TennisEdge's AI model finds no betting value in F. Cerundolo vs A. Gea (Los Cabos on 2026-07-31) — the market is efficient, with F. Cerundolo favoured at 1.46.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | F. Cerundolo | A. Gea |
|---|---|---|
| Elo ratings-based, surface-weighted |
54.6% | 45.4% |
| Machine learning 29-feature gradient-boosted model |
60.6% | 39.4% |
| Serve / return point-by-point simulation |
28.5% | 71.5% |
| Blend what the bot actually prices off |
52.8% | 47.2% |
| Market implied de-vigged from the opening odds |
64.9% | 35.1% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| F. Cerundolo | clay | 1861 | 125–69 | 194 |
| A. Gea | clay | 1586 | 72–67 | 139 |
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 May 30, 2026.
A. Gea leads 1–0 over 1 meeting. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| July 31, 2026 | Los Cabos | hard | A. Gea | 2 - 1 |
| L | A. Gea | 1 - 2 | July 31, 2026 |
| W | T. Boyer | 0 - 1 | July 30, 2026 |
| L | L. Miguel | 3 - 2 | July 18, 2026 |
| W | C. Norrie | 3 - 0 | July 18, 2026 |
| L | B. Nakashima | 1 - 3 | July 17, 2026 |
| W | D. Shapovalov | 2 - 0 | Aug. 2, 2026 |
| W | C. Wong | 0 - 2 | Aug. 1, 2026 |
| W | F. Cerundolo | 1 - 2 | July 31, 2026 |
| W | M. Zheng | 2 - 1 | July 30, 2026 |
| W | S. Kwon | 1 - 2 | July 29, 2026 |
All July 31, 2026 results & picks →
=== F. Cerundolo (A) vs A. Gea (B) - Los Cabos, hard, QF ===
MODELS (prob A / Cerundolo)
ELO 0.546 | ML 0.606 | ensemble 0.528
Market: book 1.46 = 68.5 pct | Betfair 1.54 = 64.9 pct (A); Betfair 2.66 = 37.6 pct (B)
value_pick B, edge 0.121 - models rate the dog above market.
RECENT FORM (last 5 shown; all fetched used for classification)
Cerundolo (A): 07-30 W v Boyer 1647 (set1 TB, 4-3 retirement - short, thin reps); 07-18 UTS exhibition mixed; 06-30 LOST Wimbledon v Munar 1586 6-1 6-4 6-3 (poor); 06-21 WON London title grass v Paul 1852; 06-20 W v Nakashima 1812.
Read: clay/grass player, thin/questionable HARD form, off a retirement win. TRANSITION off grass/exhibition onto real hard-court tour tennis.
Gea (B): 07-30 W v Zheng 1796 (0-6 6-4 6-0, volatile); 07-29 W v Kwon 1785 (3-set, long TB); 07-19 LOST Granby final. Strong week but GRINDY/volatile wins.
Path: 2 matches deep at THIS venue, match-sharp, SHARPENING vs Strong bracket.
FATIGUE
Gea: back_to_back 3, won 3-set matches 07-29 AND 07-30 -> 3-set hangover haircut -4.5pp applies to HIM. Age 21 (<=22) softens one tier, but load is real and points TOWARD Cerundolo.
Cerundolo: today's win was a retirement (short); fresher than Gea for the QF.
=> Fatigue favors the FAVOURITE, opposing any dog (B) bet.
ELO/RANKING SANITY
Gea rank 241 but models rate him 40-47 pct => ELO likely FARMED at challenger events (Granby/Newport/Ilkley). Discount the B edge. Live wins over Zheng/Kwon partly offset but were volatile grinds, not clean form.
MARKET DISAGREEMENT
B apparent edge: Betfair 2.66 -> eff 2.577 -> implied 0.388. ML gives B 0.394 = NO edge; only ELO/ensemble (0.45-0.47) show a gap, and that ELO is the contaminated one.
ALT MARKETS
Totals: model line 21.5, over 0.489 / under 0.511 vs market 22.0 (over 1.99 / under 1.96) - both sides negative edge, no play.
Handicap: A -3.5 @ 2.08 / B +3.5 @ 1.83, model handicap null - no margin read.
AH shadow (dog B): B +3.5 @ 1.83 - pass (no clean cover edge; dog carries fatigue).
GATES
G1 edge-origin: FAIL - strip fatigue, B edge ~5 pct marginal AND ML (0.394) matches market (0.388); the gap rests solely on farmed/inflated ELO. No clean numeric edge.
G2 consensus: stake-tier - lone contaminated signal (farmed ELO, off-surface fav), would cap 0.5-1.0u even if taken.
G3 fatal-risk: N/A (ML; passing). Self-negating - cannot back a dog whose fatigue I mark against him.
VERDICT: PASS. ML agrees with the market on the dog price; the only disagreement is a farmed-ELO artifact, and Gea's back-to-back 3-set load pushes the true number toward the favourite. No stakeable edge in any market.