In recent times, a new play has popped up quite a bit in poker, particularly in high rollers. This involves going all in on the river, minus one big blind. A lot of players when they see this bet assume it’s always value, and indeed it nearly always is in lower buyin events, where players recognize or at least fear that unsophisticated players overfold to all-in bets (thinking nobody would ever put all their chips in and risk their tournament life on a bluff), so they use an almost all-in sizing they believe gets called a lot more.

Never a bluff?

High rollers are different of course. One way they’re different is the elite players that play them will have bluffs in every spot, including this “no one ever does it as a bluff” one. A balanced player is unexploitable, and one of the ways in which they achieve this is by having the correct amount of bluffs in every spot. If they bet the river for, say, pot, then they are giving their opponent two to one on the call. That means that to be balanced they should “have it” (be value betting) two thirds of the time and be bluffing the other one third of the time. That’s a function purely of how big the bet is in relation to the pot, not in relation to the stacks. It doesn’t matter whether the pot sized bet is all-in, or for any other percentage of the effective stack: the opponent should be calling it one third of the time with their bluff catchers to avoid being exploited. And when I say “any percentage” I mean any percentage, from 1 or less to 99 or more. If the pot sized bet is for 99 of the effective stack of 100 big blinds, so only one blind is left behind, it should still be a bluff exactly one third of the time.

the nut low is often one of the best bluffs we have

What makes the spot a little funky is whenever we are bluffing, then we might get rebluffed. So what do we do in this spot if we bet 99 of our big blinds and our opponent sticks us in for our last blind? The answer, as always, is pot odds. If we bet 99 big blinds into a 99 big blind pot and our opponent effectively raises to 100 big blinds, there are now 298 blinds in the pot and we have to call only one, so we are getting 298/1 on the call. If we believe we are good more than one time in every 299 (or just over .34% of the time) then we call, but if we believe we are not, we can actually fold. You might think we can never be 99.67% sure we aren’t being bluffed, and you’re probably right, unless our opponent accidentally exposes their hand (I’ve seen it happen: believing the nearly all in player to be all in they push all their chips into the pot then table their hand triumphantly), or….if you bluffed with the nut low. And remember: the nut low is often one of the best bluffs we have because it has the least showdown.

ICM considerations

There are other reasons players make this play, whether they intend to call or fold to a shove. One is on a bubble or a payjump: say you have a hand you have to go all in with, or call an all in with near a bubble or a pay jump: well, if you move all in and get called and lose, you can’t get the pay jump. But if you move all in minus one big blind, and get shoved on, you can wait there until someone calls clock on you before calling, during which time you might get the pay jump. Or if you’re intending to fold because you are bluffing with the nut low, well, now you have until the blinds hit you to get that pay jump.

Remember how tournaments and ICM work. The bottom of your stack is worth more than the top (except in knockout tournaments). Every big blind you add to your stack is worth slightly less than the one beneath it, which in turn is worth slightly less than one beneath it, and so on. Your last big blind is your most valuable by far. This is true in all non-knockout tournaments, but in live tournaments, there’s one other factor which massively impacts this: our new friend, the big blind ante.

The big blind ante

These are now almost universal and unique to live tournaments. Online, we still used the old fashioned “every player antes every hand” ante, but live is different. And the big blind ante changes things in a way that might not be obvious at first.

Let’s say you’re under the gun with one big blind. You shove. If you win the pot, how much do you win? Well, you get the big blind ante.

The big blind ( whether the big blind calls or not: if someone else raises and they fold, you still win that big blind).

The small blind (if they fold: if they call you get another half a big blind).

One big blind for every player not in the blinds who puts money in preflop.

you’re getting 2.5 to 1 on your money against their random hand

Worst case scenario: everyone folds to the big blind, and you stand to win 2.5 bbs, so you’re getting 2.5 to 1 on your money against their random hand, so you need to win only 28%. That’s a good spot no matter what your hand is: even 32o has x% against any two cards.

Even if someone wakes up with a hand, you’re in a good spot. Say someone raises, and everyone including the blinds fold, now you’re getting 3.5/1, and only need to win 22% of the time. The more players put money in, the better your odds get. If everyone calls, you only need to win 10%. Sometimes more than one player calls preflop, but then someone bets on a later street causing others to fold away equity: even better for you.

Is two better than one?

Now consider the situation where you have two big blinds instead of one. What happens if you shove now? Once in a blue moon everyone folds, but this is such a rarety we shouldn’t really give it more consideration. The two most likely outcomes are:

(1) Someone calls or raises, and everyone else folds. In this case we win two big blinds from them, and two and a half in blinds and antes for a total of 4.5. So we are getting just over 2/1 on our money, which is nice, but nowhere near as nice as the 3.5/1 we get if we only had one big blind. Another way to look at it is in this scenario we get 3.5/1 on our first big blind, but only evens on our second: the amount we win doesn’t double, it just increases by one big blind from three and a half to four and a half.

our second blind nets a much lower return than our first

(2) It folds to the big blind and they call. Now we win only 3.5 big blinds, which again is only one more than the 2.5 big blinds we would win if we shoved for only one big blind. Again our second blind nets a much lower return than our first.

Less is more

From this, you should be getting the idea: one big blind is a much more profitable stack size than two. That’s why you’re seeing the high rollers leave exactly one big blind behind, rather than two or any other number. Believe it or not, if it were possible to leave less than a big blind behind, this would be even better, but most tournaments don’t allow this any more, in line with a recent change to TDA rules that prohibit players leave less than a big blind behind. I believe the reason for this rule change (which essentially changes no limit Holdem to “almost no limit Holdem”) was to stop players leaving the smallest chip behind then stalling. But if it were possible to leave, say, a tenth of a big blind behind, then this is an even better situation, purely because of the big blind ante. Under the old fashioned ante still used online, a player can only win the amount they shove from every ante so if it’s less than an ante they can’t win all the antes, but with a big blind ante, you can always win the full ante. So if you shove for, say, one tenth of the ante (and big blind), you can win, at the very least, 2.5 big blinds, so you’re getting 25/1 on your money!

The micro stack limp

From all of this, you should now get the idea: one big blind is much more profitable than two. So much so that if you do find yourself with two, I suggest you limp! In some worlds, the hand gets checked down, you lose, but are left with one big blind.

In my next piece, I’ll talk about a similar but related play I saw Jen Shahade make at European Poker Tour Barcelona: going almost all in before the flop, leaving one big blind behind. This play has the same advantages as the river not quite all in I discussed in this article, and a few additional ones I’ll explain in my next article.