Efficient Sit & Go Play
So I’ve been doing some research on this whole multi-tabling sit&go thing the past few days – some interesting results have occurred. At the 2+2 forums, they have a sit&go section, or what they call “stt’s”, which stands for single table tournaments. Anyways, there is a lot of discussion about how to play certain hands in certain situations, but what I was looking for, was some statistics and strategies on multi-tabling.
I found it here and if you have any interest whatsoever in multi-tabling tournaments, you should definitely give it a read. Apparently, multi-tabling sit&go’s is quite common amongst this crowd. Basically, the poster took a ton of data from a bunch of different multi-tablers in an attempt to quantify some truths about sit&go’s. Certainly, a heroic task to take on, but he did a pretty good job. He attempted to answer a list of very important questions, questions that seem to have no obvious answer, but by the use of empirical data over a large sample size, seems to have come up with a relatively reasonable evaluation for sit&go results.
The questions he asked:
Clearly, the first two questions are the only ones that have anything close to a definite answer, simply due to the fact that their answer is based purely on statistics and mathematics, with the others being mostly personally derived. So I’ll discuss his first two questions and then give my opinions on the others.
He claims that over his (large) sample size, the highest long term expected is 44.4%. Looking at my statistics, I’m sitting right around there with an average of 45.23%. Clearly, my sample size of a few hundred is no where near the infinite amount needed to define long term success, but it’s good to know I’m on the right track. He mentions that he knows a lot of people over 44.4% but eludes to the fact that variance and long term play will smooth out their percentage to this so called mathematical ceiling. Quite interesting indeed, that there is a theoretical breaking point of how often you can finish in the money. On a mathematical level, it makes sense, no one of any skill level can avoid the luck factor, so there has to be some number that is the maximum. How he derived this exact number, I’m not sure, and I suspect that despite his best efforts, he is probably not 100% accurate on that, but it’s in the right ballpark for sure.
Similarly to the ITM question, the ROI question brings up the same sort of mathematical dilemma. If you accept the fact that there is a statistical maximum of ITM%, then a statistical maximum for ROI must be a possibility as well. They are not one and the same though. Right now I could alter my game to maximize my ITM% but it would actually hurt my ROI, since I would be giving up on too many +EV situations where I would be limping into third, rather than gaining more chips for a first or second finish. But they certainly are related. He divides the expected ROI for a “great” player into the different levels of buy-ins. What I assume he means by “great” player is top 10% of all players. His results are listed below.
A great $6/$11 player can probably sustain 30% or so.
A great $22/$33 player can probably sustain 22% or so.
A great $55 player can probably sustain 18% or so.
A great $109 player can probably sustain 12% or so.
A great $215 player can probably sustain 6% or so.
So again, I’m slightly above his predictions with about a 24% ROI, but again, my sample size is somewhat insufficient to provide any real evidence of long term trends. What is interesting is the drop-off in ROI as you move up in levels. Without looking at the data, I’m not sure I would make the same assumption, or at least, I wouldn’t assume that the drop-off would be so dramatic. At the moment, I’m in a transitional period, trying to make the move from $33 to $55. If his data is accurate and I should expect a lowered ROI when I reach $55 outright, I will still be increasing my total real money gained from $7.26/game to $9.90/game. Add in the increased FPP points and increased skill and you have yourself a winning option, despite lowered ROI. These numbers are reassuring to say the least. I’m glad I’ve stumbled my way into the winning ROI levels without ever really doing a lot of research or tinkering of my game. Of course I’ve done a lot of research and studying about playing poker in general, but none to actually maximize my ROI on multi-tabling. Obviously I need to start doing that, since there is always room for improvement, but according to this poster, not much.
The last four questions are all personal preferences. I’ll give my personal short and sweet answers to each one.
When should I move up?
When I was moving up I did it by pure feel and comfort (without the use of poker tracker), so I can’t give any sort of answer that has any quantitative base. However, that comfort that allowed me to move up was essentially based on the fact that I was making enough to justify the upgrade. I honestly wouldn’t worry about ROI all that much when making this decision. If you have the bankroll to support a few lost buy-ins and you are getting a positive ROI, then move up. Why not? Just don’t move out of your comfort zone too fast as you’ll end up playing scared and getting really pissed off when you lose.
