Yesterday's picks: how the model did (July 4 MLB)

Independence Day belonged to baseball, and the Verdexed model came out of it a winner where it mattered most. With the NBA, NHL, and NFL all in their offseasons, Major League Baseball had the full Saturday board to itself, and the two sides the model priced as its cleanest, highest-confidence leans both cashed. That is the version of a favorite-leaning day that grades well. It was not a clean sweep, though. A couple of priced favorites fell, one of them ugly, so this reads as a solid winning day carried by the top tier rather than a wire-to-wire rout for the model.

The scorecard

The day was won at the top of the board. Verdexed publishes its graded record, Brier score, and per-confidence-tier breakdown on its public accuracy pages, and those exact figures for a single day were not cleanly retrievable at the time of writing, so July 4 is described here qualitatively rather than with a precise hit-rate. The shape is clear, however: the two sides the model would have rated as its safest projections both delivered, which is the outcome a favorite-leaning system is built to bank.

The high-confidence tier held. The Los Angeles Dodgers, the board's heaviest favorite and the model's cleanest lean, closed out a 3-0 win over San Diego behind Yoshinobu Yamamoto. The single widest starter-gap edge on the card, the Pittsburgh Pirates as a road favorite behind Braxton Ashcraft against Washington and Zack Littell, cashed just as cleanly at 7-1. When the two projections the model prices as its surest reads both land, the day grades as a win before the coin-flip games are counted.

The board did produce real misses, and this column names them rather than bury them. The New York Yankees, a home favorite against Minnesota, were routed. The Tampa Bay Rays, carrying the longest active winning streak in the American League, were walked off. Two priced sides falling on the same day is why this grades as a good day and not a great one. On the longer view, the model's published running record has hovered in the high-60s percent range on win accuracy, and a day where the top two leans hold is consistent with that clip, not a departure from it.

Best call / worst miss

The sharpest call was the Dodgers, and Yamamoto made it look easy. Los Angeles was the model's heaviest chalk on the card, and Yamamoto struck out 10 over seven innings while allowing three hits, with Freddie Freeman driving in two on a home run in a 3-0 shutout that handed San Diego an eighth straight loss. A top favorite winning by shutout is the cleanest possible cash for a model that leans chalk. The Pirates deserve a mention in the same breath: Ashcraft won his fourth consecutive start, allowing one run over 5 2/3 innings as Pittsburgh got back to .500 with a 7-1 win, the day's biggest edge landing exactly as projected.

The worst miss was the New York Yankees. Installed as a home favorite against the Twins, New York fell behind 6-0 before it batted twice as rookie starter Brendan Beck was tagged for three first-inning home runs, and Minnesota went on to hit six on the day, the first time the Twins had done that in nearly three years, in an 11-4 blowout. Josh Bell homered in consecutive at-bats late to bury it. A home favorite losing by seven is the single worst kind of result for a chalk-leaning model, and this one was not close. The Rays loss stings less on inspection: Tampa Bay led into the ninth before Yordan Alvarez ended it with a two-run walk-off homer, his second of the night, in a 10-8 final. That is a strong side losing to a great player on one swing, closer to variance than to a bad read.

What it means

The honest read is a good day, driven by the part of the board that should drive it. The two highest-confidence leans, the Dodgers and the Pirates, both cashed, and one of them was a shutout. That is the tier the model lives and dies on, and it held. The misses are worth naming and worth keeping in proportion. The Yankees blowout was a genuinely bad beat for a favorite, but a rookie starter surrendering three homers in the first inning is a pitching collapse, not evidence the projection was wrong to lean New York at home. The Rays loss came down to one Alvarez swing in the ninth of a game they led. Neither result changes the underlying picture. Across the longer run the model is still tracking near its usual high-60s clip on win accuracy, and July 4 is a data point in favor of that, with the sides it priced hardest on the right side of the ledger. A favorite is a probability, not a promise, and yesterday the safest probabilities came through while a couple of narrower ones did not.

Responsible play

**This is a record of past model output, not betting advice.** Verdexed grades its model in public for transparency and entertainment. Past results do not predict future outcomes, nothing here is a recommendation to place a wager, and no pick is ever a guarantee, even a strong model loses regularly.

**You must be 21+** (or the legal age in your jurisdiction) and located where sports wagering is legal to bet. Sports betting is not available everywhere; know and follow the laws where you are. Never bet more than you can afford to lose, and never chase losses.

If gambling is affecting you or someone you know, help is available. In the US, call or text **1-800-GAMBLER** (1-800-426-2537), or visit ncpgambling.org. Please play responsibly.