Logged picks. Every pick logged at a price before the game and graded in the open: WNBA & MLB. College football picks are tracked on their own board; their record page is pending.
The record
The picks StatIQ logged at a price before the game started, graded in the open afterwards. Wins and losses kept. Through Nov 12, 2026, we withdrew at least 87 college football picks before kickoff, so they never counted in the record. See the corrections. Loading the full ledger…
MLB · all-time logged picks2,210–1,709 · 56.4%3,919 graded picks, Jul 12, 2026 to Oct 28, 2026 · flat 1 unit a pick · return on stake +8.18% · 3,550 of 3,919 at licensed books · as logged +8.20% · 3,550 graded picks at licensed books, Jul 17, 2026 to Oct 28, 2026includes corrections through Nov 16Every MLB pick and loss
WNBA · recovered pre-restart window382–314–8 · 54.9%704 graded picks, Jul 8, 2026 − Sep 27, 2026 · flat 1 unit a pick · recovered from our daily-board archive, never added to the live recordThe recovered WNBA record
Past results, not a forecast. Record read Nov 18, 5:04 PM ET.
Logged picks. Every pick logged at a price before the game and graded in the open: WNBA & MLB. College football picks are tracked on their own board; their record page is pending.
Cumulative units by game day, Jul 12, 2026 to Oct 28, 2026 (98 game days, evenly spaced: order, not calendar time). Model change Jul 28, 2026, marked. Included: the whole official MLB record, both eras; the shaded span is the era the headline counts. Ends at +321.49u. This line is as logged (every pick at the price we locked); the headline’s units use licensed sportsbooks only.
Why the chart never resets
The line is never restarted. A model change is drawn as a rule rather than as a restart, because restarting the chart at every recalibration is how a losing record disappears one cutover at a time. The shaded span is the era the headline counts; the other era stays on the same axis. The best and worst points are both labelled: a chart that only marks its peak is an advertisement.
The same picks: licensed prices (the headline) and as logged
Where the price came from: the same 681 settled tracked StatIQ picks (bets we made), previous model, read two ways
Licensed books only · the headline−3.16%−13.02u359–322 · n=681 · units over 412412 graded picks at licensed books · MLB · Jul 17, 2026 to Jul 27, 2026includes corrections through Nov 16
Other venues−4.14u · −8.80%47 excluded picks · MLB · Jul 17, 2026 to Jul 27, 2026
No book recorded+12.14u · +12.26%99 excluded picks · MLB · Jul 12, 2026 to Jul 27, 2026
Every day, and every pick inside it
one row per game day · won–lost, units, running total
Every pick (CSV)The displayed_prob column is filled for members; for everyone else it is empty, by design.
How this is drawn
One row per game day: won–lost, that day’s units, and the running total after it. Open a day to see the individual picks: the side we took, the price we took it at, how long before the game we took it, what we projected, and what actually happened. Collapsing a day never hides its result: the totals stay on the header either way.
13 game days · 681 graded picks · 85 void · 0 still open. The day totals are as logged and add to −6.08u. The headline, at licensed books: −13.02u · 412 of 681 at licensed books · as logged −6.08u.
Founder No pre-tip calls open in this MLB ledger
MLB · no open game date · Counted from the open rows of this ledger whose games have not started. Every projection and line below stays free.
A research tool, not a tip service: projections for every sport we project, and graded picks in the sports where we publish a record.
Pre-game picks in WNBA, MLB, CFB and CBB, each in its season. Our committed call on the upcoming slate: the exact side we would take, published before tip and then graded in the open like everything else. The board still shows every projection; this is our side, on the record before the games.
Free here: Every projection, the market line, and the full graded record (every graded result, every loss, the CLV decomposition) stay free. The pre-tip call itself is the paid part.
Jul 23, 202615–10+2.57urunning +5.08uprevious model
Byron Buxton›Minnesota Twins vs Cleveland Guardians · total bases · priced 11h 11m before the gameOver 1.5+133 at a US-regulated bookproj 1.93 → 1+0.11%CLV vs close✗ Lost−1.00u
The position, as logged
Our sideOver 1.5 total bases
Price we took+133 at a US-regulated book
Locked before the game11h 11m before the scheduled start
First price we ever saw+135
Where our side closed+135
What we projected, and what happened
Our projection1.93 total bases
The market’s line1.5
Probability we displayedMembers see the model's pre-game chance
Actual1 total bases
Showing picks 301–325 of 766 logged in this window, newest first. The rest are on the other pages, counted in every total above and in the CSV. Download every pick (CSV)
Every value on a receipt is read off one row of the MLB board. Where a field is not in the payload the receipt says so rather than filling it in. Every row is in the CSV, unchanged.
A graded pick is graded from the box score at the price we took, never at a better price found afterwards. Every row is permanent: positions settle, they never disappear. An open pick is logged and priced; it will settle where it is, win or lose, and it will not move.
A void put no stake at risk, so it settles at zero and sits in neither the wins nor the losses. It stays on the record because we did take that line, and that is the claim.
