Cash games, unlike tournaments, are about finding any advantage in a given situation, no matter how slight. In theory, if you have a .01 percent edge in a one million dollar pot you are giving up $100 if you do not take it. Of course bankroll considerations always are factors in these types of situations. However, if you are properly rolled for the games that you are playing and if you have good emotional control you should have no problem getting into spots that are close to neutral expected value.
Much like the article I wrote about not running the cards more than once a few months ago as professionals we need to use any edge that we can get versus recreational players. One of the biggest things that professionals do better than non-professionals is tilt control. I would have no problem flipping for large sums of money if I knew that my opponent played badly if they got stuck. I have confidence in my ability to stay level headed even when I am down. Most non-professionals cannot do this. They start to play poor when losing and we gain even more of an advantage over them.
One of the other reasons that you should take neutral EV situations if your bankroll can withstand them is because of image. If you have listened to any of my training material you will know that I strongly believe that live image has to do with winning or losing not looseness or tightness (check out my Deuce Plays free episode Image). When you are winning you can get away with a lot more at the table. People believe your bets, give you more respect and tend to stay out of your way. This is a really good thing in holdem when more often than not people do not connect with the flop. If you take a neutral EV situation and win a large pot knowing how to correctly leverage your image will give you a huge advantage over the table. It is almost a kin to having a big stack in a tournament. In fact, if you can correctly adjust to having a losing image (tightening up, bluffing less, and value betting more) than winning a big pot is way more influential than losing one. Because of this we want to get ourselves into situations when we can win big, so long as the EV is at least neutral. You can even make the case that you can take slightly the worst of it if your opponent is prone to badly tilting. The small percentage points that you give up in the actual hand can be more than made up by your opponents’ future bad play.
On the flip side of this argument if you are constantly getting into close to neutral EV situations and your roll is not large enough to withstand the variance there is a good chance of going broke or in mathematical terms what we call risk of ruin. Poker players traditionally underestimate their risk of ruin. This is why, when playing high enough, no matter how bad the opponent is if he has an unlimited source of money there is always a chance that he can break you. As the skill gap diminishes and the stakes get larger, especially in games where the EV runs a lot closer like PLO, this risk of ruin becomes a real possibility.