Ozzy Analytics
60-Day Check-In · 2026 Season

60 days in.
The model took a punch. And it kept its feet.

A month ago we published the 30-day check-in: $10,000 to $32,141 in the first month, run line crushing, plus-money winners doing the work. The honest question was whether any of it would survive a real losing streak.

It got one. From Apr 30 to May 20 the bankroll dropped from $42,939 to $22,772. A 47% drawdown from peak across 20 days. The model didn't change. Quarter Kelly held the bet sizes. The edge thresholds held. And from May 21 to today, the bankroll went from $22.8K to a new all-time high.

Here's the full 60-day picture.

Started with
$10,000
60 days later
$49,054.04
+$39,054  ·  +391%
+390.5 units · 10.1% ROI on $386.5K wagered
Daily bankroll · Mar 26 → Jun 2 69 days. One real drawdown. One real recovery.
$10k $15k $20k $25k $30k $35k $40k $45k $50k Mar 26 Apr 18 May 11 Jun 2 $10,000 start $22.8k low $49.1k today

By the numbers

+391u
Profit
10.1%
ROI
68%
Run-line hit rate
10,000
Sims per game

"A 47% drawdown is what separates a model from a hot streak. The math didn't flinch. The bet sizes didn't flinch. The recovery took 13 days."

What's actually working

Run line picks
42–20 · 68%
+23.5% ROI · still the model's sharpest edge after 62 picks
Moneyline picks
128–134 · 49%
+6.3% ROI · plus-money winners paying for chalk losses
Survived the drawdown
−$20,167
Apr 30 → May 20 · −46% from peak · zero panic
The recovery
+$26,282
May 20 → today · 13 sessions · new all-time high yesterday

And one thing we still don't bet: totals (over/under). Our 2025 backtest showed no profit on totals, so we're iterating on the model rather than putting real money behind it. We'll never bet a market we haven't proven out first.

What's under the hood

⚙️
Monte Carlo simulation
Every MLB game simulated 10,000 times, plate appearance by plate appearance.
📊
Statcast + Bayesian regression
Pitch-level data (launch speed, launch angle, xwOBA, xBA), regressed for small samples and weighted by park and platoon splits.
💥
Hard-hit & barrel rates
Barrel %, hard-hit %, average exit velocity, handedness-split. We trust how hard the ball was hit, not BABIP luck.
🎯
Advanced K & whiff metrics
Strikeout rate, walk rate, whiff %, chase %, per batter and pitcher, vs L and vs R. The DNA of a plate appearance.
🧬
Cutting-edge sabermetrics
Skill-based player ratings from premier modern baseball research: contact quality, plate discipline, projected rate stats, layered into every plate appearance.
☁️
Weather & park factors
Wind speed, wind direction, temperature, and park-specific HR factors baked into every game's run environment.
🏃
Speed & baserunning
Sprint speed and baserunning ratings drive triples, doubles taken, and infield-hit probability. Small effects that stack.
🔄
Elo-blended priors
Team-strength signal merged with the simulation output so power-rankings disagreements get reconciled, not ignored.
🛡️
Advanced bankroll management
Fractional-Kelly stake sizing with edge thresholds, hard caps, and dynamic scaling. The reason the drawdown didn't end the season.

"The first 30 days were the highlight reel. The next 30 were the test. Win rate dropped from 56.6% to 48.8%, the edge held, and the bankroll still went up."

How it actually went

Phase 1: Mar 26 to Apr 24 (the build). 152 picks, 86–66, +$22,141. Run line crushed at 75%. We wrote the 30-day check-in on Apr 25 with the bankroll at $32,141.

Phase 2: Apr 30 to May 20 (the gut punch). Started hot, bankroll peaked at $42,939 on Apr 30. Then May 1 (−$5.9K), May 3 (−$8.0K), May 9 (−$7.6K), May 10 (−$3.7K), May 20 (−$3.8K). In 20 days the bankroll dropped to $22,772. A 47% drawdown from peak. The kind of stretch that makes most models get tweaked, abandoned, or "improved" mid-season.

Phase 3: May 21 to today (the recovery). No model changes. Same Quarter Kelly, same edge thresholds, same betting universe. +$26,282 over 13 sessions. New all-time high yesterday at $49,054. The math didn't change; the variance just turned back the other way.

Every pick. In the open.

All 324 picks of the season, every result, every dollar. Nothing edited after the fact. Hover any pick for the date, score, and P&L. Click any pick to see the full game analysis.

170 wins 154 losses 324 picks · 60 days +$39,054 net
Apr 2
No picks · model passed on the slate
Apr 6
No picks · model passed on the slate
Apr 24
+$291
Apr 25
No picks · model passed on the slate
Apr 26
No picks · model passed on the slate
May 7
+$1,943
May 21
+$614

Free daily picks. Receipts on the site.

Every pick, every result, every dollar tracked publicly. No paywall. No hype. Just the model and the math.