[Poshua]

[ QUOTE ]
In lower limit games, this can add up to quite a large percentage of winnings over time. If everyone at the table is paying for random stuff out of their stacks, you could be seeing $20-$30+/hour disappearing off the table in addition to rake and tips.

[/ QUOTE ]

Unless you're playing in California (as I discuss in the footnote at the bottom of my post) I think this is probably not important enough to worry about, for the following reasons:

1. While you can control your own spending practices, you can't make your opponents pay out of pocket, so it only makes sense to analyze how drink tips affect your stack size, not your opponents'.

2. Marginal increases in your stack only impact winnings in hands where you are all-in and your opponent covers you; such situations should only account for a small fraction of your net win. Also, if you are +EV, you should be accumulating chips and covering your opponents more often than they cover you, further diminishing the importance of marginal increases in your stack size.

3. As you point out, while having a bigger stack will increase your net if you are +EV, this will increase both your wins and your losses. The net effect will be a small fraction of the gross positive effect.

4. The main situation where the positive effects would manifest, as you describe, are situations where you repeatedly double through other players (with you being covered) rendering the effect exponential. For this to happen, it would have to be a game where lots of other players at the table have also built stacks of several times the max buy-in, enabling them to cover you after you have doubled up once or twice. At most games, this would be an unusual circumstance.*

*One key exception would be a game with a very low buy-in cap, where players frequently bust out and re-buy, and other players routinely build monster stacks many times the buy-in. I have seen games like this, particularly in the Bay Area, and I understand this is common in LA. In this case your strategy might indeed have a measurable (though, I would still contend, not particularly large) effect.