ok lets do us some math

Heres an imaginary hand that you decide to open push. Let me know if you dont like my assumptions.

PokerStars No-Limit Hold'em, $1.00 BB (9 handed) Hand History Converter Tool from FlopTurnRiver.com (Format: FlopTurnRiver)

MP1 ($100)
Hero ($100)
MP3 ($100)
CO ($100)
Button ($100)
SB ($100)
BB ($100)
UTG ($100)
UTG+1 ($100)

Preflop: Hero is MP2 with A , K .
MP1 raises to $4, Hero raises to $100 and is allin . . .

Ok. First off assume everyone else folds except MP1 who is a sensible TAG preflop VPIP 17%/ PFR 7%

From this position hes probably raising . We'll say he raises AA-77, AK-AJ.

Hes a pretty good hand reader with x-ray vision so hes going to call your raise with everything better than AK and fold everything worse. So after you push hes calling with AA-77, and AK.

Hes raising 7.2% of hands and only calling your push with 4.8% so your push makes him fold 2/3 of the time (Ignore the fact that you already hold 1 of the Aces. It doesnt make much difference for the purpose of this argument).

2/3 of the time you win his $4 bet and the $1.50 blinds for a total of $5.50

1/3 of the time he'll have 77+, AK and hell call. Against this range you will win $101.50 41.97% of the time and lose $100 58.030% of the time

1,047,930,048 games 3.745 secs 279,821,107 games/sec

Board:
Dead:

equity (%) win (%) tie (%)
Hand 1: 58.0304 % 49.40% 08.63% { 77+, AKs, AKo }
Hand 2: 41.9696 % 33.34% 08.63% { AKo }

So the total EV calculation is

2/3 * +$5.50
+ 1/3 * 0.4197*101.5
- 1/3*0.58030*100

= -$1.47

NOTICE YOUR EXPECTATION IS NEGATIVE

In a real poker game it will be far more complicated than this. You may find players that will call with AQ or AJ and you are very likely to find players who will fold 77 but it's still proves the important point that knowing villain "might" have a strong hand massivly reduces the EV of the situation. Actually it is even worse than this in a real game since any one of the players to act after you could also wake up with a hand.