Closing line value: positive means we locked a better number than the market settled at. It can be negative and it is shown when it is. Beating the close and still losing is normal, and so is winning at a worse price than the close. Never read it as a return: break-even is roughly half the hold we paid, about +2.25 points. A row where CLV is not measurable is shown as unmeasurable rather than as zero, because those are different statements.
Price movement before the game is not CLV yet. The close is the last price we record before the game starts; until then a row shows the latest price we have seen, and the number moves with it. It freezes at the start, and it counts toward no CLV figure on this site until the pick is graded.
Why a de-vigged row shows two numbers. Its two raw legs (our price against the open, the open to the close) telescope to the raw figure, because an opening quote for the other side was never captured and only raw prices can be chained. The CLV we publish is the de-vigged number, which strips the book’s margin off both ends. They differ by the change in that margin, and we never average them together.
Cite this record: StatIQ Sports, “The public record”, https://www.statiqsports.com/record; MLB graded through Oct 28, 2026; WNBA graded through Oct 31, 2026. The graded rows are a public CSV, one file per sport: MLB · WNBA. Reuse is governed by the Terms of Use.
Your 7-day Founders trial starts as soon as your account is confirmed. No card.
Every tool and every sport’s projections, plus our pre-game picks in the sports where we publish them, posted when qualifying opportunities exist. Nothing is charged when it ends. Your account goes back to the free tier.
This page: logged pick performance (tracked, graded). The other kinds of evidence, never blended with this one: Projection accuracy (measured vs real closing prices) · Measurement-only forward tests (no bet placed).
How we grade: how a pick is logged before the game, priced and settled, and what would prove us wrong.
Eras and this season · current, previous and all-time, and how they add up
MLB: the 2026 season, next to every season on record
this calendar year beside every season on record
How this is drawn
Two windows, always both on screen. Once MLB has more than one season logged, the prior ones live in the all-time row below, not in a reset scoreboard. This is a calendar year, not a model era (that split is further down, and the two are different axes over the same rows).
Window
W−L
Graded
Hit rate
Return on stake
Net units
2026 season4,290 rows logged this year3,919 graded picks · MLB · Jul 12, 2026 to Oct 28, 20263,550 graded picks at licensed books · MLB · Jul 17, 2026 to Oct 28, 2026
2,210–1,709
3,919
56.4%
+8.18%as logged +8.20%
+290.43u3,550 of 3,919 at licensed books · as logged +321.49u
All-time official record3,919 graded picks · MLB · Jul 12, 2026 to Oct 28, 20263,550 graded picks at licensed books · MLB · Jul 17, 2026 to Oct 28, 2026every graded season together
2,210–1,709
3,919
56.4%
+8.18%as logged +8.20%
+290.43u3,550 of 3,919 at licensed books · as logged +321.49u
Three windows, all on screen at once
current + previous = the all-time official record
How this is drawn
A model cutover starts a new era. The previous era stays published and unedited: a scoreboard you are allowed to reset is not a scoreboard. So rather than hide two of the three numbers behind the one that flatters us, all three sit in one table, and the top two add up to the bottom one. Pick a row to make it the headline.
Window
W−L
Graded
Hit rate
Return on stake
Net units
Current model3,238 graded picks · MLB · Jul 28, 2026 to Oct 28, 20263,138 graded picks at licensed books · MLB · Jul 28, 2026 to Oct 28, 2026
1,851–1,387
3,238
57.2%
+9.67%as logged +10.12%
+303.45u3,138 of 3,238 at licensed books · as logged +327.57u
Previous model681 graded picks · MLB · Jul 12, 2026 to Jul 27, 2026412 graded picks at licensed books · MLB · Jul 17, 2026 to Jul 27, 2026
359–322
681
52.7%
−3.16%as logged −0.89%
−13.02u412 of 681 at licensed books · as logged −6.08u
How to read this page · one sport, one era, the n=20 rate gate, signs not colours
One sport at a time. You are looking at MLB. Each sport is a different model on different markets and their records are never added together, not even to make a bigger number.
One era at a time, but all three are on screen. The headline counts the era in the switch above. The other two are in the table below it, and the first two add up to the third.
A rate needs a sample. Below 20 decided picks we show the raw won–lost and a dash instead of a percentage. That rule is applied to good samples and bad ones identically.
Colour is never the only signal. Every gain or loss carries an explicit + or − and a word. The page reads correctly in greyscale.
Any number that is not the headline says so. Every subset on this page states its own n, how that n relates to the headline’s n, and what happened to the rows it excludes. If you find one that does not, that is a bug and it is the worst kind we can ship.
Voids are not wins and not losses. A player who never played settles at 0 and stays on the record labelled as such.
The cutoffs, in full. The official MLB record starts Jul 12, 2026; the current model era starts Jul 28, 2026. Anything before the first date is kept and shown but not counted. It was graded under rules we have since disowned. The excluded stretch is itemised at the foot of the era table, including when it helped us.
Return on stake is net divided by the total staked: a flat 1 unit on every pick priced at a licensed sportsbook, at the best licensed price we locked, before any bonus, boost or fee. Picks with no licensed price are counted in the won–lost and left out of the units; the as-logged reading is shown beside the headline. The denominator is the amount staked, not the number of picks, so the percentage is the same whether you read it in units or in dollars. It describes picks already graded. It is not a forecast and it is not annualised.
