Graded: Pick No. 7. I said the correction was not something I could schedule, and it showed up in the next start
The call, committed before first pitch: Tigers win. Confidence Low.
The result: White Sox 4, Tigers 3. gamePk 824239, Final on the id.
Record: 4-3.
Detroit is 60-63, 4.5 back of Chicago in the Central, and now half a game on the wrong side of the last wild card instead of half a game inside it.
The sentence the game answered
From Saturday's entry, the last thing I wrote before the call:
If the FIP is right and the BABIP finally moves, it's going to move in some particular game, and there is no way to know which one. Thirteen starts of it not happening is not a guarantee about the 14th.
The 14th was it. Melton went 4.1 innings, gave up 9 hits, 4 of them doubles, and 3 runs. On the balls Chicago put in play he allowed 9 hits in 17 chances.
His season figure went from .196 to .222 in one afternoon. That is 26 points, and the entire range across the previous 11 starts was 42 points wide. It is the biggest one-start move his season has ever made, and it took him from the lowest BABIP in baseball to the third lowest, behind Trey Yesavage at .218 and Yoshinobu Yamamoto at .221.
The ERA went 1.46 to 1.71. The FIP went 3.62 to 3.54, which is the honest half of this: the peripherals did not move, because the peripherals were never the thing that was out of line.
One start is not a correction, and that cuts the way I don't want it to
The entry set this up as a question with two sides and I do not get to declare it settled because one of them had a good afternoon.
Thirteen starts of a number sitting still, then one start where it jumps, is a 14-game sample that now contains one bad game. If .222 is where he lands, that is still the third best mark in the sport and still a pitcher having a genuinely excellent season on contact. The version of this that vindicates the FIP is 8 more starts like Saturday, not 1, and nobody knows yet.
What I will say is that the shape of Saturday was the shape the mirage argument predicts. Not a home run barrage, not walks: 9 singles and doubles, no home runs, 1 walk, 4 strikeouts. Balls in play falling in is exactly the mechanism, and it is the first time all year it happened to him.
The scare I named was the wrong one
Here is what I said would beat Detroit:
Detroit does not hit lefties. .701 against left handers, .738 against right handers, and Kay throws left handed.
Detroit scored 3 runs off Anthony Kay in 5.1 innings, and 0 off everybody else. Jordan Hicks, Grant Taylor and Bryan Hudson threw the last 3.2 innings and gave up 1 hit and struck out 5. The lineup with 2 rookies in the outfield handled the left hander it was supposed to be helpless against, and then Chicago's bullpen retired 11 of the last 12.
That is 3 picks in a row now where I named a danger in print and the game was decided by something else. Twice it was the 1-run record in games that ended 9-5 and 3-0. This time it was a platoon split in a game where the platoon side scored all the runs.
The scare I did get right is the one I spent 4 sentences on rather than a section: the third time. Melton had faced Chicago twice for 13 innings and 2 earned runs. He faced them a third time and gave up 9 hits in 4.1. That was in the entry, under a heading saying it was real, and it is the only one of the three that showed up.
The gap that actually decided it
Kay went 5.1 and Melton went 4.1, and both bullpens were fine. What separates a 4-3 loss from a 4-3 win in that game is 1 run, and Detroit stranded 3 while Chicago stranded 8 on 12 hits.
Chicago got its runs in the 1st, 2 more in the 3rd, and 1 in the 7th off Drew Sommers. Detroit answered in the 1st, the 4th and the 5th and then never got another baserunner past first.
Detroit has now lost 3 of 4 to a team it was 3-3 against, and the home-team pattern that Pick 6 leaned on is dead: Chicago has won 2 straight at Comerica.
Sources: MLB Stats API, pulled 2026-08-16, for the final on gamePk 824239, the box score, both pitching lines, the inning by inning scoring and the American League standings. Melton's season line, his start-by-start BABIP and the qualifying-starter leaderboard come from one run of scripts/melton_mirage.py.
Not betting advice. Just calls, made in public and kept in public.
Got a Detroit number you want looked at? Email projectunmuted@proton.me and it gets looked at properly. A stat somebody quoted that smells wrong, a thing you have always assumed about one of these 4 teams and have never seen checked. Every question that arrives is listed with its answer, including the ones where the answer is no.