How important are the numbers

[James Boston]

I've been re-reading TOP recently. Many of Sklansky's hand examinations are very mathematically oriented. I can't do this kind of math on the spot. I doubt I am the only person with this problem. So this brings me to my question. Are the concepts more important or is getting the math exactly right more important? Will just knowing the concepts and estimating the math hurt me in the long run?

P.S. I'm not talking about pot odds and things like that. I'm more referencing his game theory of bluffing formulas. Some of those things just involve too much math for me to do on the spot.

[George Rice]

You could win a lot of money playing low and middle limits even if you never heard of game theory. As you gain experience you will have a sense of when bluffs are profitable and when to call trying to snatch a bluff. Those more advanced will do better, but you will do fine too, especially if you practice good game selection.

[stripsqueez]

i enjoyed reading the theory of poker, but since posting here it has become apparent to me that Sklansky is in truth a frustrated mathematician

you definately dont have to know with precision the sort of maths discussed in the theory of poker whilst playing - you correctly suggest that what is of assisstance is knowing the concepts

bluffing is very much an art not a science - you need to experiment with it and think about it - mainstream theory about bluffing, such as is contained in the theory of poker, is of limited use until you have some experience upon which to base a decent understanding of the theory

stripsqueez - chickenhawk

[SlipFits]

I am in the same boat as you, I'm an idiot when it comes to the number crunching, but with some practice you can learn some of the basics.

[bigpooch]

Definitely the most important aspects of playing poker
with any great success are:

1. Game selection

2. Theoretical and practical understanding of the specific
game being played

3. Understanding how opponents play (which would include
psychology and reading hands)

Clearly, the mathematically and statistically inclined will
have an advantage when examining any of these areas. Even
in the third aspect, it is often important to know the
distribution of possible hands your opponent has so that
you may be able to play your hand well.

Of course, if your opponents are quite hopeless (or nearly
so) a successful (or winning) player may also be quite
incompetent if playing among experts! With the invasion of
the great number of newbies that play online, I suspect that
there aren't really that many skills that are required at
present, especially at the most heavily trafficked sites.

You only really need the more complicated defences such as
game theory against competent opponents. But it is still
useful to know the correct technical decisions as a baseline
for any common problems that are often encountered at the
tables.

An important consideration that is often overlooked is
approximating the loss rate of the opponents that you
commonly face. By doing so, you may have an excellent idea
of how good a game really is!

[LetsRock]

The more you understand and the more "precise" you can be, the better off you are (obviously).

I'm no master, but try to understand the concepts and apply them in a general-loose sense. Even if you could memorize all the equations and have immediate precise numbers (like bluffing%) to plug in, it would be difficult to be able to complete such calculations mentally at a live game and downright impossible to do on-line (cuz you just don't have time).

But you can take basic generalizations into acount when making decisions (have I seen this guy bluff?, is the pot big enough to warrant this action?, etc)

As far as pot odds is concerned, I taken the trouble to build my own chart to indicate how many SB/BB need to be in the pot for each of the number of outs on the turn and river so I can quickly determiine if the pot is at (or at least near the proper size) to make an action correct or not.

Of course this works well on-line, but wouldn't be something I'd do live. It has been my experience that practical use of something leads to memorization of key borderline spots and I find that I rarely consult the chart.

[morgant]

lets rock, would you mind posting that chart up here? i think it would help a bunch........if not cool.
thnks
-morg

[Louie Landale]

Anybody who says they do bluffing frequency calculations quickly and accurately just about every river is lying to you. OK, maybe not "anybody"; but I doubt there's more than a handful that can and do.

The "perfect" strategies are designed to eliminate any mistakes you might make. A side affect is that it will eliminate any mistakes your reasonable opponents might make. So long as you are playing reasonably, it won't matter what you do if the opponent is playing perfectly. Only if you are a maniac or hopelessly weak-tight will you start to lose money.

The difference between a "perfect" stategy and an "almost perfect" strategy is only pennies per session. In a 10-bet-pot, the difference between estimating that the opponent will bluff one-time-in-11 and one-time-in-9 is also pennies (actually about 2/100ths of a bet).

No. Understand the principles and SWAG the math. That's plenty good and its also pretty darn realistic.

Having said that, I highly recommend that you DO work out interesting hands after most sessions. Then when you see a similar situation you can be confident you got the math close enough. It also gives closure to the session so you can deal on an emotional level with the Lady friend, but I digress...

- Louie

[LetsRock]

I would happily supply the chart, but I'm not 100% sure that it's 100% correct and would hate to pass along bad information.

It's not that hard to do yourself. I just converted the draw X% fro each out count from a published list into X:1 and then multipield X * SB for the turn and X * BB for the river.

I can just glance down and know that I need to see 8BB in the pot to correctly call with a flush draw on the river, for example.

(Please don't pounce - I know that pot-odds is just part of the equation that other factors contribute to the overall effectiveness of using mathematics to determine correct play.)

[James Boston]

I use what something called a bet multiplier. I don't remember where I found it. Memorize the odds of making various hands in terms of X:1. Then add X+1=bet multiplier. EX. flush draw is 4:1, then the bet multiplier would be 5. Take the amount you have to call and multiply it times the bet multiplier. If that amount or more is in the pot then you can call, if not fold. So for the flush example if it's $5 to you, you would need at least $25 in the pot ($5 X 5 bet multiplier = $25).

[jedi]

I heard it explained somewhere that you only need a general sense of the numbers. If you estimate a decision one way, and you're close but wrong (getting slightly the worst of it), then it's not too much of a wrong decision. The obvious decisions should be easy to make.

