Sandlot Analytics

PhD researcher by day, baseball stats junkie by night.

[Sandlot Analytics @SandlotStats]

The betting slate from the Almanac is looking good so far with the top six wagers all projected to hit.

A lot of season left, and there are crazy storylines every year (2024 Mets, 2025 Guardians), so we'll see how it all plays out!

[Sandlot Analytics @SandlotStats]

25% of the MLB season is in the books and HOBIE is maintaining a narrow lead over the pack.

[Sandlot Analytics @SandlotStats]

May 1 Update: HOBIE is sitting at #1 on the 2026 in-season accuracy tracker with 378 total error, juuuust ahead of @fangraphs. @SeanZerillo and @baseballpro are close behind. Season is 19% complete.

Live in-season tracker + free 2026 Almanac at https://t.co/3O4maeOM7j.

[Sandlot Analytics @SandlotStats]

The race is on @fangraphs @baseballpro @SeanZerillo @espn @usatodaysports @keithlaw!

[Sandlot Analytics @SandlotStats]

That's about 15% of the MLB season in the books.

The way-way-way-too-early leaderboard has HOBIE out front if we project that teams will end the season with the same winning percentage they have today.

Spoiler alert: they will not.

Follow every projection system's accuracy all season long at https://t.co/vY7niF6me8

[Sandlot Analytics @SandlotStats]

Pirates OVER 78.5
Dodgers UNDER 103.5
Rays OVER 76.5
Astros UNDER 86.5

This model is not afraid to be contrarian.

[Sandlot Analytics @SandlotStats]

Happy Opening Day!

Just dropped my MLB win total projections: https://t.co/mNn8ytsbnF

(a few bets I like in comments for those doing research)

[Sandlot Analytics @SandlotStats]

I’m with you on the OVER for COL, but teams coming off extreme negative WAR seasons rarely eliminate all of it in one year.

The Rockies had 20 hitters combine for -11.5 fWAR last season. This year they've got seven hitters projected to total -0.8 WAR. That amount of negative WAR disappearing almost never fully materializes unless the roster structure changes meaningfully, which I don't think has happened.

Only 67% of COL’s projected PA are going to players projected above 0.4 WAR. Compare that to BAL (also coming off a down year) at 90%.

I’ve got COL in the low 60s after a haircut for volatility and unreliable roster structure (otherwise my model would have had them up closer to your 67 too).

[Sandlot Analytics @SandlotStats]

@DSzymborski @fangraphs @fangraphs (powered by ZiPS) had a pretty great 2025.

Looking forward to seeing how 2026 shakes out across all the models.

Public projection scoreboard is live at https://t.co/3O4maeOM7j for anyone who wants to track model accuracy or see performance over the last few years.

[Sandlot Analytics @SandlotStats]

@usatodaysports You guys crushed it last year @usatodaysports.

Let's see how 2026 shakes out across all the models.

Public projection scoreboard is live at https://t.co/3O4maePjWR for anyone who wants to track model accuracy or see performance over the last few years.

[Sandlot Analytics @SandlotStats]

@SeanZerillo You had a solid 2025 season @SeanZerillo.

Let's see how 2026 shakes out across all the models.

Public projection scoreboard is live at https://t.co/3O4maePjWR for anyone who wants to track model accuracy or see performance over the last few years.

[Sandlot Analytics @SandlotStats]

@BretJutras You're the Ethan Hunt of AI.

[Sandlot Analytics @SandlotStats]

Just launched Sandlot Analytics as the new home of HOBIE, my MLB prediction model that’s outperformed public projections since I built it four seasons ago.

The site has an MLB model scoreboard, a public prediction contest, and lots of nerd talk.

https://t.co/L2nn2YJF2Z

[Sandlot Analytics @SandlotStats]

Fun to cash this one with nine games to go 😉

[Sandlot Analytics @SandlotStats]

My favorite 'Way Too Early' @MLB bet:

@Mariners OVER 83.5 on @DKSportsbook -- third order wins tell a story of bad luck last year and my model currently has them at 94 wins for 2025.

Throw them into a Division Winner parlay with @Orioles @Cubs @Dodgers to make it extra spicy.

[Sandlot Analytics @SandlotStats]

The six projections in the comparison were @baseballpro, @fangraphs, @espn, @keithlaw, Clay Davenport, and my own.

[Sandlot Analytics @SandlotStats]

I compared six MLB projection models across the last three seasons to see which was most accurate.

THEN, I developed a betting strategy based on those results that I used to go 14-3 last season with a 64% ROI.

Check it out: https://t.co/pg7tmYVISx

[Sandlot Analytics @SandlotStats]

Year 4 of the model is live!

It's 61-29 picking o/u the last three seasons (68%); better than both FanGraphs (50%) and PECOTA (47%) over those years. And for the nerds out there, it has a lower RMSE and MAE too.

The more it diverges from Vegas, the more accurate it is.

[Sandlot Analytics @SandlotStats]

Projection for @Dbacks : 77 wins

Improved rotation w/ Burnes, but Naylor can't make up for losing Walker and Joc; meanwhile, most of the non-Corbin offense comes back down to earth

[Sandlot Analytics @SandlotStats]

Projection for @Braves : 91 wins

The MUCH improved offense if everyone is healthy makes up for losing Fried and some bullpen regression

[Sandlot Analytics @SandlotStats]

Projection for @Orioles : 89 wins

Young core moves offense forward, but losing Corbin hurts the staff