Bankroll Requirements

[GrunchCan]

I have always agreed with pzhon that buying short can be the most +EV play, depending on who you are and where you're playing.

But I disagree that buying for 50BB is ever right, if 50BB isn't the minimum buyin. If buying short is the most +EV move for you at a particular time, then there are two overriding reasons why that is so. First, it limits the amplitude of your errors. Second, it reduces the chance that you can make a mistake postflop, since you are going all-in preflop so frequently.

Buying in for 50BB rather than the absolute minumum comprimises both of these considerations. Since you bought in for more, you can lose more when you're wrong, and you'll get all-in preflop infrequently which exposes you to postflop play.

So basically, if it's more +EV for you to play short, then the best ammount to buyin for is the absolute minimum.

In fact, I might go so far as to say that if it's theoretically most +EV to play short, but the minimum buyin is 50BB, then in reality it's most +EV to play deep at a (edit) smaller game. The reason for this is becasue the amplitude of your errors is unchanged in reall dollars, but becasue your stack is so much bigger in relation to the pot, you will not often have to make crying flop calls becasue you were potstuck. Being deeper allows you to dodge losing situations more often.

[ajmargarine]

Interesting thread. I agree with Grunch that 50bb's is not a good buy-in level for the short player. And I stand by my previous assertion that an ~80% buy-in is OK if you have bankroll considerations.

I still don't agree with the shortstack mentality however. I don't believe that is more +EV than buying in full. Yeah, I find it a nuisance when I limp with 44 and the SS PFR's and I have to fold, but so what. I find it easy to play against SS's. You just have to make a few adjustments, something any decent player should be able to do.

At NL200 and below, villians make so many mistakes. The greatest EV comes when you can capitalize to the maximum on their mistakes. If they have a 140bb stack, your greatest EV against them will occur when you have 140 bb's as well and they make a mistake.

I read the Tommy Angelo thing and I see his point. He is not talking about playing the perpetual shortstack. It's the Ed Miller SS thing that I think of when people start talking about buying in short, that IMO isn't good for the avg. 2+2'er.

[pzhon]

[ QUOTE ]
I have always agreed with pzhon that buying short can be the most +EV play, depending on who you are and where you're playing.

[/ QUOTE ]
Thanks. I'd keep saying it even if (or especially if) no one believed me, but it's nice to see some agreement.

[ QUOTE ]

But I disagree that buying for 50BB is ever right, if 50BB isn't the minimum buyin.

[/ QUOTE ]
In that case, you disagree with El Diablo's tactic of buying in for $1000-$1500 in a live $10/$20 game at first, and buying in for $2500 in the UB $25/$50 game. I don't play in these games, so I don't fully understand the context of his decisions, but he has discussed some reasons. IIRC, one point he made was that it is valuable to get more information before buying in deep. Another was that many people don't play their A game when there is a stack disparity. They feel they only need to play carefully against someone who can take their whole stack. In a tough game where an expert has a low win rate, getting your opponent to tilt in this fashion can be valuable relative to a normal advantage. So can the ability to steal the limps a bit more frequently.

[ QUOTE ]
If buying short is the most +EV move for you at a particular time, then there are two overriding reasons why that is so. First, it limits the amplitude of your errors. Second, it reduces the chance that you can make a mistake postflop, since you are going all-in preflop so frequently.

[/ QUOTE ]
There are quite a few other considerations. As zippy pointed out, "Buying in for 50BB instead of 100BB certainly strips away implied odds for large stacks looking to play speculative hands." Look at what happens with the 5%-10% rule: If you raise to 5 BB after someone limps with a speculative hand, they have to call 4 BB, or 4% of their stack if the effective stack size is 100 BB. That's an easy call with a low pocket pair, and a marginal call (or marginal fold) with a suited connector out of position. If you have 50 BB, the call is 8% of the effective stack sizes, which makes it a clear fold with a suited connector, and only a marginal call with a pocket pair.

[ QUOTE ]
you'll get all-in preflop infrequently which exposes you to postflop play.

[/ QUOTE ]
My goal isn't to avoid postflop play. Having a different stack size from what people anticipate when they choose which hands to play can give a short stack an advantage in post-flop play. A short stack can play pair poker (and semibluff aggressively with draws), just as people did successfully in Party's old 50 BB structure.

[ QUOTE ]

In fact, I might go so far as to say that if it's theoretically most +EV to play short, but the minimum buyin is 50BB, then in reality it's most +EV to play deep at a (edit) smaller game. The reason for this is becasue the amplitude of your errors is unchanged in reall dollars,

[/ QUOTE ]
I don't think your errors or your opponents' are the same size in real dollars.

I win many pots by open-raising preflop to 4 BB, getting one caller. I bet 6 BB on the flop, and take down an 8.5-9.5 BB pot. Many opponents call preflop, then fold on the flop so frequently that they have a huge leak somewhere, possibly calling with garbage like QTo or A9o and then folding unless they have top pair. The size of that leak is in proportion with the blinds, not the stacks. I'm getting almost the same value from that leak whether I have 50 BB or 100 BB.

[Isura]

Interesting post pzhon. I am going to try out 4-tabling 1/2 this afternoon with a 40bb stack. I'll post results in a bit, should be a sufficient samplee. [img]/images/graemlins/wink.gif[/img]

[teamdonkey]

Back when Party max buy-in was 50BB, wasn't the generally considered "optimal" win rate for SSNL much lower than it is now? I don't play at Party, but i remember people speculating that maybe 8-10BB/100 was possible now, and other saying there's no way anyone could sustain that, even with bigger buyins.

Shouldn't that answer this question pretty definatively?

[pzhon]

[ QUOTE ]
Back when Party max buy-in was 50BB, wasn't the generally considered "optimal" win rate for SSNL much lower than it is now?

[/ QUOTE ]
I don't know. I doubt it was only half as much in BB/100.

[ QUOTE ]
I don't play at Party, but i remember people speculating that maybe 8-10BB/100 was possible now, and other saying there's no way anyone could sustain that, even with bigger buyins.

Shouldn't that answer this question pretty definatively?

[/ QUOTE ]
There may be some good data there in old threads. As I recall, though, the change to a 100 BB structure happened around the time the highest stakes increased from NL 200 to NL 1000. I don't think the win rates should be compared between the old NL 200 and the new NL 200, as the new NL 200 is much softer. In the Mid-High NL forum before the recent forum split, 5 BB/100 was considered good in NL 600+. I'd like to see comparisons between NL 50 and NL 100 before and after the change.

I don't this will be definitive, though. Buying in for 50 BB when everyone buys in for 50 BB may be very different from buying in for 50 BB when most people buy in for more.

[CarlSpackler]

I've been learning and playing only SSNL for the last month (before this I was primarily playing STT's/MTT's). FWIW, whenever I see someone join my table without buying in full, especially if they only buy in for around half or less, I immediately categorize them as weak-tight/scared money/fish until they show me otherwise. Perhaps this is another advantage of buying in short at a higher level -- that many opponents will erroneously think you're weak-tight/scared money/fish when you buy in short, and thus won't play correctly against you (at least in the short term).