NFL. Pre-registered Aug 6, 2026, before the season: a measurement of a rule, not picks we sold, and no ROI is claimed. The focused test’s rows are also in the broad test, with correlated players and markets, so the two records overlap and are never added together.
CFB. A pick is settled when the box-score stats post, which can come before the game is final. If a later stat correction changes a result, it is added alongside the row; the graded row is never deleted.
By market · the same picks, split by market
The same picks, by market
an exhaustive partition of the headline’s picks: the rows add to its W–L and to the as-logged units
Why this section is here · Split
The split is an exhaustive partition of the picks in the current scope: the rows add to the headline’s W–L and to the as-logged units, and the table prints the total so you can see that they do. Rates are withheld below n=20, and on a small sample almost every cell is withheld, which is the honest answer rather than a table of noise.
A descriptive split, not a finding: it shows where the result came from, not a market we claim to be good at.
Market
Picks
W−L
Hit rate
Return on stake, as logged
Net units, as logged
Contribution
total bases
472
235–237
49.8%
−3.25%
−15.35u
hits
209
124–85
59.3%
+4.43%
+9.26u
All 2 groups
681
359–322
52.7%
−0.89%
−6.08u
This split partitions the headline exactly 2 groups
Subsets that are not the record · exchange prices and closing-line value
Two numbers that are not the record
different rows from the headline, each with its own n
Why this section is here · Subsets
Both of these count a different set of rows from the headline, and that is exactly how a number stops reconciling and starts looking like a contradiction. So each one states its own n, how that n relates to the headline’s n, and what happened to the rows it leaves out.
Kalshi fee applied0.07 × p × (1−p) per contract, rounded up to the cent
and it reconciles to the headline
✓This identity holds:222 comparable + 459 with no fillable Kalshi price = 681 graded picks
✓This identity holds:−10.77u + +4.68u = −6.08u, the headline net
A price comparison on the 222 picks where Kalshi listed our line at a fillable pre-game price (5¢–95¢; the tails are stale or untradeable and a “win” at 1¢ pays 49× and fakes the whole number). The other 459 picks ran and count in the headline only. Kalshi charges no sportsbook vig but we line-shop across many books, so the book beating the exchange is a finding, not a bug.
Voids and exclusions · kept on the record, out of the W–L
Voids, pushes and the picks still open
not in the W–L, not in the return, kept on the record
Why this section is here · Nothing settles at “never happened”
Rows that could not win or lose money. They are not in the W–L and not in the return, because no stake was ever at risk on them, and they are not deleted, because “we took this line” is the claim, and the claim stands whether or not the player played. Every count here is scoped to the era selected above, the same as the headline.
Voids: nobody could win or lose 85 rows
85 settled at 0.00u
85 void picks · MLB · Jul 17, 2026 to Jul 27, 2026 · previous model
did not play51
the game was postponed; it was never played34
✓This identity holds:681 graded + 85 void + 0 open = 766 rows we logged in this scope
The voids are in that total and not in the headline’s 681. Counting a scratch as a win would be a lie; deleting it would be a different one.
Where we cannot name the reason precisely we say “no action” rather than guess “did not play”. A postponed game is not a player sitting out, and a label that is only probably true has no business on a record page.
Pushes and open picks
Pushes (landed exactly on the number)0
Logged, game not finished0
Logged before the game, and the worst stretches · timing proof, best and worst days, drawdown
Append-only, and here is what that costs us
the worst stretches, printed at the same size as the best
Why this section is here · The promise
“Append-only” is easy to write and hard to prove. The proof is that the worst things that have happened to us are still on this page, at full size, in the same type as the good ones.
Since Nov 13, 2026, each time we update a board, we save a private, time-stamped copy of every pick as we published it. If you want to check our record against those copies, write to us and we will go through it with you.
Logged before the game, every time
✓ 675 / 675
675 timestamped picks · MLB · Jul 16, 2026 to Jul 27, 2026 · prices locked strictly before the scheduled start
6 picks have no lock timestamp out of 681 graded picks in this scope, so they cannot be checked this way and are excluded from the fraction above rather than counted as passing.
earliestgame startstightest 5h 56m
Earliest1d 5h before
Median14h 10m before
Tightest5h 56m before
This is the claim the whole record rests on, so it is measured rather than asserted: for every pick we hold the moment we locked the price and the scheduled start, and the gap is checked on every single one. The tightest was .
Arithmetic checks and CSV · every total recomputed
Check the arithmetic
every identity re-derived from the raw rows
Why this section is here · Don’t take our word for it
Every identity this page relies on, re-derived from the raw rows. If a subset near a headline cannot be reconciled with the headline, that is the worst bug this product can ship, so these are printed rather than promised, and a failing one renders as a failure on the page instead of quietly disappearing.
