Every pick. Every result. Every reason.
Every pick is published before the game starts, graded against the real result, and shown win or lose. No deleted losers, no hidden record. For our most-developed model, you can open any pick and see the exact factors that drove it.
What the model is betting today.
The transparency isn't decoration — it's the product.
Why this pick
Most services tell you what to bet. We show you why. Open any MLB pick and you'll see the specific factors the model weighed — starter matchup, recent form, park, and lineup — each with its actual weight, win or lose. MLB has full factor explanations today; NHL, NBA, and WNBA roll out the same view as each model's calibration is validated. We'd rather ship it one league at a time and have it be right than ship it everywhere and have it be hand-wavy.
No confidence theater
We don't show confidence tiers. We used to — and when we audited them, they didn't actually predict outcomes. So we retired them. Instead you see what we'd genuinely do with the bankroll: BET, TRACK, or pass. A label that doesn't survive its own backtest is marketing, not information.
Corrections published
When we find we've been wrong about our own numbers, we publish the correction. In May we caught our MLB moneyline model reporting 75% confidence on picks that historically won 53% of the time, fixed it, and wrote it up publicly. That's the kind of thing that usually gets quietly patched. We publish it because trust is the product.
What we can't do yet
What we don't have yet: deep factor explanations for every sport (they're rolling out per league), and a long enough track record on some bet types to make strong claims. Where a number is thin or a model is young, we say so on the page. We'd rather under-promise than launder uncertainty.
Get notified when new sport explanations roll out, or when we publish another calibration note. Low volume, no spam — just the next thing worth knowing.
Everything is tracked. Nothing is hidden.
Every pick is published before games start and graded against real results. Browse the history yourself.