nhl | Total | 2026-04-30
Dallas Stars @ Minnesota Wild - Over 5.5@ -118
Over heuristic (scoring inflation in pro leagues)
/ PUBLIC RECORD / LIVE FROM DATABASE
Every pick posted. Every pick tracked.
Append-only ledger. Rows are locked at post time by a Postgres trigger, so retroactive edits are impossible. ROI, yield, hit rate, and CLV are computed live at request time across 7d, 30d, 90d, and all-time windows.
Third-party reviewers: pipe export.json or export.csv into your own CLV/ROI computation. Schema at /docs/RECORD_EXPORT.
Live schedulerNext series-tipoff-tick fires at 8:51 PM ET(in 1 min)
Showing only NHL picks. AllTime ROI / hit rate / units below recompute against this slice.
/ ALL-TIME
Every settled pick. All sports. All markets.
SAMPLE
26
Settled picks
WIN %
46.1%
Hit rate
ROI
-18.66%
ROI
EX-VIG
-28.23%
ROI ex-vig (CLV-derived)
YIELD
-18.66%
Yield (flat)
UNITS
-4.85u
Units won
CLV
-0.50%
CLV avg
Small sample (26 settled). Not statistically meaningful. Track, do not tail.
/ TRANSPARENT STUB MODE
100% of recent picks (last 90 days) are sharp-stub-v0 heuristics, not XGBoost model output. Each stub pick is honestly labeled in its why-row and model_version column (178 of 178). The trust surfaces below show real published-pick outcomes, but a +CLV reading on heuristic picks proves nothing about a model that doesn't exist yet. Methodology stage 3 (XGBoost ensemble) is still pre-train; when it ships, the same surfaces flip to real model picks and this banner disappears.
How to read the ledger below
Three trust surfaces stacked, each answering a different question. Numbers come live from the picks table; rows are locked at post time by a Postgres trigger.
Daily P&L heatmap
One cell per day, colored by net units that day. Green = profit, red = loss, gray = no picks. Shows publishing cadence and day-by-day variance at a glance.
CLV trend
Closing-line value tracked over time. CLV is the gap between the price we posted and where the line closed. Positive = we beat the close. The leading indicator of long-run edge.
Odds distribution
Every settled pick bucketed by the odds we shipped. Lets a viewer see if a high hit rate is from grinding heavy favorites or from genuine +EV pricing across the board.
Daily P&L · last 90 days
178 picks across 6 active days · -4.02u net · 4 green · 2 red
2026-01-31 → 2026-04-30 (ET)
Bankroll simulator · last 90 days
Flat stake on every settled pick over the period. No martingale, no pyramiding, no cherry-picking. 153 wins+losses across the window. Past results are not indicative of future returns. Entertainment only.
$5 per pick
-$20.10
Wagered: $765Returned: $745ROI: -2.6%
$10 per pick
-$40.20
Wagered: $1,530Returned: $1,490ROI: -2.6%
$25 per pick
-$100
Wagered: $3,825Returned: $3,725ROI: -2.6%
$50 per pick
-$201
Wagered: $7,650Returned: $7,449ROI: -2.6%
CLV trend · 14-day moving avg
Closing-line value vs market across the last 60 ET days. Positive = we beat the closing number; negative = market closed tighter than where we posted.
-0.65%
▼ -0.65% (14d)
Odds distribution · 166 settled picks
Pikkit-style anti-cherry-pick view. Hit rate alone hides whether a capper grinds -300 favorites for cheap wins. The split below shows where picks actually live + whether each tier is profitable.
/ FACT-CHECK
Verify yesterday's results yourself, pick by pick
Pick a date below. We render every pick we fired for that ET day with a one-click ESPN box-score link per game so you can confirm the bot's W/L/Push grade against ESPN's authoritative scoreboard - the same source our espn-grader service uses to settle picks.
/ RECENT EDGES
Latest picks, with the WHY the model used
PENDING
nhl | Spread | 2026-04-30
Dallas Stars @ Minnesota Wild - Minnesota Wild -1.5@ +210
Home team at posted spread (home-court bias heuristic)
nhl | Moneyline | 2026-04-30
Dallas Stars @ Minnesota Wild - Minnesota Wild ML@ -122
Pre-tip moneyline favorite captured at fire time
nhl | Total | 2026-04-30
Utah Mammoth @ Vegas Golden Knights - Over 5.5@ -132
Over heuristic (scoring inflation in pro leagues)
nhl | Spread | 2026-04-30
Utah Mammoth @ Vegas Golden Knights - Vegas Golden Knights -1.5@ +140
Home team at posted spread (home-court bias heuristic)
/ WINDOWS
Rolling lookbacks
7D | 2026-04-24 -> 2026-05-01
Picks178
Settled166
Hit rate54.3%
ROI-2.63%
Yield-2.42%
CLV-0.67%
30D | 2026-04-01 -> 2026-05-01
Picks178
Settled166
Hit rate54.3%
ROI-2.63%
Yield-2.42%
CLV-0.67%
90D | 2026-01-31 -> 2026-05-01
Picks178
Settled166
Hit rate54.3%
ROI-2.63%
Yield-2.42%
CLV-0.67%
ALL | genesis -> 2026-05-01
Picks178
Settled166
Hit rate54.3%
ROI-2.63%
Yield-2.42%
CLV-0.67%
/ BY SPORT
Per-sport slice (all-time)
mlb
Picks108
W/L/P/V51/44/2/8
Hit rate53.7%
ROI+0.02%
Yield+0.02%
CLV-0.39%
Modest sample (105 settled). Directional signal, not proof of edge.
nba
Picks41
W/L/P/V20/12/0/3
Hit rate62.5%
ROI+2.53%
Yield+2.31%
CLV-1.44%
Small sample (35 settled). Not statistically meaningful. Track, do not tail.
nhl
Picks29
W/L/P/V12/14/0/0
Hit rate46.1%
ROI-18.66%
Yield-18.66%
CLV-0.50%
Small sample (26 settled). Not statistically meaningful. Track, do not tail.