✓ 10 of 10 checks passed · re-derived from 766 rows in this scope
✓Passed:the eras partition the record: 1,851–1,387 + 359–322 = 2,210–1,709
✓Passed:era units add up: +303.45u + −13.02u = +290.43u
✓Passed:the market split partitions all 681 picks and −6.08u (as logged) exactly
✓Passed:the at-Kalshi subset plus the picks it excludes equals the as-logged total, in count AND in units
✓Passed:every graded pick has exactly one CLV status: 614 counted, 6 we never priced it, 44 the book moved the line, 17 no closing quote
✓Passed:675 timestamped picks were priced before the game started
If you had followed it
A backward simulation over the public record at the price each pick was logged and graded at, run as an account: each stake goes down when the price was logged and comes back the morning after the game, and the unit is re-sized only after a 20% move, the way StatIQ sizes units. Nobody placed these bets. It is not a forecast.
681 graded MLB picks under the previous model, Jul 12, 2026 to Jul 27, 2026. Voids and DNPs are left out, as they are out of the win-loss.includes corrections through Nov 16
Compounding assumes a stake that grows with the bankroll is accepted at the same price. The depth we have seen on one exchange says it would not be: hand-collected reads of ProphetX MLB player props in August 2026 found from nothing usable to about $216 resting at our price, about $71 a row on average one morning, and a read of the venue’s test environment across 4,850 price levels put the median level near $16. A stake past what rests moves the price. Sportsbooks limit stakes in their own ways, which we have not measured. The depth measurements
Past results over this window only. A different window, unit or starting point gives a different path, and the next pick is not more likely to win because these did.
The unit is 1% of the bankroll, the app’s default unit setting. Every pick is staked at the record’s 1 unit.
$1,000.00 start
StartAfter 13 game days (626 graded picks, includes corrections through Nov 16)
Re-sized unit (teal above the start, amber below)
Unit never re-sized (dashed)
Bankroll starts at $1,000.00 and ends at $872.09 after 13 game days and 626 graded picks; with the unit never re-sized it ends at $872.09. The picks includes corrections through Nov 16.
605 stakes cut to the cash on hand; 55 picks not placed
Each pick is staked at the time its price was logged, from cash not already in play; stakes and winnings come back at 6:00 AM ET the morning after the game, and the unit is re-sized only then. Picks on the same day never fund each other. 6 picks without a logged time are placed at the game's scheduled start, or just before the day's results come back when the start is missing too. Figures are exact arithmetic, not rounded to the cent per bet; a stake under $0.01 is not placed, and nothing is placed after the bankroll falls under $1.00.
Every return figure here describes picks already graded. This window holds 681 graded picks. A single pick’s return swings by about one unit, so an average return over n picks is known only to about ±100 divided by the square root of n percentage points: about ±3.8 points here, ±2.6 at 1,500 picks and ±1.6 at 4,000. That is fewer than the roughly 4,000 picks we use as a rule of thumb for a return that says much, and a rule of thumb is not a pass mark. Our pre-registered bar for a forward profitability claim is a return two standard errors from zero on at least 1,500 picks, with the error clustered by game day. Whatever its size, every window stays a record of what happened, not a promise about the next pick. How the pre-registered bar was set. Every pick in this record is in the ledger above, page by page, and in the CSV. Picks we withdrew before kickoff are listed in the corrections. The CSV’s last three columns hold the original result when a correction applies. Download every pick (CSV). The displayed_prob column is filled for members; for everyone else it is empty, by design.
Settled
−1.00u
Closing line value
CLV: the number we publish+0.11%
Basisde-vigged, both sides frozen at lock
Raw close vs our price−0.37%
Our price vs the open (raw)−0.37%
Open to close (raw)0.00%
Raw legs: −0.37% + 0.00% = −0.37%; the +0.11% above is de-vigged.
Chase DeLauter›Minnesota Twins vs Cleveland Guardians · total bases · priced 13h 8m before the gameOver 1.5+136 at a US-regulated bookproj 2.16 → 0+1.71%CLV vs close✗ Lost−1.00u
The position, as logged
Our sideOver 1.5 total bases
Price we took+136 at a US-regulated book
Locked before the game13h 8m before the scheduled start
First price we ever saw+130
Where our side closed+130
What we projected, and what happened
Our projection2.16 total bases
The market’s line1.5
Probability we displayedMembers see the model's pre-game chance
Actual0 total bases
Settled−1.00u
Closing line value
CLV: the number we publish+1.71%
Basisde-vigged, both sides frozen at lock
Raw close vs our price+1.11%
Our price vs the open (raw)+1.11%
Open to close (raw)0.00%
Raw legs: +1.11% + 0.00% = +1.11%; the +1.71% above is de-vigged.
Kyle Manzardo›Minnesota Twins vs Cleveland Guardians · total bases · priced 11h 1m before the gameOver 0.5-103 at an offshore bookproj 0.94 → 0+0.35%CLV vs close✗ Lost−1.00u
The position, as logged
Our sideOver 0.5 total bases
Price we took-103 at an offshore book
Locked before the game11h 1m before the scheduled start
First price we ever saw-120
Where our side closed-106
What we projected, and what happened
Our projection0.94 total bases
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual0 total bases
Settled−1.00u
Closing line value
CLV: the number we publish+0.35%
Basisde-vigged, both sides frozen at lock
Raw close vs our price+0.72%
Our price vs the open (raw)+3.81%
Open to close (raw)−3.09%
Raw legs: +3.81% + −3.09% = +0.72%; the +0.35% above is de-vigged.
