Zheng, future mistress of Rome?

There’s something almost unfair about how this match is being presented to the general public. Bondar painted as the outsider, Zheng as the clear favorite, and yet the 52-week clay stats tell a much more nuanced story. The 28-year-old Hungarian shows a Match Efficiency on clay of 2.441, slightly above the Chinese player’s 2.336. Her Dominance Efficiency climbs to 1.419 versus 1.069 for Zheng. She wins 76.1% of her service games on clay over this period, saves 65.6% of break points, and posts 58.3% in tie-breaks compared to a disastrous 25% for Zheng on the same surface. She’s not a player riding a lucky wave since Madrid, she’s a clay specialist in full control of her environment. Nineteen Challenger and ITF titles, sixteen of them on clay, a career built entirely around the red dirt. Her run in Madrid — beating Svitolina 6-3 6-4, Samson 7-6 6-1, and a courageous loss to Andreeva 6-7 6-3 7-6 — was no accident.

Zheng comes in with undeniable surface pedigree, 69.1% career WTA win rate on clay, a semifinalist here in 2025 after beating Sabalenka in the quarters, but also with an extremely small 2026 clay sample. Two matches on clay this season: the first lost to Rybakina after snatching the opening set, the second a labored win over Kenin with a first set dropped 1-6. Her return from elbow surgery is still visible in the serving numbers: just 54.7% first serves in, well below her usual standards, and that disastrous 25% rate in clay tie-breaks says something deep about how she handles key moments when her timing isn’t fully back yet. On return, Bondar wins 37.3% of points on her opponent’s first serve versus 43.2% for Zheng, a real but limited gap in the face of the Chinese player’s power when she lands her first serve with authority.

The H2H reads 2-0 for Zheng, with the only documented meeting having taken place right here in 2023, ending 7-6 6-4, a tight match that Bondar could have swung her way. Rome is the only place where they’ve crossed paths, and the Hungarian knows the script.