Quote:
Originally Posted by JimHalpert
Ah, na a tie is certainly better than a loss. I agree that it's way higher than I thought it would be and the more I think on it the more I think it likely that I'm misunderstanding the output or it doesn't handle this particular scenario well?
It's just a more complicated situation because the tie figures in so prominently (a lot of the time with that calculator you'd be dealing with very binary W/L situations, or situations in regulation where even if OT is a possibility, the tie is a distant and still unlikely outcome).
I mean even if you ignore the loss chances, you're talking about going from an almost assured 1/2 win (the tie) to 55% of a win (the EV of making it) by going for it. Once you figure in the odds of losing once you get stopped, it makes zero sense. Of course at this point in the season you still have variance-embracing meta considerations like "our team sucks and either this daring move might inspire them and start an unlikely winning streak, or just earn us a better draft pick" that might make some sense.