TennisEdge's AI model finds no betting value in A. Ilagan vs R. Bertola (Washington on 2026-07-25) — the market is efficient, with R. Bertola favoured at 1.46.
A. Ilagan (age 25, rank 332) vs R. Bertola (age 27, rank 255) — Washington, hard
MODELS vs MARKET
ELO 45.0% A | ML 47.5% A | Blended 43.2% A
Book: 2.65 / 1.46 (implied A 35.5%) — stale vs exchange
Betfair (executable): 2.44 / 1.57 -> A implied 41.0% (norm ~39.2%)
Commission-adjusted A: eff 2.37, implied 42.2% -> blended edge ~+1.0%; ML-only ~+5.2%; ELO ~+2.8%
Backing B: eff 1.54, implied 64.9% vs model 56.8% -> negative edge
ELO/RANKING SANITY
Ilagan: models near coin-flip vs a player ranked 77 spots higher, yet Ilagan is ranked 332.
History shows rating built at weak fields: Jiujiang Qualification (opp ELO 1127, 1284),
Wuxi (opp ELO 1293, 1500, 1506). Classic farmed-ELO pattern -> the model lean toward A
is discounted, not trusted.
RECENT FORM (last 5 each)
Ilagan: L Watanuki (1609) 7-6 7-5 GRIND; L C. Smith (1686) 3-set TB GRIND; L Broady (1582) TB;
W Noguchi (1587) 3 sets; W Rybakov (1550) 3 sets. Three straight losses coming in.
Wins concentrated vs Avg/Weak opponents; 0 DOM wins vs Strong/Elite in window.
Bertola: L Fearnley (1624) 6-4 6-4; L S. Johnson (1551) 6-1 6-4; L Mannarino (1773) 6-2 6-1 blowout;
L Wimbledon quali final (5 sets); W Rodionov (1667). Also cold: three straight losses,
two lopsided, but against the market-favored side that cuts both ways.
FATIGUE (precomputed)
Ilagan: 2 matches/14d, 6 sets, 266 court min, last 07-21
Bertola: 2 matches/14d, 6 sets, 166 court min, last 07-21
Diff 100 min -> Ilagan side disadvantaged; age 25 = baseline, no softening. (Context only —
cannot create an edge, and here it points AGAINST the only candidate pick.)
NARRATIVE
Let-down: none live (Ilagan's Newport final win 18d ago, outside window; Bertola Nottingham title 06-14).
Venue: no Washington history for either. Surface transition: none — both on hard last match.
Alt markets: handicap/totals null -> ML only.
GATES
G1 edge-origin: FAIL — blended model vs commission-adjusted Betfair = ~+1% on A, below the 5% bar; only the lone ML model reaches ~5%, and that model's input (Ilagan ELO) is flagged as farmed.
G2 consensus: FAIL — market and blended model both sit under the ML lean; qualitative evidence (3 straight losses, farmed rating, heavier court time) runs AGAINST Ilagan, not for him.
G3 fatal-risk: FAIL — the thesis for A requires trusting a rating built on ELO<1300 fields while the player arrives on a 3-match losing streak with a 100-min court-time deficit.
VERDICT: PASS. The book's 2.65 looks soft only against the model; the exchange at 2.44 removes
the edge, and every qualitative signal argues the models are overrating Ilagan rather than the
market underrating him. No stakeable alt markets. Predicted winner: R. Bertola.