Dr. Sun's Poker Lab

[SunPowerGuru]

@WannabeCoder

You could as easily comment on the relative rarity of being dealt 22-44 compared to TT+. It’s far likely to be a quirk of the data than anything meaningful.

I wasn’t tracking how many times I was dealt a specific pair. I just saw flops less often with smaller pairs. This seems reasonable to me.

33 in particular started out something like 4 of 14, which is probably a fluke. It seemed to be evening out, but never fully recovered.

Yes, the sample sizes for each hand value are small. For that matter, 1,000 instances total isn’t as large a sample as I would like. It is, however, as much as I was willing to do. If soneone wants to go to 5,000 instances of each pair, be my guest.

Anyway, within the limitations of what I was doing, the results are good enough to satisfy my curiosity. I think at least that aspect of the pRNG is legit. I provided what data I did collect, draw your own conclusions.

[Comicguy]

@SunPowerGuru

Anyway, within the limitations of what I was doing, the results are good enough to satisfy my curiosity. I think at least that aspect of the pRNG is legit.

Its not surprising that your results came out nearly perfectly aligned with statistical expectations once you told them you were looking into it.

Couldn’t resist

[SunPowerGuru]

Haha, I actually thought about that too.

I eventually concluded it would have taken them at least 3 or 4 years to change the algorithm in order to make it look fair.

[JanCee]

[feelmysins]

My with/without showdown numbers are pretty close to a perfect 50/50 with only 17 of a difference over 4741 wins. Does anybody else care to share a screenshot??

[BluffaloKing]

Differences are that I fold 74% to your 70% preflop and my showdown stats are 56/44, so apparently you bluff more than I do, or the extra folds mean that I am playing slightly better hands and winning more on the showdown. I also have a tendency to try to play my monster hands down to the river and build the pot as large as possible rather than force opponents to fold at the flop.

Although I have played slightly more pots, we have both won almost exactly the same number of pots, so your percentage is better.

[feelmysins]

With/without showdown numbers differ by 1 over 4883 wins.

[Grateful_ed]

That would drive me crazy. I’d need to go all in until I won a pot without showdown. I guess then I’d have to take up golf…

[waidus]

|Pots Won: |19% (7,601)|
|At showdown: |66% (4,983)|
|Without showdown: |34% (2,618)|

|Total hands played: |40,047|
folded: |59% (23,805)|
|Rank: |42,121|

|Flops seen: |24,866|
|While in big blind: |5,482|
|While in small blind: |5,075|
|While not paying blind: |14,309|

[BigFish1337]

I think I might have a bluffing problem… Please don’t call.

[ValueFish]

@waidus

olded: |59% (23,805)|

Gotta get a bit tighter maybe !

[SunPowerGuru]

Puting a Bad Beat on a Fella

So, the question i have been playing with lately is this: can a -EV play be +EV. The answer seems to be, “Yes, if viewed in the proper time frame.”

Making a bad call and sucking out seems to have an effect well beyond the current hand, at least in many cases. There is a psychological impact, and many players just can’t handle the strain. Now, you might think this only effects less experienced players, but you would be wrong to assume that.

I have been finding that it has less to do with poker experience, and more to do with personality type. I consider human nature to be one of the few things worthy of study, mainly because it has such a widespread and meaningful impact is so many aspects of our daily lives. Understanding someone’s personality type helps us to understand their approach to the game, and this, in turn, allows us to have a better idea of their ranges, what their actions mean, and so on.

I have identified about 6 basic personality types. At least 2 of them are prone to losing it when someone puts a bad beat on them, and they will respond in a predictable and exploitable manner. One other of these base types usually, but not always, responds poorly.

Calling bad and sucking out advertises to the whole table that you are an idiot, and this can result in you winning much, much more those times you aren’t being a moron. Lose a little now to win a lot later.

So yes, -EV plays can sometimes be +EV.

Now let’s see what the accountants think. And please try to answer without using the word “spew”. Thanks!

[puggywug]

Having a lot of experience with this, I can agree with your findings.

