R. Zarazua vs T. Korpatsch — Prediction & Analysis

TennisEdge's AI model finds no betting value in R. Zarazua vs T. Korpatsch (Toronto on 2026-08-03) — the market is efficient, with R. Zarazua favoured at 1.80.

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

R. Zarazua leads 2–0 over 2 meetings. Full record.

Recent form

All Aug. 3, 2026 results & picks →

Full analysis

═══ Zarazua (A) vs Korpatsch (B) — WTA Toronto R1, hard ═══

MODELS vs MARKET
  ELO A 59.7% | ML A 65.7% | blend A 60.5% | Pinnacle no-vig A 59.3%
  Betfair ML: A 1.66 (eff 1.627 -> 61.5%) | B 2.40
  Book: A 1.80 | B 2.02. value_pick A (+7.6% vs soft book, NOT vs exchange).
  ELO + sharp Pinnacle both sit AT/below the Betfair 61.5% — consensus with the
  exchange. Only the inflation-prone ML model diverges upward.

RECENT FORM (last 5)
  A Zarazua: Memphis run on HARD, match-sharp. L SF vs Vidmanova(1788) 6-3 6-3;
    W QF vs Zakharova(1787) 6-4 6-0 (DOM over Strong); W vs Maria(1500) 6-4 6-4;
    W vs Kessler(1500) 6-4 6-1 (DOM); L Evansville SF vs Dolehide. SHARPENING on
    surface but heavily played.
  B Korpatsch: just WON Hamburg CLAY title 07-26 — W Final Bondar(1819) 6-3 6-3
    (DOM Strong), W SF Sherif(1855, ret), W QF Kalinina(1837) 3-set GRIND,
    W Stusek 3-set, W JimenezK 3-set. Strong CLAY form, many grinds.

PATH / SHAPE / SURFACE
  A: 4th match-day cluster, same surface (hard) as this — match-sharp, played-in.
  B: TITLE Hamburg 7d ago (clay) -> SURFACE TRANSITION clay->hard = level
    depression + wider margin variance; WEAK let-down (title 5-7d). Her recent
    wins are all clay; hard-court form unknown.

FATIGUE (Step 2 calibrated)
  A: matches_14d 7, matches_7d 4, back_to_back 3, 511 court-min — heavy cumulative
    load on the MODEL-FAVORED side. But last match (08-01 SF) was a straight-sets
    LOSS, not a 150min+ 3-set win -> no 3-set-hangover haircut. Cumulative load is
    CONTEXT only, caps any A lean; A age 28 baseline. Net: no fatigue EDGE either
    way, but removes any reason to press A beyond market.
  B: last match 07-26, rested but rusty + off-surface.

NARRATIVE
  Surface transition on B (dog) could support an A ML thesis — BUT ELO (which
  carries surface) already prices A at 59.7% == market. The transition is
  apparently already in the price.

GATES
  G1 edge-origin: FAIL — at executable Betfair 1.66 (eff 1.627 -> 61.5%) the
    numeric edge is negative on ELO/blend/Pinnacle; only ml (+4.3%) beats it,
    below the 5% bar. min_odds 1.05/0.605=1.74 > 1.66 -> auto-PASS on price alone.
  G2 consensus: stake-tier — moot (no edge to stake).
  G3 fatal-risk: n/a for ML.

MARKET CHOICE
  ML A: eff 1.627, edge negative-to-+4.3% (< 5% bar), price below min_odds -> PASS.
  Handicap A -2.5 @1.88: cover 55.7% vs eff-implied 54.5% = +1.2%; high-variance
    tier (lay -2.5 needs ≥10%) -> PASS.
  Total 20.5: over 51.5% vs 53.1% implied (neg); under 48.5% vs 51.3% (neg) -> PASS.
  AH: A -2.5 @ 1.88 (favourite lay, not a dog ML — shadow n/a).

VERDICT: PASS. Sharp market + ELO consensus with the exchange at ~60% for A; only
the inflation-prone ML diverges; the surface-transition angle is already priced;
and the model-favored side (A) is the fatigued one, so no qualitative edge to press.
No market clears its edge/min-odds bar.