Yesterday's picks: how the model did (September 28)

Verdexed's graded line for Monday, September 28, is zero picks, zero correct, zero incorrect. The slate was a single game, the Eagles at the Bears on Monday Night Football, and it did not produce a scored prediction here because the model's per-game probability was not published. By this column's running count that makes a fifty-third consecutive slate opening with the same number. Monday's picks column said so plainly before kickoff: no edge was claimed and no side was stated, so the result on the field, a 27-7 Chicago win, does not belong on the model's record in either direction.

The scorecard

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

Major League Baseball contributes nothing to the ledger either. The regular season ended Sunday and the Wild Card round begins Tuesday, so Monday was an off day across the sport. One NFL game was the entire graded universe, and it went ungraded.

No accuracy percentage and no Brier score is cited for Monday. Verdexed's public methodology page describes a point-in-time logistic-regression model, with every prediction logged before the game and scored afterward on win accuracy and on Brier score, where lower is better, but the dated ledger behind it is not retrievable from this publishing environment. The 65 percent and 70 percent confidence tiers went unpopulated for a fifty-third straight slate, so whether high-confidence chalk held is a question Monday cannot answer.

The hand-kept ledger of stated sides holds at 13-10 across 23 decisions, unchanged, because no side was stated.

What the picks column did commit to was a market read, which is a borrowed scoreboard rather than a record. Philadelphia was quoted as a favorite in the 3.5 to 4.5 point range, with moneylines reported between roughly minus 180 and minus 230. The total had fallen from as high as 46.5 into the 41 to 42.5 band as Chicago's quarterback room thinned. The quoted favorite lost outright by 20. The game finished with 34 combined points, under every total the column cited.

Best call / worst miss

Nothing was picked, so the model has no sharpest call and no worst miss. What follows grades the column's own reads.

The best read: the falling total. The column called the total the more interesting half of the market and noted how far it had traveled toward the under as Case Keenum moved into the starting job. Result at Soldier Field: Bears 27, Eagles 7. Thirty-four points cleared nothing in the 41 to 42.5 band. The under did not land because Chicago's third quarterback struggled, which was the premise the market had priced. It landed because Philadelphia's offense produced a single touchdown, a Jalen Hurts quarterback sneak as time expired in the first half.

The worst read: the quarterback premise, and the favorite. The column framed the game as one in which Chicago might hand the ball to a quarterback it did not plan on using, and the market's conclusion was a short Philadelphia favorite. Keenum, starting for the first time since 2023 with Caleb Williams out and Tyson Bagent behind him, completed 24 of 34 passes for 247 yards, threw touchdown passes including an 8-yarder to Luther Burden on the opening drive and a 41-yard strike in the second half, and ran in a 1-yard score of his own. He accounted for all three Chicago touchdowns. The defense forced three turnovers, including a Tank Bigsby fumble and a rookie Dillon Thieneman interception of Andy Dalton in the back of the end zone while Hurts was off the field being evaluated for a concussion after a late hit. Hurts finished 16 of 25 for 153 yards with no touchdown passes and an interception. The column's "what could go wrong" section flagged Philadelphia's thin receiving group and the variance of a one-game board. It did not flag the possibility that the replacement quarterback would be the best player on the field, and that is the miss worth owning.

What it means

A quoted favorite losing by 20 is not the model's losing night, in the same way Sunday's 3-2 run by quoted sides was not its winning day. Neither scoreboard is its own. The result is the zero, the fifty-third of its kind.

Monday does offer one sober lesson about the market read itself. The line and total moved all week on the quarterback news, and both moves pointed the same way: fewer Chicago points and a comfortable Philadelphia edge. The first half of that held and the second did not. Whether the market overreacts to injuries is the kind of question a logged, graded model is supposed to answer, and it cannot answer it while its probabilities stay unpublished.

The durable read has not moved. The hand-kept 13-10 across 23 stated sides remains the only graded figure this desk can point to, and at 23 decisions the distance between 13-10 and 10-13 is a few coin flips, so it deserves to be held loosely rather than quoted as a track record.

The calendar now removes the last excuse. Tuesday brings four MLB Wild Card series and the NHL's opening night, and the NFL's Week 4 follows on Thursday. Fifty-three consecutive slates without a graded prediction is a publishing failure rather than a modeling result, and it remains the only figure here reported with confidence.

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.