/ PROOF

We don't force picks where the data says we lack edge.

Name: NuroPicks
Availability: InStock
Author: NuroPicks LLC

Every selectivity gate below cites its sample. Every CLV figure comes from the live picks table. If a sport×market combination doesn't clear the bar, the model stays silent — that's how we keep the slate honest.

Currently paused buckets

Paused

WNBA × all markets

n=14 · hit 57.1% · ROI -22.3%

Sample too small to validate. 14 picks at -22% ROI is more likely noise than signal. Pause until 100+ pick sample lands.

Paused

NHL × PLAYERPROP

n=15 · hit 33.3% · ROI -51.3%

Worst bucket in the entire 30-day audit. The model's player-prop signal hasn't survived playoff hockey — different ice time distribution, defensive matchups, line shuffling. Pause until regular-season returns or a retrain.

Paused

NBA × PLAYERPROP

n=31 · hit 35.5% · ROI -24.5%

31-pick sample at 35.5% hit / -24.5% ROI. The current prop signals don't beat the juice on NBA. Pause; revisit when prop-pick-builder gets a redesign.

Higher edge bar (still publishing, just stricter)

+350bps edge required

MLB × PLAYERPROP

n=64 · hit 53.1% · ROI -16.1%

Workhorse with 64 picks of sample but bleeding at -16% ROI. Raise the edge bar +350bps before publishing to filter the marginal calls.

+200bps edge required

NHL × ML

n=34 · hit 61.8% · ROI 3.45%

ML is +3.45% over 34 picks but playoff variance is real. Double the regular-season edge bar through end of NHL playoffs.

+150bps edge required

MLB × ML

n=102 · hit 57.1% · ROI -4.51%

Workhorse with the largest sample (102 picks) but -0.44pp CLV and 57% hit. The model is selecting prices the market later TIGHTENS against — that's anti-edge. Lift the implied-prob bar +150bps and re-audit in 2 weeks.

+250bps edge required

MLB × SPREAD

n= · hit % · ROI %

MLB chalk spreads pay -230 to -260 typical. The 30d sport-level CLV is -0.38pp; the spread sub-bucket inherits the same bleed without the ML's positive-EV moments. +250bps demands real edge before we publish.

+200bps edge required

MLS × ML

n=5 · hit 50% · ROI -16.67%

Tiny n=5 sample with +1.55pp CLV (model picking right side) but -16.67% ROI from chalk pricing. Tightened bar lets the rule survive when MLS is re-added to FORCE_CORE_SPORTS later.

Closing-line value (CLV) vs ROI by sport · 30 days

CLV measures whether the model is selecting prices the market later tightens against. Positive = real edge. Hit rate and ROI can lie in a 100-pick sample; consistent positive CLV is the proof. We show BOTH side-by-side here because the gap between them is the actual diagnostic:

🟢 ALIGNED — ROI within ±3pp of CLV. Bottom line tracking the model.
🔴 LEAKING — CLV ahead of ROI by 3pp+. Model picking the right side, variance against us. Honest place to be; this is where edge-weighted sizing compounds.
🟡 LUCKY— ROI ahead of CLV by 3pp+. Wins above model expectation. Don't celebrate; mean reversion is coming.

WNBA

🔴 leaking

+1.42pp

CLV

-28.26%

ROI · vig-aware

n=4 · hit 50.0% · gap +29.68pp

UFC

⚪ n/a

+0.66pp

CLV

—

ROI · vig-aware

n=10

BOXING

⚪ n/a

+0.65pp

CLV

—

ROI · vig-aware

n=5

NHL

🟡 lucky

+0.34pp

CLV

+9.17%

ROI · vig-aware

n=35 · hit 65.7% · gap -8.83pp

MLB

🟢 aligned

-0.35pp

CLV

-2.77%

ROI · vig-aware

n=108 · hit 56.9% · gap +2.42pp

NBA

🔴 leaking

-0.48pp

CLV

-6.84%

ROI · vig-aware

n=42 · hit 66.7% · gap +6.36pp

MLS

🔴 leaking

-5.61pp

CLV

-16.67%

ROI · vig-aware

n=7 · hit 50.0% · gap +11.06pp

How to read this

Hit rate is the percentage of graded picks that won. Looks great in marketing copy but lies in small samples.
ROI is profit divided by amount risked. Tells you whether the past N picks made money, but variance dominates short horizons.
CLV (closing-line value)is the only metric that survives variance. If the model bought a price the market later tightened against, that's an edge — regardless of whether that specific pick won.
The gates aboveare how the model enforces discipline. When the data on a bucket says we don't have edge, we don't publish picks there. No forced bets to keep the slate looking busy.

The raw data

→ /api/model/selectivity — gate matrix JSON
→ /api/model/clv — 7d/30d/90d CLV by sport+market
→ /record — live append-only ledger of every pick

Pre-gate baseline (30d, $25 flat staking): 530 picks · 67.3% hit · -2.4% ROI. Gates reviewed monthly. Entertainment only. 21+. 1-800-GAMBLER.

Must be 21+. Gambling problem? Call 1-800-GAMBLER. NJ/PA: 1-800-GAMBLER · MI: 1-800-270-7117 · VA: 1-888-532-3500 · CO: 1-800-522-4700 · NY: 877-846-7369 (HOPENY).

NuroPicks LLC · 30 N Gould St Ste R, Sheridan, WY 82801