So I have not spent enough time doing EV calculations away from the table. Although I am able to do the simple ones, I never learned how to calculate in these situations. I have a few questions. Sorry if this is something that's already been covered extensively.

The situation was: 6max, two limps, I pot in CO, BN shoves for 12.4 BB (with very wide range), BB + limpers call. I bet 20BB on flop, two players call (BB folds). UTG shoves 78.8 BB on turn, MP folds.

my hand: A♥-5♥-6-4 // board: A♦-9♥-6♥-T♠


The main pot was 60.6BB, the side pot on turn was 61.9BB and would be 211.1BB (after rake) if I called.

Having never done it before and lacking direct relevant material, I applied my current understanding to the problem and came up with:

EV = (eq[main] * 60.6) + (eq[side] * 211.1) - ( (78.8 * (1 - eq[main]) * 0.22353375) + (78.8 * (1 - eq[side]) * 0.77646625) ) with the decimals representing the ratio of pot size main:side


The result for the situation I was calculating was significantly higher than I was expecting, so if my equation is wrong I would appreciate some one telling me.

I would imagine this is mostly because BN's range was so varied that the impact of their equity was negligible given the hands involved. Knowing the UTG player I could be certain they had either a straight or monster draw like K♥-Q♥-J (which somewhat cancelled out blockers to my draws) and so I was essentially just getting 2.5:1 for a fairly standard decision. But it makes me wonder how reliable this kind of calculation can be when short-stack fish shove a wide and indefinable range, or when multiple shorter stacks contribute to a main pot with hard to define ranges? Are there ways to make it more reliable?

I also obviously am unable to input such ranges into Odds Oracle as dead cards and so had to calculate side pot equity as if there is no one else in the hand at that point. I'm no wiz so I don't know how much that would affect the result, but I feel like it will become more significant in the future if I end up doing calculations of similar situations where the short-stack has, say, a definable 10% preflop shoving range, or when they have a definable-enough gii range on the flop and the action continues much deeper. Could a wiz-er person give me an idea?

I'm also wondering if anyone has a trick or a way around this? Would taking the ratio of the active players' equity when calculated with these dead ranges accounted for, then applying it to 100% work well enough for estimating side pot equity?

Since this is a well thought out honest question I'll go through it in some detail.

I'm not going to check your equation, because it seems a bit too complicated

How I approach these side pot calcs is to break it into two different calls, which it appears you sort have did.

Here you do not have any cost associated with calling for the main pot so:

EV_main = Eq_main * size_of_main

EV_side = EQ_side * size_of_side (after you call) - cost of call

That's it. Then add those up.

So, you give the first all in villain a range - you said it's wide so let's say it's top 25% of hands. Now, you give UTG a range - eliminate the top 10% of hands as he was a limper, and say his range is 10%-30%AT, 78, (QJ, Q8, KQ):hh, TT).

EQ_side = 0.3745



EQ_main = 0.31288


EV_main = 18.97
EV_side = 0.26
EV_total = 19.22

So this is a hugely profitable call based on those ranges.

Don't worry about the dead cards - it is too hard to estimate exactly what they folded and easiest and simplest, and it is probably best to simply ignore them and treat them as unknowns. Only consider dead cards if you see them and can for sure eliminate them.

This is really helpful thanks.