TennisEdge's AI model finds no betting value in L. Fernandez vs M. Linette (Washington on 2026-07-27) — the market is efficient, with L. Fernandez favoured at 1.41.
═══ L. Fernandez vs M. Linette — Washington, hard, 1/16-finals ═══
MODELS vs MARKET (player A = Fernandez)
ELO A 0.542 | ML A 0.605 | ensemble A 0.563
Betfair ML: A 1.58 (impl 0.633) / B 2.70 (impl 0.370); book 1.41 / 2.90
Value tag: B, edge +0.11 (book-based — stale; exchange is truth)
Rank: Fernandez 22, Linette 55. All of ELO/ML/ensemble/market favor FERNANDEZ.
SIMULATION — LONE OUTLIER, INPUTS SUSPECT
Baseline sim makes LINETTE favorite 0.595 (Fernandez 0.405) — contradicts
ELO/ML/ensemble/market, all of which favor Fernandez. Suspect input:
Fernandez spw 0.57 over 30 cross-surface matches built during a poor
clay/grass slump on her WEAKER surfaces; rank 22 and hard-court pedigree
(US Open finalist) say she is much stronger on hard than that stat implies.
ELO/RANK SANITY: Fernandez low ELO vs rank 22 = ELO farmed-down by
off-surface losses → ELO UNDERSTATES her; do not back Linette on that gap.
Bounded correction (input: Fernandez spw, baseline 0.57):
+0.03 -> spw 0.60: Fernandez 0.562 (lands on ensemble), margin +1.0
+0.05 -> spw 0.62: Fernandez 0.662 (lands on market), margin +2.1
Corrected toward consensus, the Linette-favorite read evaporates.
RECENT FORM (all matches used; most-recent 5 shown)
Fernandez (3-9 window, FADING; mostly clay/grass = weak surfaces):
06-29 Wimbledon(gr) L J.Tjen 1464/Weak 1-6 6-7 — bad loss to weak opp
06-23 BadHomburg(gr) L Xin.Wang 1800/Strong 6-3 6-4
06-21 BadHomburg(gr) W K.Boulter 1765/Strong 7-6 3-6 3-6 GRIND, over Strong
06-16 Nottingham(gr) L Z.Sonmez 1709/Strong 6-4 7-6
06-09 London(gr) L K.Boulter 1731/Strong 3-6 7-6 7-5
Linette (mixed; grass peak faded to two straight losses):
07-13 Athens-Q L M.Hontama 1500/Avg 4-6 5-7 — lost in QUALIFYING to weak opp
06-29 Wimbledon(gr) L M.Andreeva 1644/Avg 5-7 4-6
06-26 Boodles(exhib) W Chwalinska 3-6 6-3 6-10 — exhibition, discounted
06-13 Hertogenbosch(gr) L B.Krejcikova 1887/Strong 6-3 7-6 SF, close
06-12 Hertogenbosch(gr) W Z.Sonmez 1704/Strong QF
SHAPE / PATH / SHARPNESS
Fernandez: FADING (1-6 6-7 to a 1464 the last match), BUT slump is on
clay/grass — her weak surfaces; now on hard, her best surface. 28-day
layoff since Wimbledon = RUST RISK; grass->hard TRANSITION.
Linette: FADING (two straight losses incl. a qualifying loss to a 1500);
competitive grinder (3-setters, tiebreaks vs Swiatek/Krejcikova). 14-day gap.
PATH: neither has a Washington match — both first-round, both rusty, neither
match-sharp. No edge either direction on path.
FATIGUE
Both well-rested / rusty rather than tired. Fernandez matches_14d 0 (28d gap);
Linette matches_14d 1 (last 07-13, 83min). No extreme spot either side —
fatigue is a non-factor.
NARRATIVE
No Finals in either window -> no let-down. No Washington venue history.
No title recency. Surface: both grass->hard TRANSITION (widens margins,
cuts both ways — no support for a tight cover).
GATES
G1 edge-origin: FAIL — the only numeric edge (Linette ML / +2.5) exists ONLY
in the uncorrected outlier sim; corrected toward ensemble/market it falls
to +4% then negative. No durable model-vs-market edge before adjustments.
G2 consensus: FAIL — a Linette pick sits against BOTH market and ML (both
favor Fernandez); support is a lone sim/ELO split; no H2H, no live-event
evidence. ELO itself understates Fernandez (rank 22 vs 55).
G3 fatal-risk: FAIL — fading Fernandez ignores her hard-court class; if she
plays her true hard-court level (market/ML/rank thesis) she wins clearly,
negating both a Linette ML and a +2.5 cover. Units cannot buy this back.
MARKET-BY-MARKET (edge vs commission-adjusted Betfair)
ML Fernandez: market-impl 0.645; model 0.542–0.662 -> +1.5% best case, mostly
negative. No value (market >= model).
ML Linette: market-impl 0.382; corrected model 0.34–0.44 -> no edge once
sim corrected; G2 fail regardless.
AH Fernandez -2.5 @1.83: cover 0.47–0.57 -> edge -9% to +1% (high-variance
tier, needs +10%). No.
AH Linette +2.5 @2.10: cover 0.53 (moderate corr) to 0.43 (market corr) ->
edge +4% to -6%, straddles zero, below 5% bar and rests on the suspect
sim direction; G3 self-contradiction (can't fade F's win AND bank F not
covering). No.
Totals Under 22.0 @1.97: ~0.524 vs 0.520 -> +0.4%. Coin-flip. No.
VERDICT: PASS. The lone signal (raw-stat sim favoring Linette) is an outlier
that converges to the market/ensemble once Fernandez's off-surface serve stat
is corrected. Market is efficient here; no market clears its bar. Predicted
winner: Fernandez.