A. Michelsen vs J-L. Struff — Prediction & Analysis

TennisEdge's AI model finds no betting value in A. Michelsen vs J-L. Struff (Montreal on 2026-08-03) — the market is efficient, with A. Michelsen favoured at 1.33.

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. 4, 2026.

Head to head

A. Michelsen leads 2–0 over 2 meetings. Full record.

Recent form

All Aug. 3, 2026 results & picks →

Full analysis

A. Michelsen (A, 21, #42) vs J-L. Struff (B, 36, #83) — Montreal R1, hard.

MODELS vs MARKET
  ELO A 77.6% | ML A 69.5% | model A 77.6% | Pinnacle no-vig 71.9%
  Betfair ML A 1.37 (eff-imp 74.0%) / B 3.45 (eff-imp 30.1%). Book 1.33/3.40.
  All estimates cluster 69.5–77.6% around a 74% market. Tight consensus.

RECENT FORM (last 5)
  Michelsen: 07-31 L Fritz(1941,Elite) 6-1 3-6 7-6 (competitive loss, positive shape);
    07-29 W Mannarino(1718) 6-4 7-6 GRIND; 07-28 W McDonald(1695) 6-4 7-5 GRIND;
    07-26 W Fearnley(1702) Final 7-6 6-3 GRIND (Bloomfield Hills title);
    07-25 W Suresh(1666) 6-2 6-7 6-2 GRIND. All HARD. Match-sharp, SHARPENING.
  Struff: 07-22 L Navone(1821) 0-6 3-6 CLAY blowout; 07-21 W Shevchenko 3-6 7-6 6-4;
    then a strong grass Wimbledon QF a month ago (beat Medvedev/Hurkacz). Last actual
    match a clay beatdown 11 days ago; comes to hard cold. FADING recent, off-surface.

PATH / SHAPE
  Michelsen deep hard-court run (Challenger title + Washington ATP QF, pushed Fritz),
    conditions grooved on hard — the strong qualitative side. No surface transition.
  Struff: 11 days idle since a clay loss = rust + clay->hard TRANSITION; grass-pumped
    ELO likely overstates him on hard. But well-rested, so no fatigue edge either way.

FATIGUE (Step 2)
  Michelsen heavy load: 8 matches/14d, 914min, back_to_back 5 — but age 21 (recovers
    fast), 2 days rest, and his last match was a LOSS, so NO 3-set-hangover haircut
    (haircut is winners-yesterday only). Cumulative minutes alone = context, not edge.
  Struff well-rested (11d). No extreme spot on either side.

NARRATIVE
  Let-down: Michelsen won Bloomfield Hills Final 07-26 (7d ago) = WEAK, not live.
  Transition: Struff clay->hard (risk to him). Michelsen none.

GATES
  G1 edge-origin: FAIL — best ML edge is ELO 77.6% vs Betfair eff-imp 74.0% = +3.6%,
    below the 5% bar; ML model 69.5% is BELOW market (negative). No numeric edge exists
    pre-narrative. The Michelsen story is real but already fully priced at 74%.
  G2 consensus:  n/a (stake tier only) — full 3-model + market + Pinnacle agreement.
  G3 fatal-risk: n/a — no bet.

MARKET CHOICE
  ML A: edge +3.6% (ELO high estimate only) / negative on ML model — below 5% bar. PASS.
  HC A -3.5 @1.86: model cover 0.5466 vs eff-imp 0.550 = -0.4% (and high-variance lay). PASS.
  HC B +3.5 @1.88: -9.1%. PASS.
  Totals: Over22 @1.84 -6.8%; Under22 @1.88 -3.3%. PASS.
  AH: B +3.5 @ 1.88 (passed, negative edge).
  All priced markets fail the bar. No executable edge.

VERDICT: PASS. Qualitative read favours Michelsen (match-sharp on hard vs rusty,
  transitioning 36yo) but the market already prices him at 74% and no model clears a 5%
  executable edge on any market. Efficient. No bet.