% of her service points. That's enough to gift free break chances to any returner with discipline, and Mikulskyte profiles as exactly that type, breaking 41% of her own return games. Badosa's hard-court losses this year cluster against lower-ranked grinders rather than top players, names like Selekhmeteva, Sasnovich, Putintseva, and Jovic, and that's the same shape of opponent she's facing here. Layer on three retirements in the past twelve months, and there's a live tail risk that this match doesn't finish clean even if Badosa is the better player throughout.
Mikulskyte's route into this draw
Mikulskyte didn't qualify for this event outright. She lost the final round of qualifying 3-6 4-6 to Chloe Paquet two days ago on this same court, then got a late call-up as a lucky loser when another player withdrew from the main draw. This is her first WTA Tour main-draw match at age 27, and no verified top-100 win appears anywhere on her record. Her results this year, including a W75 final in Vitoria-Gasteiz and a WTA 125 quarterfinal in Warsaw where she beat a seeded player before losing to the eventual runner-up, are legitimate form for her level. But they sit a tier below tour competition, and her freshest data point is a straight-sets qualifying loss on this exact court just two days ago.
The number that matters
-2500 is an extraordinarily heavy price for a matchup carrying this much noise: a leaky server against a disciplined returner, a three-retirement injury history, and a lucky loser who still has real hard-court and ITF-level results behind her. That combination is why this one lands as PASS. The model favors Badosa's side of the matchup. We're just not staking it.
The verdict
Call it what it is: PASS. We like Badosa to win this match, but this is not one we're backing with official money. No play on the card today.
- Event · Sao Paulo · WTA (2026) · Wednesday, September 16, 2026 · 4:30 PM ET
- Market · Paula Badosa -2500 — 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.

