TennisEdge's AI model finds no betting value in J. Duckworth vs F. Cobolli (Wimbledon on 2026-07-02) — the market is efficient, with F. Cobolli favoured at 1.36.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | J. Duckworth | F. Cobolli |
|---|---|---|
| Elo ratings-based, surface-weighted |
38.0% | 62.0% |
| Machine learning 29-feature gradient-boosted model |
35.0% | 65.0% |
| Serve / return point-by-point simulation |
25.5% | 74.5% |
| Blend what the bot actually prices off |
31.2% | 68.8% |
| Market implied de-vigged from the opening odds |
28.1% | 71.9% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| J. Duckworth | hard | 1685 | 190–126 | 316 |
| F. Cobolli | clay | 1896 | 132–82 | 214 |
Our own ratings, computed from match results since 2018. Players with fewer than 5 rated matches are omitted rather than shown at a provisional starting value. Last updated Aug. 3, 2026.
F. Cobolli leads 2–1 over 3 meetings. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| July 2, 2026 | Wimbledon | grass | F. Cobolli | 3 - 1 |
| Oct. 17, 2025 | Almaty | hard | J. Duckworth | 2 - 0 |
| Aug. 27, 2024 | ATP US Open | hard | F. Cobolli | 3 - 1 |
| W | C. O'Connell | 2 - 0 | Aug. 3, 2026 |
| L | C. Norrie | 2 - 0 | July 30, 2026 |
| W | A. Mayo | 1 - 2 | July 29, 2026 |
| L | F. Cobolli | 1 - 3 | July 2, 2026 |
| W | T. Griekspoor | 1 - 3 | June 30, 2026 |
| L | R. A. Burruchaga | 0 - 2 | July 15, 2026 |
| L | A. Fery | 0 - 3 | July 8, 2026 |
| W | A. De Minaur | 0 - 3 | July 6, 2026 |
| W | K. Khachanov | 2 - 3 | July 4, 2026 |
| W | J. Duckworth | 1 - 3 | July 2, 2026 |
All July 2, 2026 results & picks →
═══════════════════════════════════════════ J. Duckworth (A) vs F. Cobolli (B) Wimbledon — grass — 1/32-finals — 2026-07-02 ═══════════════════════════════════════════ MODELS (prob A / Duckworth): ELO 0.380 ML 0.350 Serve/Ret 0.255 Blended 0.312 | Market implied A 0.281 Value pick A, edge +3.1% (after 30% shrink) ELO + ML both slightly ABOVE market on A — small consensus dog edge. FORM — Duckworth (A), grass: W Griekspoor (1647 Strong) Wim 1/64 — GRIND (6-4 4-6 7-5 6-4) L Hussey (1427 Weak) Eastbourne SF L Vukic (1636 Avg) London SF L Lehecka (1777 Strong) Stuttgart — GRIND 3-setter W Dedura (1500 Avg) Stuttgart 1/16 Grass record ~2-4. Best results were CLAY (Mexico City / San Luis titles, April). FORM — Cobolli (B): W Navone (1441 Weak) Wim 1/64 — GRIND (4-set, two TBs, 305min) L Tiafoe (1643 Strong) Halle 1/16 FRENCH OPEN FINALIST (L Zverev 5 sets, 6/7) — beat FAA, Svajda, Tien en route Beat Medvedev (Madrid), Zverev (Munich SF) this season. Class ceiling far above A; grass sample thin (real risk). FATIGUE: A: last match 6/30, 4 sets, ~246min, 2 in 14d B: last match 7/1, 4 sets, ~305min, 2 in 14d B one fewer rest day + more court time → slight edge A, but both fresh off long matches. Minor. NARRATIVE: Let-down: B's last Final = French Open (LOST), 25d ago — too old, no fresh letdown. Venue: neither has Wimbledon momentum history. Title recency: none relevant (A's titles were clay, April). VERDICT: PASS Model shows a marginal +3.1% dog edge on Duckworth, but the qualitative read contradicts it: Cobolli is a French Open finalist with a big class/level gap, while Duckworth's recent grass form is poor (2-4) and his form spike was on clay. Cobolli grinds tight matches but wins them. A 3.1% post-shrink edge is not enough to back a dog the eye-test argues against. No actionable edge.