TennisEdge's AI model finds no betting value in M. Trungelliti vs A. Davidovich Fokina (Umag on 2026-07-15) — the market is efficient, with A. Davidovich Fokina favoured at 1.28.
═══ M. Trungelliti vs A. Davidovich Fokina — ATP Umag R16 (clay) ═══ MODELS vs MARKET Odds: Trungelliti 3.75 (implied 26.7%) | Davidovich Fokina 1.28 (implied 78.1%) ELO: 46.4% A | ML: 34.1% A | Blend: 42.8% A Value flag: A, edge +17.4pts — all three models price the dog above market. RECENT FORM (last 5 shown; all fetched matches used) Trungelliti (A): 07-13 Umag R16 W Jacquet (1603 Avg) 6-2 6-3 DOM 06-29 Wimbledon R128 L Damm (1715 Strong) 4 TB sets GRIND loss, competitive 06-22 Eastbourne L Pinnington Jones (1650) 3 sets 06-21 Eastbourne F L Hussey (1473 Weak) straight sets 06-20 Eastbourne SF W Gill (1433 Weak) incl. TB GRIND Profile: wins almost exclusively vs Weak/Avg (challenger context); losses to sub-1300 (Svajda) and Avg opponents mixed in. One DOM vs Strong (Gojo 1672). Davidovich Fokina (B): 07-05 Wimbledon R16 L Auger-Aliassime (1762 Strong) 5 sets GRIND, elite level 07-03 Wimbledon R32 W Fucsovics (1735 Strong) 7-6 6-2 6-3 GRIND 07-01 Wimbledon R64 W Marozsan (1597 Avg) 6-3 6-0 6-3 DOM 06-29 Wimbledon R128 W Cerundolo (1445 Weak) incl. 7-6 GRIND 06-27 Mallorca FINAL W Quinn (1676 Strong) incl. TB GRIND — TITLE Profile: 8-2 in last 10, beat Dimitrov/Fucsovics/Carreno-Busta, clearly a tier above. FATIGUE A: 1 match / 14d, 3 sets, ~91 min court time, last match 07-13 (fresh, one easy opener) B: 3 matches / 14d, 9 sets, ~445 min court time, last match 07-05 (10 days rest) Court-time diff >60 min against B, but 10 days off largely neutralizes it. Slight edge A. NARRATIVE Let-down: B won Mallorca title 18d ago — outside the 7d window, letdown = None (stale). Venue: A is 1-0 at Umag this week (DOM opener on this court). B: no Umag matches in sample. Transition: B's first clay match since late May after full grass swing — opener rust risk, muted because clay is B's strongest career surface. VERDICT: LEAN Trungelliti — no bet. Model-market disagreement is real (even ML alone: 34% vs 26.7%), but the ELO input is likely contaminated by challenger-context opponents. Qualitative read only half-validates: ADF is rested, native to clay, and operating a full tier above Trungelliti's recent opposition; the dog's angles (surface-switch rust, venue comfort) are soft. True prob for A estimated ~30-32% vs 26.7% implied — thin, uncertain edge that fails the skeptic test.