TENNISEDGE.GG

T. Preston vs M. Sawangkaew — Prediction & Analysis

Vancouver · Vancouver - Semi-finals · July 31, 2026 · 19:30 UTC

TennisEdge's AI model finds no betting value in T. Preston vs M. Sawangkaew (Vancouver on 2026-07-31) — the market is efficient, with M. Sawangkaew favoured at 1.46.

Surface hard Odds T. Preston 2.62 · M. Sawangkaew 1.46

Verdict: no bet

The model found no value at current prices — the market has this one priced about right. That verdict is part of the public record too.
Profiles: T. Preston · M. Sawangkaew · head-to-head record · Vancouver

Model probabilities

Independent estimates of who wins, each from a different method. The blend is what the bot prices against the market.

Method T. Preston M. Sawangkaew
Elo
ratings-based, surface-weighted
 50.0% 50.0%
Machine learning
29-feature gradient-boosted model
 41.5% 58.5%
Serve / return
point-by-point simulation
 43.2% 56.8%
Blend
what the bot actually prices off
 42.0% 58.0%
Market implied
de-vigged from the opening odds
 35.8% 64.2%

We publish how well these score against the closing market, including where they lose: how accurate are tennis predictions?

Elo ratings

PlayerSurfaceEloRecordMatches
T. Preston hard1773 139–62201
M. Sawangkaew hard1718 145–79224

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.

Head to head

T. Preston leads 1–0 over 1 meeting. Full record.

DateEventSurfaceWinnerScore
July 31, 2026VancouverhardT. Preston2 - 1

Recent form

T. Preston

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

M. Sawangkaew

L R. Sramkova 0 - 2 Aug. 1, 2026
L T. Preston 2 - 1 July 31, 2026
W E. Jones 0 - 2 July 30, 2026
W N. Hibino 1 - 2 July 29, 2026
W E. S. Liang 0 - 2 July 28, 2026

All July 31, 2026 results & picks →

Full analysis

Vancouver (WTA, hard) SF — T. Preston (A, #126, age 20) vs M. Sawangkaew (B, #175, age 24)

MODELS vs MARKET
  ELO 0.500 (A) | ML 0.415 | model 0.420 | sim 0.4325 — all four cluster A ~42-50%.
  Market: Betfair B 1.56 (~64% B / ~36% A), book B 1.46; A 2.74 exch.
  All models say A is MORE competitive than the market's ~36%. Market is the outlier
  making B a clear favourite. But ELO 0.500 is undifferentiated/likely stale, and the
  reason the market favours B is qualitatively sound (see shape).

RECENT FORM / SHAPE / PATH
  Preston (A): Vancouver R32 6-0 6-2 vs Kaji(1527) DOM; R16 tight grind vs Hon(1500);
    QF 6-3 6-3 vs Dart(1710 Strong) DOM. 3 matches this week = match-sharp, playing
    well on hard. BUT prior form poor (Evansville R32 loss, French QF loss, several
    grass losses). Shape: STABLE, sharp this week only.
  Sawangkaew (B): R32 6-2 6-4 vs Liang(1632); R16 3-set grind vs Hibino(1589);
    QF 6-3 6-2 vs Jones(1756 Strong) DOM. Plus Wimbledon TITLE 06-25 and Birmingham
    TITLE 05-31. Shape: SHARPENING, genuinely hot. Her straight-set wins are frequently
    5-7 game blowouts (6-3 6-2 vs Jones, 6-2 6-3 vs Ma/Miyazaki, 6-4 6-2 vs Kinoshita).

FATIGUE
  Both won 2-set matches yesterday (07-30) -> NO 3-set-hangover haircut on either side.
  Sawangkaew heavier cumulative load (442 min/14d, back-to-back 4 = 4th consecutive day)
  vs Preston 322 min. Cumulative minutes alone = context, not an edge; the market
  over-prices visible fatigue and her title-run form dominates. Age 24 baseline. Not
  enough to bet against her.

NARRATIVE
  Sawangkaew: TITLE Wimbledon 06-25 (grass) + Birmingham 05-31 (grass); live hard form
  strong (Gifu title, Vancouver DOM vs Jones). Live form outweighs stale ELO -> the
  market's B favouritism is justified, not soft. No surface-transition risk (both
  played 3 hard matches this week).

MARKET CHOICE
  ML A: sim 0.4325 vs eff 2.653 -> edge 5.6%, min 2.43. Clears numerically, BUT the
    qualitative read validates the MARKET (B genuinely stronger via hot form ELO misses),
    so the sim's 0.4325 likely overstates A. Thin/illusory edge -> PASS.
  Handicap A +3.5 @ 1.94: sim covers 0.6126 vs eff 1.893 -> edge 8.4%, min 1.71.
    Nominal edge, tight tier. VETOED at G3: self-negating margin risk — the very form
    that makes B the favourite (frequent 5-7 game straight-set blowouts) directly
    threatens the +3.5 cover; sim's -0.5 expected margin looks optimistic vs her
    demonstrated blowout margins. Units never buy back a negating margin risk.
    AH: A +3.5 @ 1.94 (shadow, vetoed).
  Total: sim ~0.49 over 21.0 -> edge ~0, PASS.

GATES
  G1 edge-origin: PASS (ML/cover edge exists numerically pre-narrative) — but see verdict.
  G2 consensus:   stake-tier only (never blocks).
  G3 fatal-risk:  FAIL (margin bet) — B's in-form 5-7 game straight-set blowouts
                  self-negate the A +3.5 cover thesis; HARD VETO on the only market
                  carrying edge.

VERDICT: PASS. The model-vs-market gap is real numerically, but the qualitative read
(Sawangkaew's genuine title-run form, which stale ELO and the sim underweight) validates
the MARKET's favouritism rather than the dog. The only market with nominal edge (A +3.5)
is the audited leak — a cover thesis fatally exposed to the favourite's blowout tendency.
Pass.

More from Vancouver