Yesterday's picks: how the model did (September 10)

Verdexed stated no sides for Thursday, in either sport that played, so the graded line for September 10 is zero picks, zero correct, zero incorrect. By this desk's running count that is thirty-seven consecutive slates with nothing belonging to the model to score. The NFL played its first regular-season game in Australia and baseball ran a five-game getaway board, and the model neither gained nor lost anything on any of it.

The scorecard

The zero holds in both leagues: no NFL side, no MLB side. The NBA and NHL remain in their offseasons.

The stated-side ledger does not move. It remains 12-7 across 19 decisions, unchanged since August 4, and this column will not present a five-week-old figure as current form.

There is no high-confidence tier to report, because no confidence tier was attached to anything a reader could open. The 65 percent and 70 percent bands this desk exists to audit went unpopulated again. No Brier value and no trailing accuracy percentage is claimed here, because none was retrievable: the verdexed.com prediction, edge and accuracy pages are blocked from the environment these columns are written in, and search returned only the published methodology.

What can be graded is how the four prices the picks columns quoted landed, all labeled at publication as market reads rather than model output.

The football column quoted Los Angeles at minus 3.5 with a total of 48.5, near minus 198 on the moneyline, and stripped the margin out of that price to put the market's own view of the Rams near 64 percent. San Francisco won **27-7**. The favorite lost outright by twenty, and 34 combined points finished about fifteen under the quoted total.

In baseball, Seattle was quoted near minus 130 against Texas and won **4-3**. Philadelphia was quoted near minus 172 against Houston and lost **2-1**. New York was quoted near minus 334 against Colorado, the widest price on the board, and won **10-3**. Two of the three quoted baseball favorites held. Across both sports, four quoted favorites went two and two. That is not the model's record, and four prices is too thin to carry meaning.

Best call / worst miss

There is no best call and no worst miss belonging to the model, because no side was stated. Naming one after the fact would be inventing a record. What follows is graded as shape, not as picks.

The read that landed was the baseball column's warning about September workloads. It named managed innings on contending clubs as the day's largest risk and flagged Jacob deGrom specifically. deGrom then threw six scoreless innings of one-hit ball with 12 strikeouts and no walks, carrying a perfect game into the sixth, and was done. Texas lost anyway. Logan Gilbert answered with 11 strikeouts, and Julio Rodriguez hit a go-ahead three-run homer in the seventh to decide it for Seattle. The best pitching line of the day belonged to the losing team, which is the ordinary argument for grading probabilities rather than box scores.

The read that came out worst was the football column's only quantitative content. It put the de-vigged market price near 64 percent on the Rams, noted several public prediction models in the low to mid 60s on the same side, and concluded there was no meaningful disagreement to report. Every number available to that column pointed one direction and the game went the other by twenty. Los Angeles gained 122 yards. Matthew Stafford finished 15 of 25 for 155 yards with an interception, while Brock Purdy went 25 of 34 for 205 yards and three touchdowns and was not sacked once. A 64 percent side is supposed to lose about one time in three, so the loss by itself proves nothing. The manner of it is what matters.

Two other finals belong in the ledger, neither priced in the column and neither graded. Atlanta beat Tampa Bay **3-1** on an eighth-inning rally to avoid a sweep. Pittsburgh beat Chicago **2-0** to complete a sweep of its own.

What it means

One day of nothing is noise. Thirty-seven in a row is a pattern, and the honest reading is that the pattern describes publication rather than the model. The engine may be scoring predictions normally behind surfaces this desk cannot reach. A reader cannot verify that, and an absent record is evidence of neither a good one nor a bad one.

The Melbourne result is the useful lesson, and it cuts against confidence rather than for it. The market, the public models and the column all agreed on the Rams inside a narrow band, and a consensus at 64 percent is still a number that loses. Week 1 makes that worse, because a team-strength model leans on in-season results that do not exist yet. The caution the football column attached on exactly that point was the part of it that held up.

The only verified figure remains 12-7 across 19 stated decisions, last moved on August 4, a sample small enough that the distance between 12-7 and 9-10 is a handful of coin flips. Until stated sides with confidence tiers return to a page a reader can open, the accurate summary of Verdexed's recent model performance is that it is unknown.

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.