BY 
Where can you find the best sports picks today, free and verified, and how much should you actually bet on them? CAPTRACKER handles the first half. Its free leaderboard ranks 900+ tracked handicapper profiles by units won and ROI, and every pick is auto-settled against ESPN results. The second half is your job.
Short answer: choose cappers by their settled record, then ignore how much they staked. Bet a flat 1% of your own bankroll on each pick you follow. Units on a leaderboard belong to the capper, and yours are a different size. Copy the pick and leave the sizing behind.
Best sports picks today start with a settled record
A verified record means the pick was timestamped before the game and nobody could edit or delete a loser afterward. The result comes from ESPN data, not from anyone's memory. You can see how that works on the methodology page.
That fixes the honesty problem. It doesn't fix the sizing problem. A capper who fires ten units on a hunch and one unit on a grind gets the same single line in the pick count. You have to work out what the totals imply.
Free verified analytics still leave the unit question open
Take X_Underscore_X (football): 27-17-7 over 51 picks, +389.8 units, 51.6% ROI. Divide 389.8 by 51 and you get about 7.6 units of profit per pick. At one unit a pick, a 27-17 record can't come close to that. The stakes ran far bigger.
Now look at Timely-Conclusion532 (MLB): 83-38 over 121 picks, +61.4 units, 19.8% ROI. That's roughly half a unit of profit per pick. Two cappers, two unit sizes. Setting their totals side by side is like comparing dollars to yen.
ROI is the fairer yardstick because it scales with what was risked. Units show the size of the swing. ROI shows the return per unit risked. The glossary defines both if you want the exact terms.
Hot week versus the season base rate
The recent window looks thrilling. TheLineTamer (MLB) is 7-3 over 10 picks, +68.2 units, 94.8% ROI. Bugleboy98 (MLB) is 9-2 over 11 picks at 76.7% ROI, and Jared2338 (MLB) is 8-5 over 13 picks at 71.2%. Those are tiny samples.
Ten picks at one unit each can't produce 68.2 units, so those stakes ran big. Big stakes on few picks is exactly how a record swings hard. One bad night and that ROI looks very different.
Now the base rate. Across the full pool, 614 of 1183 cappers (52%) are profitable over 7204 settled picks, with a median ROI of +8.3%. Among the 83 cappers with 15+ settled picks, 45 (54%) are in the black. A hot week is a headline, and a full season is evidence.
Hockey's my beat, so GuyBostonSports (Hockey) caught my eye: 273-307-3 over 583 picks, -59.2 units, -10.2% ROI, a big sample and an underwater result. SportsBettingPicks (MLB) sits at 354-312-7 over 673 picks, +4.7 units, +0.7% ROI, barely above zero. Large samples don't manufacture an edge. They just deliver the truth sooner.
I'll admit my three blown bankrolls shared one habit. I sized by how good a pick felt. Feel is an expensive way to stake.
A stake plan for today's verified picks
This is the plan I'd hand a new bettor:
- Set one unit at 1% of your current bankroll, and keep it fixed for the week.
- Bet one unit on each pick you follow, whatever size the capper used.
- Cap your total risk on any single day at five units.
- Log the price and result of every bet so you can check your record against the tracker's.
Why so strict? Five straight losses at 10% of bankroll each leaves you with about 59% of what you started with. Five straight losses at 1% leaves about 95%. Streaks find everyone, and at -110 you need to win 52.38% just to break even.
Where the daily feed fits in
The daily feed lists today's picks, and CAPTRACKER's convergence signals flag when independent cappers agree on a side. Agreement is a reason to look closer. Keep your unit the same size anyway.
The leaderboard is free, and picks can't be edited or deleted after they're posted. So you can check every number in this column yourself. No record guarantees the next pick, which is exactly why the sizing matters. Pull up a name, read the sample size, and stake it on your own bankroll.