How big of a bankroll do I need to play SNGs?
This, I have no clue. The response given by the poster is something ridiculous like 50 buy-ins in case of a terrible swing. But I just don’t see it. I don’t see it possible to loose 50 buy-ins if you have a decent game and a +ROI over a decent sample size. However, this is the point that the poster is trying to hammer home – you aren’t as good as you think you are and you will run unlucky eventually. So if I had to give my personal answer, without his post in mind, I would say that you need about 20 buy-ins if you are 6-tabling. Now I did just pull that number out of my ass, but it seems about right. Of course you need enough just to buy in to your target games, which is 6*buy-in right there. And if you happen to bust out of all of them, you’ll need more. But busting out more than 3 times (18 games total with no money finishes) seems completely unreasonable. It would have to be an extraordinary run of bad luck to lose that much. However, this being said, I assume you are a winning player with a +ROI and multi-tabling experience to attempt this in the first place. Remember no one starts playing 6 tables at $55. I would imagine that pretty much everyone but the already wealthy start playing at the $5 games and work their way up.
Should I quit my job today, or after 500 more SNGs?
Well, I think the existence of this blog answers this question. But what is interesting, is that a lot of people prefer SNGs to cash games for income. Where it would seem that cash games, if you are skilled enough, is where the real money is at. At this point in time, I definitely prefer SNGs, but that is just because I’m a pussy, and can’t stomach the big losses in cash games. Anyways, if you are getting 20%+ ROI on your game, you can theoretically do that all day and make a good profit. The problem becomes stamina and boredom. The poster keeps on the fact that no matter how good you think you are, you are going to lose eventually. Certainly a grim perspective, but its always good to have a reality check, especially from someone who at least has some data to back it up.
How can I tell if I'm really good at this game, or if I'm just a luckbox on a turbo-steamer?
The poster goes into huge detail here asking what reasons you think you are good at poker. He goes a little overboard implying that unless you’ve graduated from MIT with 40 years of poker study, played 5,000,000 hands, you are probably just lucky. Personally, I don’t figure myself to be an amazing poker player. I have way too many weak leaks in my game, and I know I can be way too passive and “foldy”. I can also outtrick myself too often, and end up pushing my stack in on bluffs for no good reason. However, I do know that I am better than the average player. I’ve done a lot of research, I’ve played a lot and I think about poker all the time. I have to be better than the “average” player. I’m not the best, far from it, but I am decent, which if you are honest and analytical will your results; you can come to this conclusion yourself.
So that’s that. A pretty good post by him if you ask me, despite the multiple references to the doomsday scenario. On one hand this post reassured me that at the moment, I’m doing very well. On the other hand, it kept my head level by reminding me that it can all go down in flames, so you must remain sharp and improving. I think it is pretty safe to say that I’ll be sticking with my SNG strategy until I’m either bored, burnt out or go broke, as it seems to be a relatively safe way to make money. Off to the tables I go, wish me luck, it seems I need it.
I found it here and if you have any interest whatsoever in multi-tabling tournaments, you should definitely give it a read. Apparently, multi-tabling sit&go’s is quite common amongst this crowd. Basically, the poster took a ton of data from a bunch of different multi-tablers in an attempt to quantify some truths about sit&go’s. Certainly, a heroic task to take on, but he did a pretty good job. He attempted to answer a list of very important questions, questions that seem to have no obvious answer, but by the use of empirical data over a large sample size, seems to have come up with a relatively reasonable evaluation for sit&go results.
The questions he asked:
- What's the highest ITM (in the money) possible at each level?
- What's the highest ROI possible at each level?
- When should I move up?
- How big of a bankroll do I need to play SNGs?
- Should I quit my job today, or after 500 more SNGs?
- How can I tell if I'm really good at this game, or if I'm just a luckbox on a turbo-steamer?
Clearly, the first two questions are the only ones that have anything close to a definite answer, simply due to the fact that their answer is based purely on statistics and mathematics, with the others being mostly personally derived. So I’ll discuss his first two questions and then give my opinions on the others.
