V. Jimenez Kasintseva vs A. Parks — Prediction & Analysis

TennisEdge's AI model finds no betting value in V. Jimenez Kasintseva vs A. Parks (Toronto on 2026-08-03) — the market is efficient, with A. Parks favoured at 1.62.

Model probabilities

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

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

Elo ratings

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 July 21, 2026.

Head to head

A. Parks leads 3–0 over 3 meetings. Full record.

Recent form

All Aug. 3, 2026 results & picks →

Full analysis

V. Jimenez Kasintseva (A) vs A. Parks (B) — WTA Toronto R1 (1/64), hard. RE-CHECK of prior LEAN.

MODELS vs MARKET
  ELO_a 0.430 | ML_a 0.528 | blend 0.509 | Pinnacle no-vig_a 0.426 | book implied_a 0.413
  Betfair ML: A 2.28 / B 1.69. Executable A price prob = 1/(1+1.28*0.95) = 0.451.
  Edge on A: blend +5.8%, ML-only +7.7% — BUT ELO(0.430), Pinnacle(0.426) and book(0.413)
  ALL cluster ~43%. ML is the lone dissenter pushing A to a favorite. ELO agrees with the
  market AND with ranking direction (B rank 79 > A rank 111). Lone-ML split, ELO credible.

RECENT FORM (last 5)
  A (JK, 20, rank 111): 08-01 W Toronto QUALIFYING Final vs Bassols(1575) 6-0 2-6 6-3 GRIND
    (won ugly vs Avg = under-performing); 07-21 L Hamburg clay vs Korpatsch(1700);
    07-13 L Iasi 6-0 6-1 vs Charaeva(1692) blowout loss; 06-29 L Wimb 6-2 6-3;
    06-13 L Modena SF. Correction applied: 08-01 was a QUALI final, not a main-draw title —
    her main-draw form is poor. Match-sharp on hard (qualified) but level is quali-grade.
  B (Parks, 25, rank 79): 07-20 L Prague clay 2-6 4-6 vs Havlickova(1523) bad loss;
    07-17 L Athens quali vs Sakkari(1819); 07-15 W Athens quali 7-6 2-6 5-7 GRIND;
    07-13 W Athens quali; 07-01 L Wimb 5-7 0-6. FADING, rusty (13 days off), poor form.
  Shape: both FADING. A match-sharp on hard, B rusty+out-of-form. Neither is convincing.

FATIGUE (Step 2 calibrated)
  A won a 3-SET match YESTERDAY (08-01, 191min/14d) -> 3-set-hangover haircut -4.5pp,
  softened one tier for age 20 (recovers fast) -> ~-2.5pp on A.
  B last match 07-20 (13d) = mild rust, not extreme. Net: cumulative gaps are context only.

NARRATIVE
  A: prior "TITLE/venue momentum" signal DEMOTED — 08-01 was qualifying, not a title.
  A transition risk: last matches clay -> now hard (rust/looser margins); but she has been
  playing the hard-court quali so partly conditions-adjusted. B: no title/letdown signal.

MARKET CHOICE
  ML: A executable 2.28. Blend edge 5.8% rests ENTIRELY on the lone ML outlier that ELO,
    Pinnacle and book all contradict; discount ML -> A true prob ~46-48%, and the -2.5pp
    hangover pulls it to ~48% -> edge ~3%, below the 5% bar. No stakeable ML edge.
  Totals: model under 0.5434 vs under 2.00 (exec 1.95, 0.513) = +3.1%, below bar. Over none.
  Handicap: model handicap NULL, no margin read -> cannot build a cover thesis. Pass.
  AH: A +2.5 @ 1.90 (not staked; no margin model).

GATES
  G1 edge-origin: FAIL — pre-fatigue 5.8% blend edge is lone-ML only; ELO+Pinnacle+book all
     ~43% contradict it. Not a robust model-vs-market disagreement; edge is manufactured by
     one dissenting model. Erodes below bar after age-softened hangover.
  G2 consensus:   stake-tier only — lone_signal (ML only), would cap 0.5-1u even if bet.
  G3 fatal-risk:  n/a for ML.

VERDICT: PASS. Prior LEAN downgraded. Correction (quali final, not title) weakens A's form
  thesis; the only pro-A edge is a lone ML model that ELO, Pinnacle and the book all reject,
  and the thin edge is eroded by A's 3-set-hangover. Market ~43% on A is a fair, efficient
  price. No stakeable edge in any market.