Filtered picks
0
Pre-qualified before the match
AIPICKS INTELLIGENCE / PUBLIC TRACK SHEET 001
THE PROOF
ROOM.
No victory lap. No hidden losses. Every official AI pick we lock — revealed after settlement, wins and losses together. One record. Two views.
50 PICKS · ALL REVEALED FILTERED · 26 QUALIFIED PICKS · SYSTEM SELECT
▰ System overview / FILTEREDFILTERED PROGRAM · NOT LAUNCHED · 0 PICKS
W / L record
—
Close Loss counts as a loss
Win rate
—
Wins ÷ settled · voids excluded
Avg. odds
—
Mean decimal / settled
Avg. EV
—
Mean published EV / settled
Longest loss run
—
Consecutive losses — risk is real
◆ FILTERED PROGRAM
COMING SOON.
Filtered picks are the pre-qualified subset that passes our full screening before the match starts — the same picks that power the $2 / $3 / $5 packages. This performance view launches with the paid program. Today it holds zero picks, which is why every statistic in the overview above reads empty.
01 // QUALIFYFull screening locked pre-match — a winner can never be promoted in afterwards, a loser can never be removed.
02 // SAME SOURCEFiltered picks are a subset of the ALL PICKS record — both views calculate from the same underlying entries.
03 // THEN PUBLISHAfter settlement every filtered pick appears here with its receipt, exactly like the current full record.
Nothing is cherry-picked retroactively: qualification is fixed before kick-off. Meanwhile the complete official record — every win, every loss — is open for inspection.
▰ All picks performance / ALL50 RECORDS · INCLUDES THE QUALIFIED 26 · STAKING NOT PUBLISHED
Total picks
—
Every official pick in this view
W / L record
—
Close Loss counts as a loss
Win rate
—
Wins ÷ settled · voids excluded
Avg. odds
—
Mean decimal / settled
Avg. EV
—
Mean published EV · post-settlement
Best win run
—
Consecutive wins
Longest loss run
—
Consecutive losses — read rates with sample size
⌾ Signal matrix
One square = one official pick · colour = result · newest day on top
—
WIN LOSS CLOSE LOSS VOID PENDING GOLD RING = QUALIFIED (SYSTEM SELECT)
◉ Results split
ALL PICKS—DECISIONS
Wins—
Losses—
Void / pending—
A win-rate percentage is not a measure of profitability on its own — sample size is always shown.
⛨ Verification protocol
◇Pick locked privately before each match
◇Published here only after settlement
◇Full outcome trail, including losses and voids
◇Corrections kept, never silently overwritten
Current records are post-settlement imports — independent pre-match timestamping is planned, not yet live.
THE PUBLIC LEDGER.
Search every official pick. Open a record for its receipt: timing, odds, EV and settlement evidence.
| Pick ID | Date | Match / selection | Odds | Conf | Tag | Result | EV | Evidence |
|---|
—
01 / 01
HOW THE PROOF WORKS
Before a match: the AI locks its selection privately with odds and timestamp. After settlement: the pick, its odds, EV and result are published here — wins and losses alike. Everything is revealed post-match, so nobody can tail our positions, and nothing is ever removed from the record.
01 // LOCKPrivate pre-match selection + odds
02 // SETTLEResult confirmed, then published
03 // AUDITFull public export, losses included
Current entries are post-settlement imports (locked privately, published after results). Independent pre-match timestamping is planned — a log written after the fact is not independent proof, and we say so plainly.
THE MATH, WITHOUT THE HYPE
Every official pick counts once. We publish record, win rate, average odds, average EV and streaks — stakes, units and bankroll are proprietary and never shown here. EV figures appear only after settlement, never as a promise.
Win rate = wins ÷ (wins + losses) · voids excluded · Close Loss = full loss
Avg odds = Σ odds ÷ settled · Avg EV = Σ published EV ÷ settled
Stakes · units · ROI · bankroll → proprietary, not published
Avg odds = Σ odds ÷ settled · Avg EV = Σ published EV ÷ settled
Stakes · units · ROI · bankroll → proprietary, not published
Unweighted averages: every official pick counts the same regardless of size. EV is recorded post-settlement and implies no future outcome. No commissions, no execution modelling — raw results only.