Short time frame is especially applicable to tournaments. But tournaments are when you can least afford mistakes. There’s no rebuy, stacks are short, and effective stacks shrink as the blinds increase.

There are many types of -EV plays. The best is the one that you win chips on. It’s always good to win chips, but the other benefits of tilting your opponent through the bad beat, and creating a table image that you are a poor decision maker that can be exploited can work to your favor, just as you say.

The other type of -EV play (where you lose chips) can also create a false (hopefully) table image to be exploited later. It’s never good to lose chips, but the best time to do it is early in the tournament, when you are forming your table image, and when it will be cheapest to take the loss, and you will have the most opportunities to exploit your cultivated table image ahead of you.

The other time it is “good” to make a -EV play is out of desperation. When you make a decision to shove with the best hand you’re likely to see before you bleed out, it’s often a -EV move where most of the time you’re winning minimal pots with blind steals, but losing your stack a lot of the time when someone calls. As much as possible, you want to make this move holding premium cards, but in a pinch you’ll sometimes have to make do with a marginal hand. This is never a situation that you deliberately want to find yourself in, of course, and if you’re here frequently, it indicates that you should look at your strategy and figure out how play better. But when you do shove out of desperation with a small stack, and you manage to win a big pot that way, it can give you a second life at the table, almost like a re-buy.

It’s pretty rare that you can hit draws in tournaments without making calls that are at least slightly -EV. But it’s best to do so when you’re drawing to the nuts, and/or you have a high pair that will give you a bit of showdown value if you miss.

But mostly try to avoid making -EV plays.

[SunPowerGuru]

Well, this only has a very narrow set of specific circumstances where it is worth doing. I don’t think early tournament stages are one of these times. Yeah, it’s cheap, but you won’t have a good read on personalities then, and most of the people at your table won’t be around very long, and most of them won’t understand anyway. There are a few exceptions to this, but only a few.

How often have we heard, “I start with a GTO strategy, then make adjustments”?

If you have to make adjustments, your GT wasn’t all that O to begin with. There’s that. But the real issue is with the scope of these adjustments. All but the best players tend to over-compensate, and this open up huge holes in their games.

My approach is less passive and more proactive. instead of analyzing your style, why can’t i play in such a way that forces you to make adjustments that are not only 100% predictable, but also go in a direction that I find easiest to exploit?

I find this kind of play to be most effective in the mid to late stages of a tourney. This is when I usually have the chips to pull it off, understand more about my opponent’s personality types, and have the best chance to force them off their game and into my evil clutches.

[SunPowerGuru]

Please remember that my poker lab is dedicated to experimental poker. None of this is settled science.

I post my crazy ideas here to gain insights from other points of view.

[puggywug]

Re: GTO + adjust, I assume that means they switch to exploitative plays once they get a good read on their opponents.

When it comes to GTO, I mostly hear about it as an abstract concept, an ideal, and not something that is currently well understood. Very little of what I have read actually describes it in detail, so I don’t really know what it is. I could be behind the curve on this topic, I do hear people talking about it all the time, but see little in the way that would suggest that anyone knows precisely what the GTO strategy is.

[SunPowerGuru]

Right. NLHE hasnt’ been solved yet, except for heads up play. If it was solved, there would be an optimal solution to each of the 23 kazillion possible situations, and nobody would be able to remember them all.

However there are some “GTO” type ideas that can be employed in frequently encountered situations. So really, GTO mostly refers to a more general approach rather than actually claiming anything is situational optimal. Actual optimal is mostly unknown, and possibly unknowable because of the number of variable, especially the problems associated with player modeling.

[puggywug]

Oh, and of course the other type of -EV play – the bluff. Of course a bluff that works often enough is +EV. But based on pot equity every bluff is -EV at showdown.

[JanCee]

Gotta recommend this book to you, Sun!

[SunPowerGuru]

Yeah that and volume 1 are a little dated, but still have good basic info.

About 10 years later, he wrote Harrington on Modern Tournament Poker, which is also pretty good.