Lars Nootbaar›Arizona Diamondbacks vs St. Louis Cardinals · total bases · priced 8h 6m before the gameUnder 1.5-200 at a US-regulated bookproj 1.03 → 2−2.38%CLV vs close✗ Lost−1.00u
The position, as logged
Our sideUnder 1.5 total bases
Price we took-200 at a US-regulated book
Locked before the game8h 6m before the scheduled start
First price we ever saw-180 on 0.5
Where our side closed-184
What we projected, and what happened
Our projection1.03 total bases
The market’s line1.5
Probability we displayedMembers see the model's pre-game chance
Matt Olson›San Diego Padres vs Atlanta Braves · hits · priced 11h 45m before the gameOver 0.5-190 at a US-regulated bookproj 1.02 → 0+1.48%CLV vs close✗ Lost−1.00u
The position, as logged
Our sideOver 0.5 hits
Price we took-190 at a US-regulated book
Locked before the game11h 45m before the scheduled start
First price we ever saw-240
Where our side closed-203
What we projected, and what happened
Our projection1.02 hits
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Mauricio Dubon›San Diego Padres vs Atlanta Braves · total bases · priced 12h 13m before the gameOver 1.5+190 at a US-regulated bookproj 1.63 → 1+4.91%CLV vs close✗ Lost−1.00u
The position, as logged
Our sideOver 1.5 total bases
Price we took+190 at a US-regulated book
Locked before the game12h 13m before the scheduled start
First price we ever saw+190
Where our side closed+155
What we projected, and what happened
Our projection1.63 total bases
The market’s line1.5
Probability we displayedMembers see the model's pre-game chance
Actual1 total bases
Settled−1.00u
Closing line value
CLV: the number we publish+4.91%
Basisde-vigged, both sides frozen at lock
Raw close vs our price+4.73%
Our price vs the open (raw)0.00%
Open to close (raw)+4.73%
Raw legs: 0.00% + +4.73% = +4.73%; the +4.91% above is de-vigged.
Nathan Church›Arizona Diamondbacks vs St. Louis Cardinals · hits · priced 11h 13m before the gameUnder 0.5+140 at a US-regulated bookproj 0.44 → 1−0.10%CLV vs close✗ Lost−1.00u
The position, as logged
Our sideUnder 0.5 hits
Price we took+140 at a US-regulated book
Locked before the game11h 13m before the scheduled start
First price we ever saw+140
Where our side closed+141
What we projected, and what happened
Our projection0.44 hits
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual1 hits
Settled−1.00u
Closing line value
CLV: the number we publish−0.10%
Basisde-vigged, both sides frozen at lock
Raw close vs our price−0.17%
Our price vs the open (raw)0.00%
Open to close (raw)−0.17%
Raw legs: 0.00% + −0.17% = −0.17%; the −0.10% above is de-vigged.
Travis Bazzana›Minnesota Twins vs Cleveland Guardians · hits · priced 15h 8m before the gameOver 0.5-145 at a US-regulated bookproj 0.89 → 0−0.17%CLV vs close✗ Lost−1.00u
The position, as logged
Our sideOver 0.5 hits
Price we took-145 at a US-regulated book
Locked before the game15h 8m before the scheduled start
First price we ever saw-145
Where our side closed-144
What we projected, and what happened
Our projection0.89 hits
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Vladimir Guerrero Jr.›Tampa Bay Rays vs Toronto Blue Jays · total bases · priced 11h 58m before the gameUnder 1.5-114 at a US-regulated bookproj 1.32 → 2+3.99%CLV vs close✗ Lost−1.00u
The position, as logged
Our sideUnder 1.5 total bases
Price we took-114 at a US-regulated book
Locked before the game11h 58m before the scheduled start
First price we ever saw-165
Where our side closed-134
What we projected, and what happened
Our projection1.32 total bases
The market’s line1.5
Probability we displayedMembers see the model's pre-game chance
Xander Bogaerts›San Diego Padres vs Atlanta Braves · total bases · priced 10h 41m before the gameOver 0.5-145 at a US-regulated bookproj 1.06 → 0+2.31%CLV vs close✗ Lost−1.00u
The position, as logged
Our sideOver 0.5 total bases
Price we took-145 at a US-regulated book
Locked before the game10h 41m before the scheduled start
First price we ever saw-190
Where our side closed-160
What we projected, and what happened
Our projection1.06 total bases
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual0 total bases
Settled−1.00u
Closing line value
CLV: the number we publish+2.31%
Basisde-vigged, both sides frozen at lock
Raw close vs our price+2.35%
Our price vs the open (raw)+6.33%
Open to close (raw)−3.98%
Raw legs: +6.33% + −3.98% = +2.35%; the +2.31% above is de-vigged.
