TennisEdge's AI model finds no betting value in S. Williams vs M. Joint (Wimbledon on 2026-06-29) — the market is efficient, with M. Joint favoured at 1.72.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | S. Williams | M. Joint |
|---|---|---|
| Elo ratings-based, surface-weighted |
22.8% | 77.2% |
| Machine learning 29-feature gradient-boosted model |
36.4% | 63.6% |
| Blend what the bot actually prices off |
34.6% | 65.4% |
| Market implied de-vigged from the opening odds |
45.0% | 55.0% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| S. Williams | hard | 1788 | 17–5 | 22 |
| M. Joint | hard | 1725 | 84–53 | 137 |
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 Sept. 3, 2022.
M. Joint leads 1–0 over 1 meeting. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| June 30, 2026 | Wimbledon | grass | M. Joint | 2 - 1 |
| L | M. Joint | 1 - 2 | June 30, 2026 |
| L | A. Tomljanovic | 2 - 1 | Sept. 3, 2022 |
| W | A. Kontaveit | 2 - 1 | Sept. 1, 2022 |
| W | D. Kovinic | 0 - 2 | Aug. 30, 2022 |
| L | E. Raducanu | 0 - 2 | Aug. 17, 2022 |
| W | Y. Starodubtseva | 1 - 2 | Aug. 3, 2026 |
| W | A. Grubor | 2 - 1 | Aug. 1, 2026 |
| L | K. Liutova | 2 - 1 | July 30, 2026 |
| W | W. Osuigwe | 2 - 0 | July 27, 2026 |
| L | T. Valentova | 1 - 2 | July 22, 2026 |
All June 29, 2026 results & picks →
S. Williams vs M. Joint — Wimbledon (grass), 1/64-finals MODELS: ELO p(A)=0.228 | ML p(A)=0.364 | Blend p(A)=0.346 (Joint 0.654). Market: Joint 1.72 (impl 0.55), Williams 2.10 (impl 0.45). Value pick B, edge +10.4%. ALL THREE models AND the market agree Joint is favored — consensus, not disagreement. The numeric edge is ELO being more bullish on Joint than the market, not a model-vs-market split. FORM (S. Williams): NO matches in 90d and empty form_analysis — zero live read. Cannot validate the model's favoring of Joint nor assess Williams. High data uncertainty (math_only prediction). FORM (M. Joint): 1 win in last 8. On GRASS specifically: L Eastbourne 6-7 4-6 to E. Arango (1427, Weak), L Nottingham 7-6 5-7 4-6 to Starodubtseva (1561, Avg). Both grass losses, one to a sub-1450 opponent. Only quality result is a clay DOM 7-5 6-1 over Kawa (1698, Strong) at Makarska — surface mismatch. Recent record otherwise: L vs Kabbaj (1504), L 1-6 2-6 vs Potapova (1912), L 1-6 2-6 vs A.Li (1720), L vs Paquet (1479), L vs Golubic (1558). FATIGUE: Williams well-rested (0 matches). Joint 8d since last match (2026-06-22), 2 matches/14d, 187min court time — adequately rested, no fatigue edge either way. NARRATIVE: No recent finals → no let-down. No Wimbledon venue history. No title recency. Live grass form for the model-favored side (Joint) is poor — losing to weak/avg grass opponents undercuts the ELO-driven edge rather than validating it. VERDICT: PASS. Model+market consensus on Joint (no genuine split). The +10.4% edge rests on a stale-looking ELO gap, and the qualitative read CONTRADICTS it: Joint's live grass form is weak (lost to a 1427), while we have zero data on Williams. Unvalidated edge + low confidence (0.41) + no read on the opponent = no bet.