TennisEdge's AI model finds no betting value in J. Pegula vs D. Shnaider (Washington on 2026-08-01) — the market is efficient, with J. Pegula favoured at 1.31.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | J. Pegula | D. Shnaider |
|---|---|---|
| Elo ratings-based, surface-weighted |
80.4% | 19.6% |
| Machine learning 29-feature gradient-boosted model |
73.4% | 26.6% |
| Serve / return point-by-point simulation |
67.9% | 32.1% |
| Blend what the bot actually prices off |
80.4% | 19.6% |
| Market implied de-vigged from the opening odds |
74.9% | 25.1% |
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. Pegula | hard | 2077 | 192–73 | 265 |
| D. Shnaider | hard | 1841 | 98–62 | 160 |
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.
J. Pegula leads 4–0 over 4 meetings. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| Aug. 1, 2026 | Washington | hard | J. Pegula | 2 - 0 |
| April 3, 2026 | Charleston | clay | J. Pegula | 2 - 1 |
| Sept. 2, 2024 | WTA US Open | hard | J. Pegula | 2 - 0 |
| Aug. 12, 2024 | Toronto | hard | J. Pegula | 2 - 0 |
| L | A. Eala | 1 - 2 | Aug. 3, 2026 |
| W | D. Shnaider | 2 - 0 | Aug. 1, 2026 |
| W | A. Kalinskaya | 2 - 0 | July 31, 2026 |
| W | M. Frech | 2 - 1 | July 30, 2026 |
| L | C. Gauff | 1 - 2 | July 7, 2026 |
| L | J. Pegula | 2 - 0 | Aug. 1, 2026 |
| W | L. Samsonova | 2 - 0 | July 31, 2026 |
| W | A. Potapova | 2 - 0 | July 30, 2026 |
| L | L. Samsonova | 1 - 2 | July 2, 2026 |
| W | E. Lys | 2 - 0 | June 30, 2026 |
All Aug. 1, 2026 results & picks →
J. Pegula (A) vs D. Shnaider (B) - WTA Washington, hard, Semi-finals
MODELS vs MARKET
ELO A 0.804 | ML A 0.734 | model A 0.804 | Betfair A 1.30 (~0.77 impl) / B 3.95 (~0.256)
Models straddle the market (ELO above, ML below) - classic no-edge configuration.
Pegula ML: min_odds = 1.05/0.80 = 1.3125 > executable 1.30 -> FAILS min-odds even at the
most generous ELO prob. Below ELO the gap only widens. No stakeable ML edge on the fav.
Shnaider ML: ML model 0.266 vs implied 0.256 = market essentially fair; ELO 0.196 negative.
Her SHARPENING run is already priced. No ML dog edge.
RECENT FORM (last 5 shown; all fetched used for classification)
Pegula (32, rank 5) - 2 matches at Washington = match-sharp; shape STABLE-to-strong:
07-31 W Kalinskaya (1887 Elite) 6-3 7-5 DOM over Elite = OVER-performing
07-30 W Frech (1739 Strong) 3-6 6-3 6-0 GRIND (dropped set)
07-07 L Gauff (1790) 6-4 3-6 3-6 (grass)
07-05 W Jovic (1922 Elite) 4-6 6-3 6-1 (grass)
07-03 W Bouzas (1734) 6-1 6-3 (grass)
Shnaider (22, rank 20) - 2 matches at Washington = match-sharp; shape SHARPENING:
07-31 W Samsonova (1740 Strong) 6-3 6-4 DOM over Strong = OVER-performing
07-30 W Potapova (1724 Strong) 6-1 6-2 DOM over Strong = OVER-performing
07-02 L Samsonova (1783) 4-6 6-4 2-6 (grass)
06-30 W Lys (1550) 7-5 6-1 (grass)
06-22 L Tauson (1610) 6-4 6-4 (grass)
Both players sharp and grooved on this surface. No rust, no transition risk (both have
2 hard-court wins this week after grass). ELO ~ ranking for both -> ELO credible.
FATIGUE (Step 2)
Both played a 2-SET win yesterday (07-31) -> no 3-set-hangover haircut (2-set win ~neutral).
No extreme spot. Cumulative court time Pegula 170m vs Shnaider 129m is CONTEXT only.
Shnaider is 22 (recovers fast). Roughly equal.
NARRATIVE
No first-Final let-down. player_notes (undated) note Pegula historically strong this time
of year (context, no numeric edge). Both sharp -> nothing that moves a price.
MARKET CHOICE (alt_markets present)
ML: fav fails min-odds; dog fairly priced. No edge.
Handicap: Shnaider +4.5 @ 2.0 -> model cover 0.5766, eff edge ~6.4%. |4.5| = HIGH-VARIANCE
tier, requires edge >= 10% (min 1.907). 6.4% < 10% -> FAILS wide-AH bar. Pass.
Pegula -4.5 @ 1.93 -> model cover only 0.4234, negative edge.
Totals: 20.5, model over 0.508 / under 0.492 -> coin flip, no edge.
AH: B +4.5 @ 2.00
GATES
G1 edge-origin: FAIL - no numeric model-vs-market edge clears its bar (ML fails min-odds;
+4.5 cover 6.4% < 10% wide-AH bar). Nothing to stake before narrative.
G2 consensus: n/a (stake-tier only) - models split around an efficient market.
G3 fatal-risk: n/a - no candidate bet.
VERDICT: PASS. Efficient sharp-vs-sharp semi-final. Favourite's short price fails min-odds,
dog fairly priced to win, only positive-edge line (+4.5) sits below the high-variance bar.