TennisEdge's AI model finds no betting value in R. Jodar vs A. Tabilo (Washington on 2026-08-02) — the market is efficient, with R. Jodar favoured at 1.40.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | R. Jodar | A. Tabilo |
|---|---|---|
| Elo ratings-based, surface-weighted |
61.0% | 39.0% |
| Machine learning 29-feature gradient-boosted model |
65.6% | 34.4% |
| Serve / return point-by-point simulation |
74.4% | 25.6% |
| Blend what the bot actually prices off |
67.5% | 32.5% |
| Market implied de-vigged from the opening odds |
67.8% | 32.2% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
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.
Level at 1–1 over 2 meetings. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| Aug. 2, 2026 | Washington | hard | R. Jodar | 2 - 1 |
| March 6, 2026 | ATP Indian Wells | hard | A. Tabilo | 2 - 0 |
| L | T. Fritz | 2 - 0 | Aug. 3, 2026 |
| W | A. Tabilo | 2 - 1 | Aug. 2, 2026 |
| W | L. Musetti | 2 - 1 | July 31, 2026 |
| W | K. Nishikori | 2 - 0 | July 30, 2026 |
| W | A. Fils | 0 - 2 | July 28, 2026 |
| L | R. Jodar | 2 - 1 | Aug. 2, 2026 |
| W | B. Shelton | 2 - 1 | Aug. 1, 2026 |
| W | T. Atmane | 0 - 2 | July 30, 2026 |
| W | T. Griekspoor | 2 - 1 | July 27, 2026 |
| L | T. Torres | 0 - 2 | July 22, 2026 |
All Aug. 2, 2026 results & picks →
R. Jodar (A, 19, #165) vs A. Tabilo (B, 29, #35) — Washington, hard, semi-final (likely 08-02). MODELS vs MARKET ELO A 0.610 | ML A 0.656 | ensemble A 0.675. Betfair A 1.45 (impl 0.690) / B 3.15 (impl 0.329); book A 1.40. Market sits AT/ABOVE the ensemble on Jodar -> market is more confident in the favourite than the models. No ML value on A. Jodar ML edge = 0.675 - 0.7005 = -2.6% (comm-adj). Market 1.45 < min_odds 1.557 -> auto-PASS. Tabilo ML: ensemble edge -0.3%; ELO edge +6.1% but ELO is the lone/weakest model here and the qualitative read contradicts it. RECENT FORM (last 5) Jodar: W Musetti 1-6 6-1 6-4 (1924, Elite, over-perform); W Nishikori 6-3 6-2 (1752, DOM); W Fils 7-6 6-3 (1954, Elite, GRIND); L Mochizuki (grass, 07-03); W Carreno-Busta (grass). SHARPENING, 3-deep at venue = match-sharp. Tabilo: W Shelton 4-6 7-5 6-4 (1956, GRIND, 08-01); W Atmane (1755, GRIND, tb); W Griekspoor 7-6 4-6 6-4 (1814, GRIND); L Torres (1629, clay); L Rublev SF (clay). All Washington wins are grinding 3-set/tb affairs. PATH & SHAPE Jodar: dream run for a #165 19yo, beating Fils/Nishikori/Musetti in consecutive rounds. SHARPENING, over-performing bracket. High-ELO/modest-rank divergence, but VALIDATED by live elite scalps -> ELO credible, not farmed. Tabilo: also match-sharp (3-deep) but ground through elite fields; heavy load. FATIGUE Jodar: 3 matches, last 07-31 (2 days rest for an 08-02 semi), 293 court-min/14d, age 19 = fast recovery. Fresh. Tabilo: WON a 3-set match 08-01 (vs Shelton) -> -4.5pp 3-set-hangover haircut for an 08-02 semi (surface-agnostic, age 29 baseline). 691 court-min/14d, 15 sets. Fatigue-adj ELO ~0.345 -> Tabilo ML edge collapses to ~+1.6%, inside noise. Net: fatigue favours Jodar, but this is already visible and priced (why the #165 kid is a 69% fav). Confirms market direction; does not create an exploitable edge. NARRATIVE No first-Final let-down. Both surface-consistent (hard now, both came off clay/grass but have 3 hard-court wins this week -> transition dissipated). player_notes: only a generic Day-3 predictions piece for A, no injury/fitness signal. ALT MARKETS / MARKET CHOICE Handicap A -3.5 @ 1.99 (Betfair): sim covers 0.6032, comm-adj impl 0.5153 -> edge 8.8%. Line -3.5 = laying favourite wider than -2.0 = HIGH-VARIANCE tier (needs edge >=10%, hv_min 1.82). 8.8% < 10% bar. Also G3: Jodar's wins are grindy 3-setters (dropped 1-6 first set to Musetti, tb vs Fils) -> a -3.5 blowout cover thesis is self-negated by his own scoreline distribution. Margin veto. Totals under 22.0 @ 2.02: sim under(21.5) 0.532 -> ~0.56 under 22.0, edge ~5.2%. Marginal, no independent margin thesis. Pass. AH: A -3.5 @ 1.99 (high-variance, passed). GATES G1 edge-origin: FAIL — no numeric model-vs-market edge exists pre-narrative on any ML side (Jodar -2.6%, Tabilo -0.3% ensemble); the only >5% number (ELO dog / -3.5 sim) is a lone-model / high-variance artifact, not a staked edge. G2 consensus: PASS (stake-tier only; never blocks) — would be lone_signal tier if bet, but no bet. G3 fatal-risk: FAIL (margin) — Jodar -3.5 cover self-negated by his grindy 3-set profile; hard veto for the only market with a nominal >5% edge. VERDICT: PASS. Efficient market — models cluster 0.61-0.68, Betfair 0.69; the surge-and-fatigue story is already fully priced into a #165-over-#35 favourite. ML no value either side; handicap high-variance below bar + G3 vetoed; totals marginal. No exploitable edge. Predicted winner: R. Jodar.