Heriberto Hernandez›Miami Marlins vs Houston Astros · total bases · priced 16h 7m before the gameOver 1.5+140 at a US-regulated bookproj 2.02 → 3+0.99%CLV vs close✓ Won+1.40u
The position, as logged
Our sideOver 1.5 total bases
Price we took+140 at a US-regulated book
Locked before the game16h 7m before the scheduled start
First price we ever saw+110
Where our side closed+135
What we projected, and what happened
Our projection2.02 total bases
The market’s line1.5
Probability we displayedMembers see the model's pre-game chance
Actual3 total bases
Settled+1.40u
Closing line value
CLV: the number we publish+0.99%
Basisde-vigged, both sides frozen at lock
Kody Clemens›Minnesota Twins vs Cleveland Guardians · total bases · priced 16h 38m before the gameOver 1.5+140proj 2.02 → 2+2.65%CLV vs close✓ Won+1.40u
The position, as logged
Our sideOver 1.5 total bases
Price we took+140 (no book on this row)
Locked before the game16h 38m before the scheduled start
First price we ever saw+139
Where our side closed+127
What we projected, and what happened
Our projection2.02 total bases
The market’s line1.5
Probability we displayedMembers see the model's pre-game chance
Actual2 total bases
Settled+1.40u
Closing line value
CLV: the number we publish+2.65%
Basisde-vigged, both sides frozen at lock
Andres Gimenez›Tampa Bay Rays vs Toronto Blue Jays · hits · priced 11h 52m before the gameUnder 0.5+120 at a US-regulated bookproj 0.49 → 0+3.32%CLV vs close✓ Won+1.20u
The position, as logged
Our sideUnder 0.5 hits
Price we took+120 at a US-regulated book
Locked before the game11h 52m before the scheduled start
First price we ever saw+106
Where our side closed+108
What we projected, and what happened
Our projection0.49 hits
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual0 hits
Settled+1.20u
Closing line value
CLV: the number we publish+3.32%
Basisde-vigged, both sides frozen at lock
Fernando Tatis Jr.›San Diego Padres vs Atlanta Braves · total bases · priced 15h 12m before the gameOver 1.5+120proj 1.91 → 3+1.46%CLV vs close✓ Won+1.20u
The position, as logged
Our sideOver 1.5 total bases
Price we took+120 (no book on this row)
Locked before the game15h 12m before the scheduled start
First price we ever saw+120
Where our side closed+111
What we projected, and what happened
Our projection1.91 total bases
The market’s line1.5
Probability we displayedMembers see the model's pre-game chance
Actual3 total bases
Settled+1.20u
Closing line value
CLV: the number we publish+1.46%
Basisde-vigged, both sides frozen at lock
Junior Caminero›Tampa Bay Rays vs Toronto Blue Jays · total bases · priced 10h 37m before the gameOver 1.5+115 at a US-regulated bookproj 2.13 → 2−0.57%CLV vs close✓ Won+1.15u
The position, as logged
Our sideOver 1.5 total bases
Price we took+115 at a US-regulated book
Locked before the game10h 37m before the scheduled start
First price we ever saw+105
Where our side closed+120
What we projected, and what happened
Our projection2.13 total bases
The market’s line1.5
Probability we displayedMembers see the model's pre-game chance
Actual2 total bases
Settled+1.15u
Closing line value
CLV: the number we publish−0.57%
Basisde-vigged, both sides frozen at lock
James Outman›Detroit Tigers vs Chicago Cubs · total bases · priced 16h 2m before the gameOver 0.5+110 at another kind of venueproj 0.59 → 10.00%CLV vs close✓ Won+1.10u
The position, as logged
Our sideOver 0.5 total bases
Price we took+110 at another kind of venue
Locked before the game16h 2m before the scheduled start
First price we ever saw-110
Where our side closed+110
What we projected, and what happened
Our projection0.59 total bases
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual1 total bases
Settled+1.10u
Closing line value
CLV: the number we publish0.00%
Basisde-vigged, both sides frozen at lock
Jase Bowen›San Diego Padres vs Atlanta Braves · hits · priced 15h 12m before the gameUnder 0.5+110 at another kind of venueproj 0.21 → 0−0.19%CLV vs close✓ Won+1.10u
The position, as logged
Our sideUnder 0.5 hits
Price we took+110 at another kind of venue
Locked before the game15h 12m before the scheduled start
First price we ever saw-125
Where our side closed+110
What we projected, and what happened
Our projection0.21 hits
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual0 hits
Settled+1.10u
Closing line value
CLV: the number we publish−0.19%
Basisde-vigged, both sides frozen at lock
Yordan Alvarez›Miami Marlins vs Houston Astros · total bases · priced 16h 7m before the gameOver 1.5+100 at a US-regulated bookproj 2.87 → 5+2.82%CLV vs close✓ Won+1.00u
The position, as logged
Our sideOver 1.5 total bases
Price we took+100 at a US-regulated book
Locked before the game16h 7m before the scheduled start
First price we ever saw-120
Where our side closed-110
What we projected, and what happened
Our projection2.87 total bases
The market’s line1.5
Probability we displayedMembers see the model's pre-game chance
Actual5 total bases
Settled+1.00u
Closing line value
CLV: the number we publish+2.82%
