TennisEdge's AI model finds no betting value in T. Maria vs C. McNally (Toronto on 2026-08-03) — the market is efficient, with T. Maria favoured at 1.90.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | T. Maria | C. McNally |
|---|---|---|
| Elo ratings-based, surface-weighted |
29.3% | 70.7% |
| Machine learning 29-feature gradient-boosted model |
39.5% | 60.5% |
| Serve / return point-by-point simulation |
42.8% | 57.2% |
| Blend what the bot actually prices off |
41.3% | 58.7% |
| Market implied de-vigged from the opening odds |
50.0% | 50.0% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| T. Maria | hard | 1566 | 112–109 | 221 |
| C. McNally | hard | 1736 | 85–66 | 151 |
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. 4, 2026.
C. McNally leads 2–0 over 2 meetings. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| Aug. 4, 2026 | Toronto | hard | C. McNally | 2 - 1 |
| July 13, 2025 | Newport | grass | C. McNally | 2 - 1 |
| L | C. McNally | 1 - 2 | Aug. 4, 2026 |
| W | I. Marton | 2 - 0 | Aug. 1, 2026 |
| L | R. Zarazua | 0 - 2 | July 29, 2026 |
| W | T. Gibson | 0 - 2 | July 27, 2026 |
| W | K. Volynets | 0 - 2 | July 12, 2026 |
| W | T. Maria | 1 - 2 | Aug. 4, 2026 |
| L | K. Liutova | 2 - 0 | July 31, 2026 |
| W | Y. Ma | 0 - 2 | July 30, 2026 |
| W | Xiy. Wang | 1 - 2 | July 28, 2026 |
| L | E. Rybakina | 0 - 2 | July 2, 2026 |
All Aug. 3, 2026 results & picks →
T. Maria (A, 38, #54) vs C. McNally (B, 24, #63) — WTA Toronto R1, hard
MODELS vs MARKET
ELO A 29.3% | ML A 39.5% | sim A 41.3% -> all lean McNally (B ~58.7%).
Betfair ML: A 2.36 (~43.6% comm-adj), B 1.71 (~59.7% comm-adj).
Market and all three models AGREE McNally is a modest favourite = consensus, efficient.
ELO/RANK SANITY
Ranking has Maria higher (#54 vs #63) yet ELO gives her only 29.3% — "strong rank, low ELO"
suggests ELO understates Maria. But the sim already lifts her to 41.3% and the market to ~43.6%,
i.e. both already price her well above ELO. No residual mispricing to exploit.
RECENT FORM / SHAPE (margins only)
Maria: SHARPENING but on inflated fields. Won a "Toronto" event Final 08-01 (yesterday) 6-1 6-4
over Marton (elo 1228, ITF-weak); DOM over Gibson (1500) 6-0 6-3; won Newport grass title 07-12.
Level-dropper: hard-court wins built on weak/qualifier fields -> stats inflated.
McNally: match-sharp on hard, reached Memphis QF (lost 07-31 to Liutova 1745); DOM over Ma (1586),
GRIND over Wang (1500). STABLE.
Both are match-sharp on hard; neither carries rust risk.
FATIGUE (Step 2)
Maria: 3 matches/7d, 197 min, won a 2-SET final yesterday (08-01) -> 3-set hangover haircut does
NOT apply (2-set win ~neutral). Age 38 (>=30, recovers slowly) + LETDOWN spot (won Final 1d ago)
argue mildly against her, but this is narrative, not a numeric edge.
McNally: 3/7d, 257 min, last match 07-31, age 24 (recovers fast). Roughly equal; no extreme spot.
NARRATIVE
Maria LETDOWN: won Final 08-01 (1d ago). TITLE: Toronto (hard) 1d ago; Newport (grass) 21d ago.
No surface transition (both on hard). Notes: press item confirms McNally in-form (4th QF of 2026),
no injury flag. All directional-against-Maria signals point the same way the market already leans.
EDGES @ EXCHANGE (comm-adj 0.95)
ML B (McNally) 58.7% vs BF 1.71 -> edge -1.0% (book 1.90 shows +6% but exchange contradicts -> stale, killed by Pricing Discipline #1).
ML A (Maria) 41.3% vs BF 2.36 -> edge -2.3%.
AH Maria +2.5 55.1% vs 1.95 -> edge +2.6% (below 5% bar).
AH McNally -2.5 44.9% vs 1.95 -> edge -7.7%.
Totals O/U 21.0 -> both negative.
Best available edge = +2.6% (AH Maria +2.5), under the 5% minimum.
MARKET CHOICE
No market clears its bar; nothing to stake. alt_markets present.
AH shadow: Maria +2.5 @ 1.95 (not staked, dog not at >=3.0).
GATES
G1 edge-origin: FAIL — no market reaches 5% exchange edge pre-narrative (best +2.6% AH; both ML negative).
G2 consensus: PASS (stake-tier only) — all 3 models align with market; would cap tier anyway.
G3 fatal-risk: n/a — no margin bet staked.
VERDICT: PASS. Full model+market consensus on McNally with NO usable exchange edge in any market.
Maria's SHARPENING look is on inflated qualifier/ITF fields; her letdown/age lean the same way the
market already sits, so no fade to buy either. Textbook efficient price.