Yesterday's picks: how the model did (MLB, September 22)

Verdexed's graded line for Tuesday, September 22, is zero picks, zero correct, zero incorrect. Eight games were scheduled in baseball, the only league with a slate in front of this desk, and not one was scored here, because the model's per-game probabilities were not published. By this column's running count, that is the forty-seventh consecutive slate opening with the same number. The prices the picks column quoted alongside four of those games went 2-1 where a favorite was stated, which is a fact about the market and not a result the model earned.

The scorecard

Picks made: zero. Correct: zero. Incorrect: zero. Accuracy on the day is undefined, because the denominator is zero.

Tuesday's MLB best bets column said as much in its first paragraph: verdexed.com was unreachable from the environment that column publishes behind, the prediction, edge and accuracy pages did not surface through search, and no edge was claimed on any of the eight games. Nothing in it can be graded as a pick.

No accuracy percentage and no Brier score is cited here. A public search returns the Verdexed methodology page, which describes logging every prediction before first pitch and scoring it on win accuracy and on Brier score, where lower is better, It does not return the ledger underneath it. A description of a process is not a number. The 65 percent and 70 percent confidence tiers went unpopulated for a forty-seventh straight day, so no per-tier breakdown can be reported.

The hand-kept ledger of stated sides holds at 13-10 across 23 decisions, unchanged.

What can be graded is the market half, on the four games the column covered. Three carried a clear favorite: two won and one was shut out. The fourth was priced near even money.

Best call / worst miss

Nothing was picked, so there is no sharpest call and no worst miss belonging to the model. What follows grades the quoted prices and the column's own reporting.

**The price that held best.** Los Angeles was quoted at minus 185 with a total of 10, and the column noted the number had moved from near minus 160 on Yoshinobu Yamamoto's projected start. Yamamoto started and threw seven scoreless innings with 10 strikeouts, tying his major league career high for the second time this season. Result at Dodger Stadium: **Dodgers 3, Padres 0**. The favorite held and the total landed seven runs under the number, the one case Tuesday where the stated reason behind a price and the result lined up.

Philadelphia at minus 142 held as well. Zack Wheeler struck out 11, including a franchise-record seven consecutive to open the game, and Bryson Stott broke a tie with an RBI single in the eighth: **Phillies 6, Brewers 4**, against the Milwaukee club with the best record in baseball.

**The worst read: the game the column called the most consequential on the schedule.** Seattle was quoted near minus 148 against Houston at plus 126, with Logan Gilbert listed to start. Gilbert lasted four innings and gave up four runs on 10 hits, his highest hit total in a game this season. Result at T-Mobile Park: **Astros 7, Mariners 0**. Christian Walker hit a three-run homer in the third, Isaac Paredes added a solo shot in the sixth, and Yordan Alvarez hit his 41st home run and drove in three, tying Junior Caminero for the American League lead. Houston pulled even with Texas at 78-79 atop the American League West after the Rangers lost 6-3 to the Mets, with five games left and the head-to-head tiebreaker in Houston's pocket. The one game on the board with a division riding on it is the one where the quoted favorite was shut out.

**A reporting error that belongs in this column too.** The picks page listed the Pirates and Cardinals at Busch Stadium in St. Louis. The game was played in Pittsburgh, with St. Louis the visiting club. Result: **Pirates 2, Cardinals 0**, a combined one-hitter. Jared Jones struck out nine without a walk over seven innings and retired the first 16 hitters he faced, and Ryan O'Hearn drove in the first run against Andre Pallante. The market read, near even money between two clubs playing for nothing, was fine. The venue was wrong, and a desk that grades itself has to say so.

What it means

Two quoted favorites holding and one losing across a four-game read is a small sample of other people's opinions. Reporting it as a good day would be the exact dishonesty this page exists to avoid. The result that belongs on the model's record is the zero.

The durable read has not moved in nearly two months. The ledger this column keeps has been frozen since late July, when the per-game board stopped being readable from here, and a hand-kept 13-10 across 23 stated sides is the only graded figure on this page. At that sample size the distance between 13-10 and 10-13 is a few coin flips, which is a reason to hold it loosely rather than lean on it. Forty-seven consecutive slates without a graded pick is a publishing failure rather than a modeling result, and it has to be fixed before any accuracy claim here carries weight.

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 of legal betting 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.