Yesterday's picks: how the model did (August 9)

There is no record to grade for August 9. This desk published no picks column for Sunday's MLB board, and the model's own scored output for the day was not retrievable at publication. The honest entry is zero picks, zero correct, zero incorrect. That is the third scorecard in five days without a graded line, and the first of them on which there was no published column to fall back on either.

The scorecard

Zero stated directions, zero correct, zero incorrect for August 9.

Two things stay separate here, and the distinction matters more on a blank day than on a good one. Verdexed's model logs a prediction before each game and scores it once the game finishes, whether or not any column quotes it. This desk can only grade output it can actually read. On Sunday neither channel was available: no board was published for the slate, and no per-game probability, slate accuracy figure or Brier value for August 9 was retrievable from Verdexed's public accuracy surface at the time of writing.

The stated-side ledger therefore does not move. Five graded cards read 2-0, 1-1, 2-2, 3-2 and 4-2, or 12-7 across 19 decisions. It last moved on August 4 and has now sat still for five days.

The missing calibration figures are the longer-running problem. By the count kept in these columns, the streak without a retrievable trailing accuracy or Brier number had reached six consecutive mornings as of the August 7 board, and nothing published since has closed it. Readers should treat that absence as an absence. It is not a quiet good result, and it is not a quiet bad one.

Best call / worst miss

There is no best call and no worst miss, because there was no call. Naming one after the fact from a slate the desk never handicapped would be the exact failure this column exists to avoid.

What can be put on the record is the slate itself, confirmed against final scores.

**Braves 2, Yankees 1** in 10 innings at Yankee Stadium. Michael Harris II opened the 10th with an RBI single, and Atlanta avoided a three-game sweep in a game both clubs finished with four hits and one home run.

**Cardinals 7, Rockies 4** at Busch Stadium. J.J. Wetherholt broke a 4-4 tie with a two-run double to center in the eighth, scoring Masyn Winn and Nathan Church, and Alec Burleson homered.

**Mets 11, Pirates 1** at PNC Park, a day after Pittsburgh had shut New York out 9-0.

**Cubs 10, Royals 2** in Kansas City, a series-clinching afternoon for Chicago.

None of those outcomes tells a reader anything about the model, and no claim is made here about which sides it held. That counterfactual is unknowable and does not belong in a record.

What it means

Nothing about Sunday changes the read on the model, because Sunday produced no evidence about the model. The ledger stands where it stood on August 4, at 12-7 over 19 decisions, and 19 decisions carry essentially no signal in either direction. Anyone treating that number as a track record, in either direction, is reading too much into it.

The finding is about publishing continuity rather than accuracy. Three consecutive days have now passed with no board, and more than a week has passed since a trailing accuracy or Brier figure could be quoted. A model that grades itself in public is only as credible as the days it reports, and the failure mode for an accountability desk is not a losing day. It is a quiet gap, where missing days get filled in by a reader's assumption that they must have gone well.

So the gap gets published. August 5 and August 6 were blanks with a column attached. August 7 carried a board with no numbers on it. August 8 and August 9 carried nothing at all. The record for the week is thin, and stating that plainly is the only version of this that is worth anything.

The standing note from earlier cards still applies when the numbers return. A record assembled from named games in a column is a weaker instrument than the model's own scored output, and cards built that way should be weighted accordingly against the tracked accuracy figures whenever those become reachable again.

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.