TennisEdge's AI model finds no betting value in M. Inglis vs T. Preston (Vancouver on 2026-08-01) — the market is efficient, with T. Preston favoured at 1.64.
Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.
| Method | M. Inglis | T. Preston |
|---|---|---|
| Elo ratings-based, surface-weighted |
38.6% | 61.5% |
| Machine learning 29-feature gradient-boosted model |
55.9% | 44.1% |
| Serve / return point-by-point simulation |
52.8% | 47.2% |
| Blend what the bot actually prices off |
38.6% | 61.5% |
| Market implied de-vigged from the opening odds |
40.6% | 59.4% |
We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?
| Player | Surface | Elo | Record | Matches |
|---|---|---|---|---|
| M. Inglis | hard | 1671 | 154–102 | 256 |
| T. Preston | hard | 1773 | 139–62 | 201 |
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. 1, 2026.
Level at 2–2 over 4 meetings. Full record.
| Date | Event | Surface | Winner | Score |
|---|---|---|---|---|
| Aug. 1, 2026 | Vancouver | hard | T. Preston | 2 - 0 |
| Nov. 29, 2025 | W75 Playford | hard | M. Inglis | 2 - 0 |
| Oct. 6, 2023 | ITF W25 Cairns Women | hard | M. Inglis | 2 - 0 |
| Sept. 22, 2023 | ITF W25 Perth 2 Women | hard | T. Preston | 2 - 0 |
| L | T. Preston | 0 - 2 | Aug. 1, 2026 |
| W | M. Hontama | 2 - 0 | July 31, 2026 |
| W | M. Uchijima | 1 - 2 | July 30, 2026 |
| W | S. Costoulas | 0 - 2 | July 29, 2026 |
| W | G. Minnen | 0 - 2 | July 29, 2026 |
| W | M. Inglis | 0 - 2 | Aug. 1, 2026 |
| W | M. Sawangkaew | 2 - 1 | July 31, 2026 |
| W | P. Hon | 0 - 2 | July 30, 2026 |
| W | H. Dart | 0 - 2 | July 30, 2026 |
| W | H. Kaji | 0 - 2 | July 29, 2026 |
All Aug. 1, 2026 results & picks →
═══ Inglis vs Preston — Vancouver (WTA, hard), Final ═══
MODELS (prob player_a = Inglis)
ELO 38.6% (favors Preston ~61%)
ML 55.9% (favors Inglis) — OUTLIER
Sim 52.8% (Preston 47.2%), expected margin +0.5 games (coin-flip), exp total 20.9
Market Betfair: Inglis 2.32 / Preston 1.68 → Preston ~58-60% (sits with ELO)
Genuine 2-vs-2 split: ML+sim lean Inglis; ELO+market lean Preston.
ELO/RANKING SANITY
Preston ranked 126 vs Inglis 164; ELO favors Preston — ELO and ranking AGREE, ELO credible.
Sim serve inputs: Inglis hold 0.721 / Preston hold 0.538. Preston's 50-match avg is
depressed by a rough multi-surface losing stretch (Evansville, Wimbledon QF, Nottingham,
Ilkley, Birmingham, RG, Paris) — opponent-unadjusted, likely UNDER-rates her. Known sim
failure mode; ELO (opponent-adjusted + updated on the week) is more trustworthy. The
market, the sharpest single estimate, agrees with ELO.
RECENT FORM / SHAPE (both match-sharp, 4 matches deep at venue)
Inglis: SF DOM 6-3 6-4 v Hontama(1607); QF GRIND(3set TB) v Uchijima(1653);
R16 DOM 6-3 6-3 v Costoulas(1605); R32 DOM 6-1 6-3 v Minnen(1684). SHARPENING, no blowout losses this week.
Preston: SF GRIND 5-7 6-3 6-3 v Sawangkaew(1748 Strong); QF DOM 6-3 6-3 v Dart(1710 Strong);
R16 GRIND(TB) v Hon(1500); R32 DOM 6-0 6-2 v Kaji(1527). SHARPENING, marquee wins over Strong.
Both dominant this week; Preston's wins came over higher-ELO opponents. No clean qualitative
edge to resolve the model split toward the contrarian (Inglis) side.
FATIGUE (context only)
Both back_to_back 2, both played SF yesterday (07-31). Preston won a 3-SET match yesterday →
3-set-hangover haircut applies, but she is 20 (recovers fast → softened ~-2/-3pp). Inglis (28)
won a 2-set match yesterday → roughly neutral. Preston more loaded (436 vs 321 min). Slight lean
to Inglis, but fatigue is CONTEXT and cannot create the edge.
NARRATIVE
Both in the Final (big spot, max effort both sides — neutral). No let-down feeding in. Both
hard-court all week, no surface transition.
MARKET CHOICE
ML: Inglis edge only exists if sim+ML (~54%) are right; ELO+market (~42-47%) say no. Contested
contrarian edge, thin-to-negative at Betfair 2.32 (min 2.23 at 47%). No conviction.
Handicap: Inglis +2.5 @ 1.96 shows nominal +5-11% off the sim, but VETOED — see G3.
Total: sim ~50/50 at line 21, prices ~50/50. No edge.
AH: A +2.5 @ 1.96 / B -2.5 @ 1.93 (quoted).
alt_markets present (not null).
GATES
G1 edge-origin: PASS(weak) — nominal ML/sim edge on Inglis exists pre-narrative, but rests
entirely on the contested sim+ML side vs opponent-adjusted ELO+market.
G2 consensus: stake-tier — 2 models (ML+sim) vs ELO+market; capped given shared unadjusted-stat
failure mode. Not blocking.
G3 fatal-risk: FAIL (margin veto) — Inglis +2.5 cover is self-negating: Inglis's losses include
blowouts (2-6 2-6 v Zakharova, 6-2 6-3 v Naef) and Preston produced blowout WINS this week
(6-0 6-2, 6-3 6-3). Cover + two-sided blowout risk = hard veto for margin bets → PASS.
VERDICT: PASS. Coin-flip final with a genuine model split; the only juicy market (Inglis +2.5) is
G3-vetoed, and the ML "value" on Inglis is the contested side against a sharp market that agrees
with the opponent-adjusted ELO. No bettable edge with qualitative validation.