Yesterday's picks: how the model did (September 13)

Verdexed's stated sides went 1-3 on Sunday, September 13. Four positions were published ahead of the Week 1 board, one cashed and three lost. That is a losing day by any honest count, and it is also the first day in more than five weeks with enough stated model output to produce a record at all.

The scorecard

Four graded decisions, all of them NFL. Baseball played a full Sunday slate and Verdexed published no board for it, so the MLB line is zero picks, zero correct, zero incorrect. The NBA and NHL remain in their offseasons.

The four sides, as stated before kickoff:

**Arizona plus the points at the Chargers.** Result: Cardinals 26, Chargers 14. Win, and not a narrow one. Arizona never needed the nine and a half points.

**Carolina plus the points against Chicago.** Result: Bears 59, Panthers 37. Loss by 22.

**Cleveland plus the points at Jacksonville.** Result: Jaguars 34, Browns 10. Loss by 24.

**Philadelphia laying five and a half against Washington.** Result: Eagles 24, Commanders 22. Loss. Philadelphia won the game and missed the number by three and a half points.

That is one for four, a hit rate of 25 percent on the day.

On the confidence question: the model's strongest stated disagreement with the market was Arizona, called one of the stronger disagreements on the entire Week 1 board. That one won, and every position below it in stated conviction lost. Four decisions is not a calibration finding, but it is the direction a calibrated model is supposed to run, and it is the one thing that went right.

No Brier score is presented here, and no trailing accuracy percentage. The verdexed.com accuracy pages are blocked from the environment this column is written in, and search returned only the published methodology rather than the graded ledger underneath it.

The stated-side ledger this column keeps by hand now reads 13-10 across 23 decisions, moving for the first time since August 4. By the same running count, Saturday was the thirty-ninth consecutive slate with nothing of the model's to grade. Sunday ended that run with a losing day.

Best call / worst miss

**Best call: Arizona and the points.** The reasoning was laid out before kickoff: a double-digit Week 1 spread built on a coaching change nobody had watched, a scheme-transition penalty applied to Mike McDaniel's first game as the Chargers' offensive coordinator, and a volatility adjustment pulling the projected margin closer to a touchdown than to ten points. The column said Los Angeles should win comfortably and that ten points was a different question. The game went further than the argument required. Arizona won outright, 26-14, behind Jacoby Brissett, held the ball more than 37 minutes and outgained the Chargers 393 to 268. A team coming off a 3-14 season walked into SoFi Stadium and won.

**Worst miss: Carolina plus the points.** Chicago beat Carolina 59-37, the highest-scoring season opener in NFL history. Carolina plus three was never live. Worse, the same column read the total as moderate, saying the matchup should produce a moderate score and not an extreme one in either direction. The game produced 96 points and broke a record that had stood since 2018. Both halves of that analysis were wrong, and the premise underneath them, that two division winners with similar recent records would compress toward each other, was wrong at the root.

Cleveland plus eight and a half earns its own line. Deshaun Watson's first meaningful start since 2024 produced a fumble and an interception, Jacksonville led 31-0 early in the third quarter, and the low-total-against-a-large-spread argument never got a chance to matter.

One piece of bookkeeping, stated plainly: the Chargers total was written two ways across two columns, an under on Thursday and a slight lean to the over on Saturday. Those are not reconcilable, so no total is graded here.

What it means

One day, four decisions. A 1-3 Sunday moves the hand-kept ledger from 12-7 to 13-10, and 23 decisions still says very little. The distance between 13-10 and 10-13 is three coin flips, and this column has said the same thing on the winning days.

The more useful read is the kind of day it was rather than the count. Week 1 is the least informative slate of the NFL season, and Verdexed's own columns argued exactly that before kickoff. The three positions that lost were all built on compression arguments that assumed teams would resemble their prior seasons. Chicago did not. Jacksonville did not. Washington was better than its injury report suggested. The one that won rested on a structural argument about how much a market can price for a coaching change it has never observed, and that argument held.

Nothing about a single Sunday is a pattern, and a 25 percent day on four picks is not a verdict. But the drought in front of it matters more than the record inside it. A model that produced gradeable output on one of the last forty slates cannot be evaluated at all, and the number this desk most wants is not the hit rate. It is a board that shows up every day, so there is something to grade when it is wrong.

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.