Basisde-vigged, both sides frozen at lock
Marcus Semien›New York Mets vs Milwaukee Brewers · total bases · priced 11h 41m before the gameOver 0.5-103 at an offshore bookproj 1.04 → 4+1.61%CLV vs close✓ Won+0.97u
The position, as logged
Our sideOver 0.5 total bases
Price we took-103 at an offshore book
Locked before the game11h 41m before the scheduled start
First price we ever saw-140
Where our side closed-112
What we projected, and what happened
Our projection1.04 total bases
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual4 total bases
Settled+0.97u
Closing line value
CLV: the number we publish+1.61%
Basisde-vigged, both sides frozen at lock
Jimmy Crooks›St. Louis Cardinals vs Los Angeles Angels · hits · priced 11h 3m before the gameUnder 0.5-112 at an offshore bookproj 0.41 → 0+1.41%CLV vs close✓ Won+0.89u
The position, as logged
Our sideUnder 0.5 hits
Price we took-112 at an offshore book
Locked before the game11h 3m before the scheduled start
First price we ever saw-125
Where our side closed-115
What we projected, and what happened
Our projection0.41 hits
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual0 hits
Settled+0.89u
Closing line value
CLV: the number we publish+1.41%
Basisde-vigged, both sides frozen at lock
Petey Halpin›Minnesota Twins vs Cleveland Guardians · total bases · priced 14h 37m before the gameOver 0.5-115 at an offshore bookproj 1.15 → 1−0.13%CLV vs close✓ Won+0.87u
The position, as logged
Our sideOver 0.5 total bases
Price we took-115 at an offshore book
Locked before the game14h 37m before the scheduled start
First price we ever saw-103
Where our side closed-120
What we projected, and what happened
Our projection1.15 total bases
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual1 total bases
Settled+0.87u
Closing line value
CLV: the number we publish−0.13%
Basisde-vigged, both sides frozen at lock
Ryan Waldschmidt›Athletics vs Arizona Diamondbacks · hits · priced 13h 38m before the gameOver 0.5-117 at a US-regulated bookproj 0.78 → 2+1.57%CLV vs close✓ Won+0.85u
The position, as logged
Our sideOver 0.5 hits
Price we took-117 at a US-regulated book
Locked before the game13h 38m before the scheduled start
First price we ever saw-160
Where our side closed-138
What we projected, and what happened
Our projection0.78 hits
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual2 hits
Settled+0.85u
Closing line value
CLV: the number we publish+1.57%
Basisde-vigged, both sides frozen at lock
Jacob Gonzalez›Pittsburgh Pirates vs New York Yankees · hits · priced 15h 35m before the gameOver 0.5-125 at a US-regulated bookproj 0.77 → 1not measurableno closing quote✓ Won+0.80u
The position, as logged
Our sideOver 0.5 hits
Price we took-125 at a US-regulated book
Locked before the game15h 35m before the scheduled start
First price we ever saw-115
What we projected, and what happened
Our projection0.77 hits
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual1 hits
Settled+0.80u
Closing line value
Not measurable on this row: no usable closing quote was captured before the game.
Exchange cross-check
Kalshi ask, pre-game61¢
With Kalshi’s fee
Dalton Rushing›Los Angeles Dodgers vs Philadelphia Phillies · hits · priced 14h 36m before the gameOver 0.5-130 at a US-regulated bookproj 0.6 → 2−0.69%CLV vs close✓ Won+0.77u
The position, as logged
Our sideOver 0.5 hits
Price we took-130 at a US-regulated book
Locked before the game14h 36m before the scheduled start
First price we ever saw-130
Where our side closed-135
What we projected, and what happened
Our projection0.6 hits
The market’s line0.5
Probability we displayedMembers see the model's pre-game chance
Actual2 hits
Settled+0.77u
Closing line value
CLV: the number we publish−0.69%
Basisde-vigged, both sides frozen at lock
Picks 1–15 of this day’s 75. The others are on the next page; the totals above count all of them.
3,550 graded picks at licensed books · MLB · Jul 17, 2026 to Oct 28, 2026
2,210–1,709
3,919
56.4%
+8.18%as logged +8.20%
+290.43u3,550 of 3,919 at licensed books · as logged +321.49u
Before the official record, as loggedpicks from Jul 10, 2026 to Jul 11, 2026 · not counted anywhere above
11–2
13
–
–
+17.55u
Pick versions
MLB picks v1: from Jul 12, 2026.
MLB picks v2: from Jul 28, 2026. Earlier picks stay on the record as v1.
MLB picks v2.1: from Aug 5, 2026.
The two eras add up to the record computed from 3,919 rows
The pre-record row is deliberately not in that sum. It is shown because hiding it would be the cherry-pick, and it is excluded because it was graded under a method we have disowned.
The official MLB record begins Jul 12, 2026. Before that we have 11–2 over 13 picks worth +17.55u, and we do not count them. They were graded before watch-then-lock existed, under selection rules we no longer use. Note the direction. That excluded stretch was ahead: dropping it takes +17.55u away from us. A cutoff that only ever deletes losses is cherry-picking; this one deletes a win, which is the only way you can tell it is a real rule.
