A. Kalinina vs A. Bondar — Prediction & Analysis

TennisEdge's AI model finds no betting value in A. Kalinina vs A. Bondar (Toronto on 2026-08-03) — the market is efficient, with A. Kalinina favoured at 1.55.

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 Aug. 3, 2026.

Head to head

A. Kalinina leads 3–2 over 5 meetings. Full record.

Recent form

All Aug. 3, 2026 results & picks →

Full analysis

═══ Kalinina (A) vs Bondar (B) — WTA Toronto R1, hard ═══

MODELS vs MARKET
  ELO A 59.3% | Sim A 59.1% | ML A 50.2% (outlier, break-even)
  Betfair A 1.56 (64.1%) / B 2.72 | Pinnacle no-vig A 63.75% (sharpest)
  Book value_pick B, edge only +2.1%.
  Read: ELO/sim ~59% and the SHARP market (Pinnacle 63.75%) BOTH favour
  Kalinina; the market is even MORE bullish on her than ELO. The only thing
  favouring Bondar is the lone break-even ML model at 50/50.

ELO/RANKING SANITY
  Kalinina ELO ~1865 but ranked 93 (high-ELO/modest-rank → clay-farmed
  rating suspicion). BUT the discount would support Bondar only if the market
  leaned on that ELO — instead Pinnacle prices Kalinina ABOVE ELO, so the
  sharp read independently likes her. Farmed-ELO story is neutralised.

RECENT FORM (last 5)
  Kalinina (clay): 07-24 L Korpatsch(1715) 6-3 1-6 6-1; 07-22 W Carle(1561)
  DOM; 07-20 W Quevedo(1739) 7-5 6-4 GRIND; 07-14 L Badosa(1792) 6-3 6-1;
  Hamburg QF. STABLE clay form.
  Bondar (clay): 07-26 L Hamburg FINAL Korpatsch(1755) 3-6 3-6; 07-25 W
  Avanesyan(1736) 3-set GRIND; 07-24 W Charaeva(1731); 07-22 W Noha Akugue;
  07-20 W Uchijima(1650). Ran to a FINAL — SHARPENING but 5 matches deep.

PATH / TRANSITION
  BOTH players transitioning clay->hard (all recent matches on clay). The
  transition roughly cancels between them; widens margin variance for both.
  H2H: Kalinina beat Bondar 6-4 7-5 (Rabat 05-21, clay) — favours A but off
  surface.

FATIGUE
  Both last played 07-24 / 07-26; today 08-02 → 7-9 days rest each. No extreme
  spot, no 3-set-hangover (a week old). Bondar carried more clay load (final
  run) but has recovered. Roughly equal for betting purposes. Both age 29.

NARRATIVE
  Bondar lost Hamburg Final 07-26 — 7d ago, outside the 0-3d NOTE window, so no
  live let-down effect.

GATES
  G1 edge-origin: FAIL — best numeric ML edge on Bondar is +2.9% (ELO/sim
    ~41%), below the 5% bar. Reaches 5% only by adopting the lone dissenting
    ML model against the sharpest market. No pre-narrative model-market edge.
  G2 consensus:   stake-tier only (would cap lone_signal) — not reached.
  G3 fatal-risk:  n/a (ML), no margin bet.

MARKET CHOICE
  ML B @2.72 -> edge +2.9% (ELO/sim), below 5% bar. PASS.
  ML A @1.56 -> edge -6.2%, negative. PASS.
  Handicap A -3.5 @2.02 (cover 51.0%) -> edge +0.25%. PASS.
  Total Under 21.5 @1.91 (53.3%) -> edge -0.3%. PASS.
  AH: B +3.5 @1.89 (not staked; dog not at odds >=3.0, shadow n/a)
  No market clears its bar. Betting the dog here means siding with the outlier
  ML model against a sharp market that is more bullish than ELO — that is
  noise, not an articulable edge.

VERDICT: PASS. No 5%+ edge at executable prices on any market; sharp market
and two of three models agree on Kalinina, ML is a lone break-even dissent.