TennisEdge's AI model finds no betting value in E. Cocciaretto vs N. Osaka (Washington on 2026-07-31) — the market is efficient, with N. Osaka favoured at 1.13.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | E. Cocciaretto | N. Osaka |
|---|---|---|
| Elo ratings-based, surface-weighted |
35.3% | 64.7% |
| Machine learning 29-feature gradient-boosted model |
17.5% | 82.5% |
| Serve / return point-by-point simulation |
38.6% | 61.4% |
| Blend what the bot actually prices off |
26.1% | 73.9% |
| Market implied de-vigged from the opening odds |
16.5% | 83.5% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| E. Cocciaretto | clay | 1690 | 94–47 | 141 |
| N. Osaka | hard | 1920 | 80–36 | 116 |
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 May 26, 2026.
N. Osaka leads 2–0 over 2 meetings. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| July 31, 2026 | Washington | hard | N. Osaka | 2 - 1 |
| March 21, 2024 | WTA Miami | hard | N. Osaka | 2 - 0 |
| L | T. Gibson | 1 - 0 | Aug. 3, 2026 |
| L | N. Osaka | 1 - 2 | July 31, 2026 |
| W | E. Navarro | 1 - 2 | July 29, 2026 |
| W | C. Tauson | 2 - 1 | July 27, 2026 |
| L | Xin. Wang | 2 - 1 | June 29, 2026 |
| L | A. Eala | 0 - 2 | Aug. 1, 2026 |
| W | E. Cocciaretto | 1 - 2 | July 31, 2026 |
| W | A. Krueger | 0 - 1 | July 29, 2026 |
| L | K. Muchova | 0 - 2 | July 7, 2026 |
| W | A. Sabalenka | 0 - 2 | July 5, 2026 |
All July 31, 2026 results & picks →
E. Cocciaretto (A, rank 41, age 25) vs N. Osaka (B, rank 16, age 28) — Washington, hard, QF MODELS (prob A) — ELO 35.3% | ML 17.5% | blend 26.1% | book-implied 16.5% EXECUTABLE PRICES — Betfair A 3.85 (impl ~26.0%), Betfair B 1.16 (impl ~86.2%). Book 5.7/1.13 is stale; the value_edge 0.096 was computed off that stale book price and evaporates against the exchange. RECENT FORM Cocciaretto (last 5): W vs Navarro 1725 (3-6 6-4 3-6, GRIND, 3 sets); W vs Tauson 1795 (4-6 7-6 7-6, GRIND, double TB); L vs Xin.Wang 1811 (grass); L vs Tomljanovic 1746 (grass); L vs Korneeva 1592 (clay). Shape: match-sharp this week but SHARPENING-by-grinding — two back-to-back 3-setters over Strong opponents, high physical cost. Path: 2 matches deep at venue = played-in. Osaka (last 5): W vs Krueger 1727 (opp retired, 49 min); L vs Muchova 1880 (grass QF); W vs Sabalenka 1942 (grass); W vs Kasatkina 1669; W vs Gasanova 1683. Strong grass run (beat Sabalenka), now hard. Path: 1 easy match at venue, fresh. FATIGUE Cocciaretto: won a 3-SET match YESTERDAY (vs Navarro) after a double-TB 3-setter two days prior — extreme spot + calibrated 3-set-hangover haircut applies (-4.5pp, hard-court, age 25 baseline). 242 court-min over 2 days. Osaka: 49 min yesterday (retirement), fully rested. Fatigue clearly favours Osaka. NARRATIVE / ELO SANITY Strong-rank/low-ELO on Osaka: rank 16 but ELO gives her only ~64.7% vs a rank-41 opponent — her ELO is understated (comeback/rebuild profile, beat Sabalenka on grass). So the lone A-favouring model (ELO 35.3%) is contaminated; do NOT lean on it. ML (17.5%) and blend (26.1%) sit at/below the exchange. Both transitioned grass->hard but each already has a hard match this week; minor. PRICING Back A (Cocciaretto) @ Betfair 3.85: eff odds 3.71, implied 27.0%. Blend 26.1% -> edge -0.9%; after the -4.5pp fatigue haircut true ~21.6% -> edge -5.4%. Negative. Her price is too SHORT once the 3-set hangover is priced, so no fatigue-drift value on the tired side here. Back B (Osaka) @ Betfair 1.16: eff odds 1.152, implied 86.8%. Most bullish model ML 82.5% -> edge -4.3%; blend 73.9% -> edge -12.9%. Negative. AH: not quoted (alt_markets.handicap null). GATES G1 edge-origin: FAIL — no numerical model-vs-executable-price gap either side (A blend edge -0.9%, B best-model edge -4.3%); the only A-favouring number (ELO) is contaminated and still below the exchange after discount. G2 consensus: stake-tier only — n/a, no qualifying edge. G3 fatal-risk: n/a (no margin bet). VERDICT: PASS. Exchange price is efficient — model blend equals Betfair on A, and every model prices Osaka below her exchange implied. The stale book line created a phantom edge; the real market has none. Fatigue on Cocciaretto only makes her fair-priced-to-short, not value.