✓
This identity holds:
681 picks across the groups = 681 in the headline
✓This identity holds:359−322 across the groups = 359–322 in the headline
✓This identity holds:−6.08u across the groups = −6.08u as logged (the headline’s own units are at licensed books only)
0 groups are below 20 decided picks (of 2 groups), so their hit rate and return are withheld. The raw won–lost and the units are still shown, because those are counts, not rates.
+4.68u
Closing line value, on the tracked picks 614 of 681 picks
average, in probability points, against where our side closed · beat the close on 387 of 614
why 614 and not 681
✓614 counted above
–6 we never took a price on it, so there is nothing to compare
–44 the book moved the line, so our price and the close are quotes on two different bets
–17 no usable closing quote was captured before the game
✓This identity holds:614 + 6 + 44 + 17 = 681 graded picks in this scope
Never read closing line value as a return. Beating the close by less than the margin we paid to get there still loses money: break-even is roughly half the hold, about +2.25 points. The figure above is below that line.
This is the CLV of the picks in this record, and it is not the only CLV number we publish: the wider pool of every price we ever locked (including model-qualified picks we price purely to measure ourselves and never bet) is itemised on the CLV audit. Both are real and they answer different questions, which is why they are never averaged into one.
Measurement-only StatIQ picks (never in this record)
1,709
There are no pushes in this scope. Every line we took was a half-number, so nothing could land exactly on it. The row is printed anyway: a zero you can see is worth more than a category that quietly does not exist.
The 1,709 measurement-only StatIQ picks are model-qualified numbers we take a price on and grade purely to measure ourselves faster. They have their own wins and losses and none of them count here, good or bad. They are the reason the closing-line-value sample on the audit page is bigger than this record.
5h 56m
The worst of it, at full size
Over 681 graded picks · MLB · Jul 12, 2026 to Jul 27, 2026
Worst dayJul 26, 2026 · 33−36
…that day cost−9.73u
Longest losing run11 picks in a row
Deepest drawdown from a peak−28.91u
Best dayJul 19, 2026 · 38−32
…that day made+4.98u
Both directions, same type size, same panel. A record page that can tell you its best day and not its worst one is an advertisement with a table in it. Every one of these days is expandable in the ledger above. The bad ones are not archived any deeper than the good ones.
What would prove this page wrong. A pick that appears here today and is gone tomorrow; a loss that changes to a void; a start date that moves after a bad week; a subset whose n cannot be reconciled with the headline’s n. The CSV below is downloadable and diffable precisely so those are checkable rather than promised.
✓
Passed:
85 voids settle at exactly 0, so none of them moves the record
✓Passed:every logged row still carries its date and its outcome
It does not score the probability we published against what happened, and it does not repeat the backtests. That is the model report card, and printing the same calibration table in two places is how two surfaces start to disagree. Likewise the itemised list of every price we ever locked, including the measurement-only ones that are not in this record, is the CLV audit. This page answers exactly one question: what happened to the picks we published?
Raw close vs our price+0.89%
Our price vs the open (raw)+5.95%
Open to close (raw)−5.07%
Raw legs: +5.95% + −5.07% = +0.89%; the +0.99% above is de-vigged.
The picks StatIQ logged at a price before the game started, graded in the open. Wins and losses kept. Through Nov 12, 2026, we withdrew at least 87 college football picks before kickoff, so they never counted in the record. See the corrections.
Detail
Each sport on its own row, never added together: graded picks at 1 unit each. MLB and WNBA-since-restart units use licensed sportsbooks only, and each row says how many picks that covers and the as-logged figure; every other row is as logged (not filtered by book) or, for NFL, a measurement-only forward test.
No pick record: the board publishes numbers, not picks.
772 of 1,535 projections graded
Backtest, measured Aug 6, 2026
193,342 prices scored: the closing price beat our probability in every market, in both seasons scored
1,535 projections published, Nov 16, 2026 to Nov 18, 2026, across 5 markets (points, rebounds, pts+reb+ast, 3-pointers made, assists) · live board, not the backtest
Recovered382–314–8 · +5.62% return on stake over 696 decided picks
Since the restart65–48 · return on stake +8.15% (113 graded)
Since the restart105 of 113 at licensed books · return as logged +5.64%
113 graded picks, Sep 29, 2026 to Oct 31, 2026
105 graded picks at licensed books, Sep 29, 2026 to Oct 31, 2026
WNBA: 65–48 over 113 graded picks, +8.56u on flat 1-unit picks, 105 of 113 at licensed books, as logged +6.37u, graded through Oct 31, 2026
704 graded picks, Jul 8, 2026 to Sep 27, 2026 · flat 1 unit a pick · recovered pick by pick from our daily-board archive after a publishing bug wiped the live board; never added to the live figure. These recovered figures are as logged, not filtered by book
CFB. Since November 2026, a college football pick keeps the line, price and book it was logged with. Before then, they could update until kickoff. Through Nov 12, 2026, we withdrew at least 87 college football picks before kickoff, so they never counted in the record.
CBB. Each pick is logged before tip-off and graded in the open, wins and losses alike. Past results, not a forecast.
WNBA. Every recovered pick is listed and downloadable. Past results, not a forecast.
MLB. Logged before first pitch and graded in the open, wins and losses alike. Past results, not a forecast.