TennisEdge's AI model finds no betting value in T. Svajda vs E. Ymer (Washington on 2026-07-25) — the market is efficient, with E. Ymer favoured at 1.51.
T. SVAJDA vs E. YMER — Washington (hard) ═══════════════════════════════════════ MODELS & MARKET - Book: Svajda 2.39 / Ymer 1.51 (implied A 38.7%) — value_pick A, +7.3% vs book - Betfair exchange: Svajda 1.92 / Ymer 2.06 — near pick'em, SVAJDA slight fav - Executable (comm-adj): Svajda eff 1.874 (impl 53.4%), Ymer eff 2.007 (impl 49.8%) - ELO 50.9% A | ML 57.5% A | Combined 46.0% A — models scattered around the exchange price - The book/exchange gap means the book 2.39 on Svajda is a stale quote; edge must be tested vs 1.92. At 1.92 the "value on A" evaporates: combined model 46% is BELOW the 53.4% executable bar; only the ML sub-model (57.5%) clears it, and it is contradicted by both ELO and the combined number. ELO vs RANKING SANITY - Svajda: ELO-competitive vs Ymer but ranked 387 — rating built on an M25 Dallas title run (opponents ELO 1261-1587) and challenger fields. Farmed-rating flag → ELO/ML edge on Svajda discounted. - Ymer: ranked 171, age 30 — rank/ELO roughly consistent, but see layoff below. RECENT FORM (top 5 shown) Svajda (20): 07-15 Lincoln L Lajal(1658) 3 sets GRIND | 07-12 M25 Dallas F W Kim(1587) | 07-11 SF W Zhu(1374) | 07-10 QF W Forbes(1484) | 07-09 R16 W Petrovic(1350) 3 sets - TITLE: M25 Dallas 13d ago (hard). 6-1 in last 7, but essentially all vs Weak/Avg opposition; lost both recent matches vs 1600+ (Lajal, Butvilas, Smith). Zero DOM wins vs Strong+. Ymer (30): 06-24 Wimb-quali SF L Skatov(1541) | 06-22 QF W Basilashvili(1639) GRIND | 06-16 Dublin L Searle(1512) double-TB | 06-08 s-Hertogenbosch L Humbert(1634) | 06-07 F W Visker(1589) - LAST MATCH 31 DAYS AGO — long layoff. Entire recent block on GRASS → TRANSITION to hard. Mixed 6-6-ish form, grinds vs Strong opposition. FATIGUE - Svajda: 3 matches/14d, 329 min, last 07-15 → 10 days rest, no fatigue. Age 20, recovers fast anyway. - Ymer: 0 matches/14d, 31-day layoff → fully rested but rust + grass→hard transition risk. Age 30. - Fatigue edge: none (neither side is loaded). NARRATIVE - Svajda TITLE M25 Dallas 13d ago — but a Won-Final 13d back is outside letdown windows; None. - Ymer TRANSITION grass→hard + layoff — depresses level and widens margin variance. - Venue: no Washington history for either in fetched window. ALT MARKETS: handicap/totals null — ML only. GATES G1 edge-origin: FAIL — vs executable 1.92 Svajda needs >53.4%; combined model says 46.0%. Book 2.39 is stale, not value. Ymer side: 54% (combined) vs 49.8% executable = ~4%, below 5% bar and contradicted by ML (42.5%). G2 consensus: FAIL — models internally split (ELO/ML vs combined); no independent hard evidence (no H2H, no live this-event results) validating either side against the exchange. G3 fatal-risk: FAIL — Svajda thesis: farmed M25 ELO, 0 wins vs 1600+ recently. Ymer thesis: 31-day layoff + surface transition. Each side's own analysis negates its thesis. VERDICT: PASS Near-pick'em exchange price with scattered models, a farmed-rating flag on the ELO/ML side and a layoff/transition flag on the other. No stakeable edge in any market.