Loss Leaders

I haven't been playing much poker the last couple of weeks, mostly because a friend is visiting. While away from the tables, I've been asking myself a question:

Is it ever a good idea to make a -EV play?

My general answer is "no." I've always subscribed to the philosophy that I should be willing to gamble on a coin flip so that I don't look like a nit, but I won't take the worst of it like a sucker.

However, there may be times when making a play as a "loss leader" could be acceptable for metagame reasons. Here are a few possibilities:

1) Playing low pocket pairs from early position in a full ring game for deception. I accept that low pocket pairs from utg and utg+1 are probably break-even at best, but it's possible that they pay off in less obvious ways. For example, they raise your EP VP$IP, which makes it slightly more likely you'll get paid off with other premium hands from the same position. Also, I would think low pockets are more effective in lower-stakes games where opponents are more likely to stack off with overpairs. Folding these hands makes more sense in games with a lot of 3-betting.

2) Shoving all-in on the flop with a high flush draw early in a heads-up match. I don't like making this move, but it's been effective for my opponents because they tilt me easily when the flush card hits. The result of pushing a flush draw early in a match is that an opponent will read it as a fishy, overagressive play. If that kind of move is done within a broader gameplan, it could pay off later on.

3) Defending the big blind liberally. It's difficult to play out of position, but playing back at aggressive players from the blinds makes it more likely that observant stealers will make mistakes in the future. For example, they may stop stealing so much, or they may 4-bet too lightly.

I don't like making -EV plays, and I usually win by making the best decisions possible. But I wouldn't be surprised if some of these kinds of maniacal tendencies pay off by misleading opponents in their future actions.