The issue is betting value vs. expected value
given the evidence. That goes for things like looking at sires to figure out who may move up in the wet.
Yeah, you could "make a case" for an IR move up, and it was in fact made. You could make a case for a bunch of horses running a better number than IR as well. That IR ran well in the Derby does not make using the sire data as reason to bet him over others. It's a data point to put into play with others, including his odds.
JB is exactly right, and it has nothing to do with "gambler's ruin". If you're looking for a sure fire thing, get out now. Even Dutch book cases can lead to ruin for the casino, if you get a bad run. You won't have perfect evidence ever. It's just about putting yourself in the best position to win. I will book the correct side of the Monty Hall paradox all day. I could lose, but I wouldn't bet on it.
Play some hold'em. You will get beat, all in, on the river, because the other guy lands his only out. If losing the right bet is something you can't stomach, then you shouldn't bet at all.
