UTG (unknown, havent seen anything stupid, so lets assume he is a reg) opened for 3bb, fish on the button shoves for 25bb, we are on the SB with AKo.

Lets do some calculations. Okay UTG range is usually tighter than it should be, soo lets say 14% - 186 combos. 66+,A2s+,KTs+,QTs+,JTs+,ATo+, KQo
Now, what are some possible UTG calling ranges if we decide to shove:

a) I would expect that reg calls only KK+ (might be a bit nitty though).
Therefore we have 1- (6/147)=0.9592% fold equity. (ignoring the BB for a simplification). If we go AI vs the fish, I expect a range like this: 88+,A4s+,KJs+,QJs,ATo+,KQo (might be too wide though). Vs this range we do have 60% equity.

So in 95.92%, we will be in a pot 54bb with 60% equity, so we are suppose to win 54(0.6)= 32.4 which is 7.4 profit. So we are winning 7.4 in 95.92% cases --- 7.4*0.9592= 7.09808

In 4.08% we got called by UTG KK+ and we have 18.48% equity vs that range. So on average in a pot of 150, we are investing 75 and our avg return is (0.1848*150) = 27.72. It means that we are losing 75-27.72=47.28. In the side pot of 76bb, we have 14.11%equity, so our avg return is 10.72 therefore we will lose on avg 25-10.72=14.28.

Losing 14.28+47.28=61.56 in 4.08% cases --- 61.56*0.0408= 2.511648

So to summarize we are making 7.09808-2.511648= 4.586432 profit/hand (without rake and the BB player), but I still believe its gonna be slightly profitable.

b) Now lets calculate the ev if the UTG calling range is QQ+,AK.
Then our FE is 1-21/147= 0.857*7,4 = 6.3418 profit (if UTG folds)
In case we got called, we are HU with 38.82% equity in a pot of 150 which makes -16.77 and the side pot is 76bb, our equity is 25.5%, so our avg return is 19.38, so we are losing 25-19.38= 5.62

Okaay so loses: 5.62+16.77= 22.390.143= 3.20177 of loses compared to winnings: 0.8577,4 = 6.3418
*Total 6.3418 - 3.20177 = 3.14003*
Thats a bit less compared with the scneario when the utg is calling only KK+. From that we can conclude: the more nitty the UTG, the more incentivized are we to go for AI shove. I mean, with the rake consideration and the BB player left to act, we might be better off folding vs a more loose utg villain.

1a scenario if utg calls KK+ incl.rake

if we play only a small pot of 25+25+3+1+=54. The rake is 5% (but since we have almost 40% rakeback, lets assume the rake is actually at 3%). Then the pot of 54 becomes 52.38

now the pot is only 52.38 and instead of 7.4 profit, we will have only 52.38/*0.6-25 = 6.428 * 0.9592(the fold equity in 1.case) = 6.17(before the rake it was 7.1)

given the 3% rake the pot is actually 145.5, we have 18.48% so 145.50.1848-75= -48.1116
his side pot would be 73.72 and our equity is 14.11%, so 73.720.1411-25 = -14.598
All of this happens in 4.08% cases so lets add it together: (-48.1116-14.598)*0.0408= -2.56
---- 6.17-2.56= 3.61 including the rake which is almost 1$/hand less.

2a scenario when he calls QQ+,AK incl. rake
I will copy the same from the previous case: instead of 7.4 profit its 6.428 (if UTG folds) * fold equity- 0.857 = 5.508796.
The main pot will be 145.5 (instead of 150), with our equity 38.82%, we are suppose to lose 18.5169.
The side pot will be 73.72 with equity of 25.5% which makes 73.720.255-25= -6.2014
Adding those losses together and multiplying them by their probability: -6.2014-18.5169=-24.71830.143 = 3.5347169
5.508796-3.5347169 = 1.974 profit/hand - thats more than 1$ less if we are taking rake into consideration. If we would neglect the rakeback and use only 5%rake without rb, we will see this. 1% of rake = 0.389; 5% = 1.945. Thats not even 1.2$ profit, thats pretty close, because we are ignoring the BB so far.

Calculations incl. the BB player: coming soon:)