[BruceZ]

Here's my chart (http://forumserver.twoplustwo.com/showthreaded.php?Cat=&Board=probability&Number=381 698&Forum=probability&Words=pot%20odds&Match=Entir e%20Phrase&Searchpage=0&Limit=25&Old=3months&Main= 380910&Search=true#Post381698) for pot odds. The pot is counted in bets, never in dollars. There is no arithmetic done during the hand for pot odds. The size of the pot is simply compared to the number from the table correponding to the number of outs. This way you know instantly if you have the right pot odds, and you can spend your thinking time on other considerations.

[Tommy Angelo]

"In a 10-bet-pot, the difference between estimating that the opponent will bluff one-time-in-11 and one-time-in-9 is also pennies (actually about 2/100ths of a bet)."

At the moment of truth, he is either bluffing, or he isn't. My goal is to have the right answer, and react properly to it, 100% of the time. That would be true if the guy bluffed every other hand, or once in a thousand.


Tommy

[Ed Miller]

In typical pots, you won't have to actually crunch the numbers to make the correct play most of the time. If I am pretty sure I am beaten but am considering a crying call, I try to come up with a set of hands that my opponent might play the way he played that I can beat. If my set is reasonably large, like, say, "busted flush draw" (of which there are often several possible combinations) then the pot is usually big enough to toss the money in. On top of the hands that you can logically put your opponent on that you can beat, there is almost always a 2-4% "totally wacky and illogical play" fudge factor that I throw in. That is, I expect to see a bizarre and unexpected hand maybe 2-4% of the time from a typical player. So if I add my fudge factor to a reasonable set of hands that I can beat, I estimate my chances of winning at usually 10% or higher and throw the money in (assuming the pot is of typical size or larger).

If I can't come up with any reasonable hands that I can beat... or maybe one specific hand, like A /images/graemlins/spade.gif4 /images/graemlins/spade.gif, then sometimes I'll throw my hand away for one bet on the river.

Conversely, if you have a hopeless hand, it is often correct to go ahead and bet as a desperation bluff on the river if you can come up with a reasonable set of hands that your opponent could have that he might throw away for one bet on the river (again, something like "busted flush draw"). Here, the fudge factor can be typically much higher than 2-4% (assuming that you have been the aggressor so far throughout the hand) because opponents tend to call with some absurd and bizarre (and therefore, weak) holdings that you would never even consider.

In other words, if you can say to yourself objectively in two seconds, "I can imagine such-and-such reasonable scenario where this one bet might win me the pot on the river, either by calling or betting" then it's usually correct to risk that bet because the pots are so big.

Having said that, I think that Sklansky's analysis of the math behind playing when all the cards are out in TOP is terrific, and you really should do your best to understand his thought process fully.

[BruceZ]

As for this issue of the mathematics of bluffing, check raising, etc. in TOP, keep in mind that this is the THEORY of poker. In a theoretical work, one writes down the theoretically optimal solution or method, as closely as possible, without regard to whether this method is actually practical. Then once this optimal method is put forth, one can find ways to make whatever approximations and estimations that are needed to realize the theory as far as is practical. TOP mentions ways to do this to some extent, but its main emphasis is the pure theory, while other books provide practical advice that applies elements of the theory to particular games.

As an analogy to understand this, consider the game of blackjack. Before we consider any card counting systems, we can determine the optimal strategy which takes into account every card which has been seen. Such a strategy can only be played by computer, and it will perform better than any human can perform. The practical card counting systems that humans play are approximations to this theoretical optimum. We need the computer perfect strategy to provide a reference by which all other card counting systems can be compared, so we know how close they are to optimal. This is the modus operandi throughout engineering, and algorithm development in particular.

[Wake up CALL]

[ QUOTE ]
"In a 10-bet-pot, the difference between estimating that the opponent will bluff one-time-in-11 and one-time-in-9 is also pennies (actually about 2/100ths of a bet)."

At the moment of truth, he is either bluffing, or he isn't. My goal is to have the right answer, and react properly to it, 100% of the time. That would be true if the guy bluffed every other hand, or once in a thousand.


Tommy




[/ QUOTE ]

Tommy I find it odd that you have never called anyone who was "bluffing with the best hand".

[Louie Landale]

But if you are so good that you can tell WHEN he's bluffing and when he's not, the numbers don't matter at all.

- Louie

[Tommy Angelo]

"But if you are so good that you can tell WHEN he's bluffing and when he's not"

Hi Louie. All I did was say a goal, not that it had been attained, or that it could be attained for that matter. (I think we'd agree that it can't.

I did mean this though:

"the numbers don't matter at all."




Tommy

[SevenStuda]

Yeah, maybe when your "running good", "no Taco Bell tonight."

[bigpooch]

About the bluffing frequencies, etc.: they shouldn't be done
at the table on the spot. You simply do the homework for
the most common situations so that when that case comes up
at the table, you are already prepared. Also, for a game
such as draw (as opposed to LHE), game theoretical bluffs
will come up often enough to make it worthwhile to do this
homework. The same could be said of games like PL Omaha and
lowball draw: the better prepared player will be able to
make bluffs rather mechanically as they are systemic.

[RydenStoompala]

[ QUOTE ]
Are the concepts more important or is getting the math exactly right more important?

[/ QUOTE ]
The concepts are important. Getting the math right is an intellectual exercise that takes a back seat to card sense, feel, position, experience and general conceptual knowledge. Knowing your weaknesses is more important than knowing the equity value of one play versus another.