/ WEEKLY ARCHIVE
Sunday 11:59 PM ET snapshots (append-only)
2026-04-20 -> 2026-04-26
PICKS
17
HIT
72.7%
ROI
+20.97%
YIELD
+15.38%
CLV
-0.98%
2026-04-13 -> 2026-04-19
PICKS
0
HIT
0.0%
ROI
0.00%
YIELD
0.00%
CLV
-
/ HONESTY RULES
Why these numbers are auditable, not flattering
RULE 01
Every posted pick lands here
If it ran through the publisher and got a Discord message, it is on this page. No private picks, no back-channel calls excluded, no deletions when a pick busts. Voids count as zero, not as removals.
RULE 02
Rows are locked at post time
A Postgres trigger sets locked_at on insert and rejects any UPDATE or DELETE that touches a locked column. Only grading fields (result, graded_at, closing_odds) may change, and those changes are logged.
RULE 03
ROI and yield shown next to hit rate
Hit rate alone misleads. A 60% hit rate on -200 favorites loses money; 53% at +100 wins. We always show ROI (unit P/L over risked) and yield (flat-unit P/L over settled picks) beside it.
RULE 04
Vig-adjusted ROI sits beside raw ROI
Raw ROI flatters anyone who only beat the juice. We also publish ROI ex-vig: raw ROI minus the average sportsbook hold per pick (computed live from the closing line of every pick that has one; falls back to the standard -110 hold of 4.55% when closing data is sparse). If ex-vig is still positive across a real sample, the edge survives the books. Leans.ai charges $299/mo to surface this; ours is free, on every window, in the same grid as raw ROI.
RULE 05
Sample size displayed everywhere
Every window and every sport card carries a sample note. At under 50 settled picks the edge is invisible. At 1,500+ a 54% true edge reaches statistical significance. We say so.
RULE 06
Weekly archive is append-only
Sunday 11:59 PM ET the week freezes into weekly_snapshots. A database trigger blocks any UPDATE or DELETE. If a stat correction lands Tuesday, it affects next week's row, not the frozen one.
RULE 07
Publisher health is public
The hero shows 7-day uptime on the publish pipeline plus missed windows in the last 24 hours. If we stop posting, you see it. No hiding downtime.
/ FAQ
Is this record live from the database?
Yes. Every number on this page is computed from the picks table in Neon at request time with a 60-second cache. No hardcoded figures. Rows are locked at post time via a Postgres trigger, so we cannot retroactively edit or delete a pick.
What is the difference between ROI, yield, and hit rate?
Hit rate is wins divided by (wins + losses). It ignores odds. ROI is net units won divided by units risked. It treats a dollar risked at -200 the same as a dollar risked at +200 in terms of denominator. Yield is net units won divided by total settled picks, a flat-staking view. A 60% hit rate on -200 favorites is losing money; a 53% hit rate at +100 is profitable. Never judge a handicapper on hit rate alone.
Why is my sample size so important?
Because at 100 picks, you cannot statistically distinguish a true 54% edge from pure luck. You need roughly 1,500 to 2,000 settled picks before a 54% true win rate becomes detectable at p<0.05. The sample note on each window tells you where this record sits on that curve.
What is CLV and why do you show it?
Closing Line Value. The difference between the odds you posted and the final market consensus at tip-off. Positive CLV over a big sample is the single most respected metric among professional bettors because it is much less noisy than short-term hit rate. We compute it in decimal odds: (posted / closing) - 1, averaged across picks with a closing snapshot.
What is vig-adjusted ROI (ROI ex-vig)?
Raw ROI flatters anyone who only beat the juice. Vig-adjusted ROI subtracts the average sportsbook hold per pick from raw ROI. We compute the per-pick hold from each pick's closing decimal odds (implied probability times two minus one hundred) and average across the window. When closing data is sparse for a window we fall back to the standard -110 hold of 4.55%. If ROI ex-vig is still positive over a meaningful sample, the edge survives the books. Leans.ai charges $299/mo to surface this metric; ours is in the same grid as raw ROI on every window, free.
Can picks be edited after the fact?
No. Every picks row is written with locked_at set to NOW() on insert. A database trigger rejects UPDATE or DELETE of locked columns. Only grading fields (result, graded_at, closing_odds) can change, and those are append-only logs, not edits.
Is this financial advice?
No. NuroPicks is entertainment and education. Past ROI or CLV does not predict future results. 21+ only. If sports betting stops being fun, call 1-800-GAMBLER or visit the responsible gambling resources page.
Entertainment, not financial advice. 21+ only. Past ROI or CLV does not predict future outcomes. If sports betting stops being fun, call 1-800-GAMBLER or visit responsible gambling resources.
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