He certainly doesn't think too hard.
He should not have taken the risk. Granted, it would have put runners at second and third, with the winning run at second, with one out ... at the risk of significantly reducing the chance of tying, I'd say by 75%. Being the home team, you're playing for the tie.
FWIW - Run expectancy, 2d & 3d, 1 out is in the neighborhood of 1.4 runs, assuming an average hitter coming up. First and 3rd, one out, it is 1.25 runs. runner on third 2 outs, it is about 0.35, maybe a touch less (that makes sense, you need to avoid an out in that spot, so it should be close to OBP, with some wiggle room for fielding and pitching miscues). We were down 1 at the time. Moving up is almost meaningless if you are trying to extend the game an inning by a tie, and is a killer if you don't make it. If you want somebody to walk it off, then it is a very slight help but not worth the cost of being caught. IOW, if you go, it must be a lock. This doesn't account for not letting Turner hit with one out, runner on 3rd, nor does it account for staying out of a GIDP by Turner with Robles running (it happens, but is rare). If it is 2d and 3d, you can guess they walk Turner (a meaningless run since the walk off run is on 2d already), so you'd have Soto and then Zim with bases loaded. Soto has a .410 OBP, so playing that run is well above 60%, probably closer to 70% in that case.
https://baseballwithr.wordpress.com/2020/12/21/summarizing-a-runs-expectancy-matrix/There's probably a Win Probability chart out there someone else can play with.
