TennisEdge's AI model finds no betting value in J. Vandromme vs V. Valdmannova (Warsaw on 2026-08-03) — the market is efficient, with J. Vandromme favoured at 1.67.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | J. Vandromme | V. Valdmannova |
|---|---|---|
| Elo ratings-based, surface-weighted |
57.0% | 43.0% |
| Machine learning 29-feature gradient-boosted model |
36.5% | 63.5% |
| Serve / return point-by-point simulation |
54.2% | 45.8% |
| Blend what the bot actually prices off |
57.0% | 43.0% |
| Market implied de-vigged from the opening odds |
56.8% | 43.2% |
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. Vandromme | hard | 1778 | 66–19 | 85 |
| V. Valdmannova | hard | 1730 | 63–30 | 93 |
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 June 19, 2026.
| L | J. Grabher | 2 - 0 | July 17, 2026 |
| W | P. Hercog | 0 - 1 | July 16, 2026 |
| W | V. Wolff | 0 - 2 | July 14, 2026 |
| L | M. Sherif | 2 - 1 | July 12, 2026 |
| W | A. Zantedeschi | 2 - 0 | July 11, 2026 |
| W | M. Podlinska | 0 - 2 | Aug. 4, 2026 |
| W | A. Rajecki | 2 - 0 | Aug. 2, 2026 |
| W | V. Ryser | 2 - 1 | Aug. 1, 2026 |
| W | V. Allen | 2 - 1 | July 31, 2026 |
| W | I. L. Van den Broek | 2 - 1 | July 30, 2026 |
All Aug. 3, 2026 results & picks →
═══ J. Vandromme vs V. Valdmannova — Warsaw (hard) ═══ MODELS ELO A 57.0% | ML A 36.5% | model A 57.0% | market (devig) A 56.8% ELO (57%) ≈ market (57%) on A; ML is the LONE dissenter (A 36% / B 63%). ELO ≈ ranking (179 vs 218) → ELO credible; market efficient on the favourite. RECENT FORM (last 5 shown) Vandromme (18, #179): last match 2026-07-17 (~16d idle → RUST). Recent all CLAY. - 07-17 Kitzbuhel-q QF clay LOST J.Grabher (1653) - 07-16 Kitzbuhel-q 1/8 clay WON P.Hercog (1500) - 07-14 Kitzbuhel-q 1/16 clay WON V.Wolff (1500) GRIND - 07-12 Contrexeville FINAL clay LOST M.Sherif (1755) — reached final - 07-11 Contrexeville SF clay WON A.Zantedeschi (1593) DOM Shape: STABLE. Clay->hard TRANSITION + rust (16d) = level/margin risk. Valdmannova (18, #218): W50 Dublin CHAMPION, 5 matches/7d, 3 straight 3-setters. - 08-02 Dublin FINAL WON A.Rajecki (1500) — TITLE today - 08-01 Dublin SF WON V.Ryser (1500) 3-set - 07-31 Dublin QF WON V.Allen 3-set - 07-30 Dublin 1/8 WON Van den Broek 3-set - 07-29 Dublin 1/16 WON A.Reva (1500) 6-2 6-2 DOM Shape: match-sharp but grinding (three 3-setters vs ~1500 field). FATIGUE / NARRATIVE Valdmannova: EXTREME spot — back_to_back=3, matches_7d=4, 435 court-min, 3-set win YESTERDAY (−4.5pp hangover haircut applies). Age 18 softens one tier but load is extreme. Plus TITLE LET-DOWN (won Final 0-1d ago = LETDOWN). Vandromme: well-rested to the point of rust (16d off) + surface transition. Fatigue points AGAINST Valdmannova, not for anyone's bet numerically. GATES G1 edge-origin: FAIL — no numeric edge exists pre-narrative for a stakeable side. Backing A: ELO 57% vs implied ~60% = NEGATIVE edge; any A edge is manufactured from opponent fatigue (definitional G1 fail). Backing B: ML +18% exists but the qualitative read (extreme fatigue + title let-down) contradicts it entirely. G2 consensus: stake-tier only — lone-ML split; would be capped regardless. G3 fatal-risk: ML market, non-blocking. MARKET CHOICE alt_markets: none at analysis time AH: not quoted ML only. Vandromme @1.67 book implied ~60% > ELO 57% → no value. Valdmannova @2.20 book implied ~45%; ML says 63% but backing a title-winner through three 3-setters into a fatigue wall is exactly the side qualitative analysis vetoes. No exchange price (thin ITF liquidity) → book-priced, prefer PASS. VERDICT: PASS. Neither side pairs numeric edge with qualitative validation — favourite has support but negative edge (market efficient), dog has ML edge but the read points the wrong way. Textbook pass.