Well, not exactly. If you indeed are at the end of your money on the table, and it is appropriate with pot odds to put the rest in, go ahead. The full title should be,
“Always keep enough money on the table so that, with the maximum number of bets you would ever expect for any hand, you never have to go all in.”
This should be very obvious, but I see it violated every day at low limit tables, so here is this essay. I wait long enough between full houses that I want to always get the maximum payoff when I get one.
Calculate how much you need for the particular game you are in. An absolute minimum would be 2 bets for every betting round. That would be 6 big bets for flop games like hold’em and Omaha, and 8 BB for stud.
A better strategy is to actually have the maximum on the table, to cover “capped” betting on every round. [I will use the standard Arizona rule that 4 bets “caps” a betting round. For LA or LV players, sometimes it is 5. Adjust accordingly.] In fact, in hi/lo games like Omaha/8 and Stud hi/lo/8, the betting is often capped on later rounds, and this idea is more important. And I ask you: Why not do this?? If you have paid attention to advice on bankroll, you will have plenty of money to cover this amount. For a 3-6 Stud game, this would be 16 big bets, or $96. So that’s my advice: whenever your table stake falls below $96, re-buy immediately. This costs you nothing, since presumably you have a bankroll of 200 BB or so. Do it. For hold’em, which has one fewer “big bet” round of betting, 12 BB is the proper amount.
For newcomers, your money is as safe on the table as it is in your savings account. I have never heard a single story, ever, of anyone losing any money (to theft or loss) on the table. Go ahead and put as much of your bankroll on the table as necessary.
In Arizona, most games are played with a “kill.” If so, your minimum should be for the maximum number of bets for a kill pot. If the 3-6 stud game has a “full kill” at 6-12, then your minimum should be $192 on the table.
I see this concept violated all the time. One day I joined my friend Pat at a 3-6 hold’em table. As players will do, I asked her, “How are you doing today?” and she told me, “I’m already into this game for $300.” At that moment, she had about $20 sitting in front of her. (I know she always has plenty of bankroll with her – this was not her last $20 or even close to it.) Over the next 5-10 hands, that piddled away, and finally she went all in with about her last $6 or $9, and lost. She re-bought with the table minimum of $30. That piddled away again, and she again re-bought for $30. Finally that glorious hand came, when she had the Slot Gacor winner. She flopped the nut flush and it held up, with lots of betting. She had started with somewhere between $10 and $15. So there was an anemic main pot of about $40-50, which she won, and a great big side pot, well over $100, which someone else won. Students: the size of that side pot is the size of her mistake! She gained back only about $30-35 or so profit, whereas she should have made back $100 or more toward her earlier losses. She didn’t realize how much she had just “lost” (by not winning the whole pot), and continued complaining about how far down she was.
I share this hand pattern. I go for hours sometimes, folding most hands, and folding more hands just after seeing the flop, taking nothing to the river for hours. Then I will hit one or two very big hands and be back to even or ahead for the day. I have to be prepared for that, with plenty of money on the table, so I get full value when I hit a big hand. I believe that this is proper play in a loose low-limit game, because (a) you don’t want to win a lot of borderline small pots anyway, because that way you pay more rake, and (b) the loose players will still give you action even if you are usually tight and/or conservative.
There are certainly times, during particular hands, when it would be advantageous for you if you could be all in. But a priori, before the hand, you don’t know what the situation is going to be. There is also a concept by David Sklansky, where in a stud game, if you could be all in for just the ante, it would be a money winner. The idea is that you would never be forced out of the hand, while others would fold along the way, and you would have better than a 1/7 or 1/8 chance to win the main pot of antes, thereby coming out ahead. This is true, but there is no way to achieve it in practice.
One other related concept is: should you keep counting your money? Glen Campbell says no, of course –
Don’t count your money
at the table,
There’ll be time for counting
when the day is done.
But I think you should. This minimum is one reason. The other reason, for me, is that I like to know what the gain or loss was for a particular hand. Especially if I am going to post the hand on the Internet, my count of the profit or loss serves as a “sanity check” on my recollection of the bets and raises and of how many opponents were in the hand.
The final word is: Go ahead and put on the table all of the money that could possibly be bet in the largest (kill) pot, and keep that minimum on the table at all times. It may take 500 hours of play before you put all of this into one pot, but then you will definitely thank me for this advice.