TennisEdge's AI model finds no betting value in J. Pegula vs M. Frech (Washington on 2026-07-30) — the market is efficient, with J. Pegula favoured at 1.12.
J. Pegula (A, rank 5, age 32) vs M. Frech (B, rank 45, age 28) — Washington, hard, 1/8-finals.
MODELS
ELO: Pegula 87.9% | ML: Pegula 79.2% | Model: 87.9%
Market (Betfair): Pegula 1.14 (~87.7%) / Frech 6.8. Book 1.12 / 6.2.
ELO agrees with market (~88% = ~88%). ML is the lone outlier at 79% -- classic
heavy-favourite compression toward 50%, an ML failure mode, not an edge.
RECENT FORM (last 5)
Pegula: Wimbledon QF L to Gauff 6-4 3-6 3-6; W Jovic(1922) 4-6 6-3 6-1 (win over elite);
W Bouzas 6-1 6-3; W Sorribes 7-6 6-1; W Vidmanova 7-5 6-3. Strong grass run (also
Berlin final, beat Sabalenka). Elite ballast. Returning to hard = her best surface.
Frech: Washington 1/16 W A.Li(1728) 6-4 6-3 (yesterday, match-sharp on THIS surface);
Wimbledon R1 L Kalinskaya 6-7 4-6; Bad Homburg L Osaka 6-4 6-1; Berlin L Lys; Berlin
final L Zhang 2-6 4-6. Poor grass stretch, mediocre clay. Weak underlying shape.
SHAPE & PATH
Pegula: STABLE/positive (QF Wimbledon, wins over Jovic/Sabalenka). Path: FIRST match at
Washington, 22 days since last match (>21d) = rust risk. BUT grass->hard is a
transition TO her primary/home surface, so rust real but surface favourable.
Frech: FADING (string of grass losses) but match-sharp this week -- played-in yesterday.
Sharpness (+) offset by genuinely poor form (-).
FATIGUE
Frech played yesterday: a 2-set win (6-4 6-3, 72min) => roughly neutral (~-2pp,
round-progression the models already price). Age 28 baseline. No extreme spot, no
meaningful haircut. Pegula fully rested. Fatigue is not a betting factor here.
NARRATIVE
No first-Final let-down. Surface transition: Pegula grass->hard (her best surface, US
home swing -- not a foreign-surface penalty). Frech match-sharp on hard.
ELO/ranking align (elite fav, credible ELO).
GATES
G1 edge-origin: FAIL -- only Frech shows a numeric edge (+5.4% via ML), and it exists
ONLY because ML compresses a heavy favourite; ELO agrees with the market at ~88%.
No articulable, market-slow information supports the ML outlier. Backing Pegula is
below min_odds (1.05/0.879 = 1.194 vs Betfair 1.14).
G2 consensus: lone-model (ML only) support; ELO+market against.
G3 fatal-risk: n/a (no ML bet advanced).
VERDICT: PASS. ELO + exchange consensus at ~88% Pegula; the lone ML disagreement is
heavy-favourite compression, not an edge. Frech is match-sharp but in genuinely poor
form, and Pegula returns to her strongest surface at a home hardcourt event. Pegula
priced below min-odds to back. No professional edge either side.