Roman Safiullin vs Novak Djokovic: AI Prediction | Games, Spread, Aces & Double Faults

The model’s edge for Djokovic comes mainly from surface fit and overall strength. On paper Djokovic’s grass credentials show through: his Elo sits at 2036 versus Safiullin’s 1825, and his surface strength index (52.34) is markedly higher than Safiullin’s (24.61), which is the single largest driver in the prediction. That advantage is compounded by Djokovic’s higher standing in the rankings (No. 7 versus No. 132) and the model also rewards his overall season strength. Fatigue and recent workload further tilt the scales toward Djokovic — he has logged 522 minutes at the tournament compared with Safiullin’s 1,044 minutes, so Djokovic should be the fresher player going into this tie. Safiullin’s case rests largely on recent form: his form index (96.18) outstrips Djokovic’s (55.49), which is why the model still gives him a roughly one-in-four upset chance. Serve profiles are almost identical in the underlying indices (mean serve ~96 for both), so expect service potency from both men rather than a clear serving mismatch; return indices are close enough that no major gap is flagged. Over their last three matches at Wimbledon both players have advanced comfortably — Djokovic with straight and four-set wins (including Tsitsipas), Safiullin with a notable five-set win earlier and a straight-sets victory in his most recent match — suggesting momentum but also workload differences.