Rhythm in the madness II: the $55's

[Irieguy]

Well, I recently moved up to the $55's at PP and just finished 200. That's not enough yet to know really where my ITM is goint to land at this level, but I do have some ideas about how beatable this level is and what significant differences there are relative to the lower limits. So, in case anybody is interested, here's what I think:

First off, the skill level jump is very small from the $33's to the $55's, but the differences caused by the larger starting chip counts is very significant. There are still plenty of players willing to call all-in in the first few levels, so optimal strategy should be able to keep you below 5% each for finishes 10th - 8th. This, I think, is where a large portion of a skilled player's ITM will come from. Again, it's just zero-sum game theory here: if other players are willing to go broke early, they will finish in the lower places more often and you will finish in the higher places more often.

But the game really changes at around level 3/4. At the lower limits, people really get impatient with their short stacks and move in too much. That keeps the games moving quickly and makes it relatively easy to keep your 5th-7th place finishes below 8%. But with more chips, that doesn't seem to happen as much. The bubble occurs at level 7 on average, instead of level 5, so even though you start with more chips, you tend to be a shorter stack relative to the blinds when the you get down to 5 players or so. This means that optimal strategy is going to be closer to "push/fold" than it is at the lower limits for bubble play and I think it is much more difficult to exert a significant advantage with skill during this portion of the SNG. My percentages for these places are all around 9%, and while i'm sure it's possible to do better, I'm not very sure it's possible to do a whole lot better.

But once you get down to 3 players, you still have a majority of your play in levels 7 and 8, and it still maxes out at level 9 (just like the smaller limits). But with more chips in play, the shorthanded matches are a little more interesting than they are at the smaller limits. I think that skilled shorthanded play can really result in a large differential between 3rd place and 2nd place, and it should be possible to gain some value here over the less skilled shorthanded players.

Because of the above factors, I don't believe that optimal strategy can acheive the same ITM% at the $55 level as at the $33 level. But I think this is more due to the structure change than a marked increase in opponents' skill level. The good news is that skillful shorthanded play should gain you a higher ROI (for a given ITM%) than would be possible at the lower limits.

In my first "rhythm" post, I suggested that the best you could do at the $11-33 level is an ITM% of around 45% with an ROI of 36% (representing an even distribution of 1st-3rd). I still think that is true. At the $55 limit I don't think you could do any better than around 42% ITM. But whereas an even distribution of 1sts-3rds would result in an ROI of 27%, I think you could do as well as 30-32%.

Just my preliminary thoughts at this level. I know some people claim to do better than this, but until I see it happen over more than 1000 SNgs, I don't believe it.

I will post my results once I have 500 SNGs logged at this level or when I decide to move to the $100's, whichever comes first.

Play the right way,
Irieguy

[LinusKS]

Irie, how many games do you play per day?

How many did you put in at the lower levels?

I stopped playing at Party a while ago, but I tracked a little over 500 games there before I left, and now have almost another 500 at Stars.

Unfortunately, I'm putting in less than ten games a day, so it's going to take a while.

One of the things I like about PS is you start with 1500 chips. I think that tends to increase the skill/luck ratio.

Do you see the extra chips as a benefit at the $55 level?

Are you trying to make a living at playing sngs?

I've often thought if I want to get really serious about poker, I need to learn to play cash limit instead.

I've yet to see the kind of serious thread about making a living at it that you sometimes see from the limit folks.

[Irieguy]

My play varies widely depending on other things, like work. But I've been playing about 50 or more SNGs a week for several months.

I played 500 PLO-8 SNGs and 300 NLHE SNGs at the $33 limit before deciding to move up. (that is, I kept detailed records of that much. I've played thousands of SNGs over the past 3-4 years at various levels from $5-$200)

I think the extra chips are an edge insofar as it seems like you can use them to get a little more play shorthanded and increase your ROI above what you could get at the same ITM% with less chips.

The reason I like PP is because the SNGs don't take as long. Maybe with more chips you could squeeze out a higher ITM%, but I'm mostly concerned with high-volume per unit time play. I want to maximize my hourly rate.

You can make a living at any form of poker in which the rake isn't too high and your opposition is worse than you are. That potential certainly exists in the SNG format.

Irieguy

[eastbay]

[ QUOTE ]
At the $55 limit I don't think you could do any better than around 42% ITM. But whereas an even distribution of 1sts-3rds would result in an ROI of 27%, I think you could do as well as 30-32%.


[/ QUOTE ]

Those are almost verbatim my mean numbers nearly every month.

So, given that, I think a highly skilled player could indeed do better. But by the time they did, they'd be ready for $109.

eastbay

[Irieguy]

Not necessarily. If you are consistently maintaining an ITM% of 42% and an ROI of 30% (for more than 1000 SNGs at the $55 level), I would argue that you are a very skilled player and that you are playing near the limit of optimal SNG strategy.

Irieguy

[eastbay]

[ QUOTE ]
Not necessarily. If you are consistently maintaining an ITM% of 42% and an ROI of 30% (for more than 1000 SNGs at the $55 level), I would argue that you are a very skilled player and that you are playing near the limit of optimal SNG strategy.

Irieguy

[/ QUOTE ]

I have a sample of about 700 SnGs. I'm not sure how to argue one way or the other, other than to say Bozeman, a very astute poster from days gone by, claimed 50% ROI over the long term (many hundreds of games, I forget how many). Considering the quality of his posts and reasoning, I find his claim highly credible.

I played my first hand of hold 'em less than a year ago. It seems hard to believe I've found optimality that quickly. I just don't think it's true.

eastbay

[Lori]

I played my first hand of hold 'em less than a year ago. It seems hard to believe I've found optimality that quickly. I just don't think it's true.

Given the findings of the difference just 1 or 2% makes from your study a couple of days ago, I think people should consider the maximum as some way off where they are currently at, no matter how unlikely that may seem.

Lori

[Irieguy]

I agree that it's not likely you have found optimality that quickly. But I don't think that Bozeman's numbers are achieveable over the long haul. Several hundred would not be enough trials to convince me that he isn't just a very good player that is getting more than his share of breaks for a few hundred SNGs.

Still, I may be underestimating the limits of profitability at this level. I just don't think that I am, and I have never seen anybody maintain results over more than 1000 SNGs at this level anywhere near 50% ROI. Your opponents would have to be giving away a lot of money for that to be possible. They're bad, yes, but aces still get cracked.

Irieguy

[eastbay]

[ QUOTE ]
I agree that it's not likely you have found optimality that quickly. But I don't think that Bozeman's numbers are achieveable over the long haul. Several hundred would not be enough trials to convince me that he isn't just a very good player that is getting more than his share of breaks for a few hundred SNGs.

Still, I may be underestimating the limits of profitability at this level. I just don't think that I am, and I have never seen anybody maintain results over more than 1000 SNGs at this level anywhere near 50% ROI. Your opponents would have to be giving away a lot of money for that to be possible. They're bad, yes, but aces still get cracked.

Irieguy

[/ QUOTE ]

Here's the post where he makes the claim. Considering his grasp of statistics, I doubt he would use the word "long term" lightly. But he does not quote a number.

http://tinyurl.com/5vhfg

eastbay

[Irieguy]

Yes, I remember that post. In fact, it was one of the posts that inspired me to keep better stats.

I am always very curious about what happens to posters who seem to know what they are talking about, and post remarkable results, and then disappear. With the exception of Fossilman, I suspect that their results return to human levels and they lose interest in advertising their now modest success.

Irieguy