He claims that over his (large) sample size, the highest long term expected is 44.4%. Looking at my statistics, I’m sitting right around there with an average of 45.23%. Clearly, my sample size of a few hundred is no where near the infinite amount needed to define long term success, but it’s good to know I’m on the right track. He mentions that he knows a lot of people over 44.4% but eludes to the fact that variance and long term play will smooth out their percentage to this so called mathematical ceiling. Quite interesting indeed, that there is a theoretical breaking point of how often you can finish in the money. On a mathematical level, it makes sense, no one of any skill level can avoid the luck factor, so there has to be some number that is the maximum. How he derived this exact number, I’m not sure, and I suspect that despite his best efforts, he is probably not 100% accurate on that, but it’s in the right ballpark for sure.
Similarly to the ITM question, the ROI question brings up the same sort of mathematical dilemma. If you accept the fact that there is a statistical maximum of ITM%, then a statistical maximum for ROI must be a possibility as well. They are not one and the same though. Right now I could alter my game to maximize my ITM% but it would actually hurt my ROI, since I would be giving up on too many +EV situations where I would be limping into third, rather than gaining more chips for a first or second finish. But they certainly are related. He divides the expected ROI for a “great” player into the different levels of buy-ins. What I assume he means by “great” player is top 10% of all players. His results are listed below.
A great $6/$11 player can probably sustain 30% or so.
A great $22/$33 player can probably sustain 22% or so.
A great $55 player can probably sustain 18% or so.
A great $109 player can probably sustain 12% or so.
A great $215 player can probably sustain 6% or so.
So again, I’m slightly above his predictions with about a 24% ROI, but again, my sample size is somewhat insufficient to provide any real evidence of long term trends. What is interesting is the drop-off in ROI as you move up in levels. Without looking at the data, I’m not sure I would make the same assumption, or at least, I wouldn’t assume that the drop-off would be so dramatic. At the moment, I’m in a transitional period, trying to make the move from $33 to $55. If his data is accurate and I should expect a lowered ROI when I reach $55 outright, I will still be increasing my total real money gained from $7.26/game to $9.90/game. Add in the increased FPP points and increased skill and you have yourself a winning option, despite lowered ROI. These numbers are reassuring to say the least. I’m glad I’ve stumbled my way into the winning ROI levels without ever really doing a lot of research or tinkering of my game. Of course I’ve done a lot of research and studying about playing poker in general, but none to actually maximize my ROI on multi-tabling. Obviously I need to start doing that, since there is always room for improvement, but according to this poster, not much.
The last four questions are all personal preferences. I’ll give my personal short and sweet answers to each one.
When should I move up?
When I was moving up I did it by pure feel and comfort (without the use of poker tracker), so I can’t give any sort of answer that has any quantitative base. However, that comfort that allowed me to move up was essentially based on the fact that I was making enough to justify the upgrade. I honestly wouldn’t worry about ROI all that much when making this decision. If you have the bankroll to support a few lost buy-ins and you are getting a positive ROI, then move up. Why not? Just don’t move out of your comfort zone too fast as you’ll end up playing scared and getting really pissed off when you lose.
How big of a bankroll do I need to play SNGs?
This, I have no clue. The response given by the poster is something ridiculous like 50 buy-ins in case of a terrible swing. But I just don’t see it. I don’t see it possible to loose 50 buy-ins if you have a decent game and a +ROI over a decent sample size. However, this is the point that the poster is trying to hammer home – you aren’t as good as you think you are and you will run unlucky eventually. So if I had to give my personal answer, without his post in mind, I would say that you need about 20 buy-ins if you are 6-tabling. Now I did just pull that number out of my ass, but it seems about right. Of course you need enough just to buy in to your target games, which is 6*buy-in right there. And if you happen to bust out of all of them, you’ll need more. But busting out more than 3 times (18 games total with no money finishes) seems completely unreasonable. It would have to be an extraordinary run of bad luck to lose that much. However, this being said, I assume you are a winning player with a +ROI and multi-tabling experience to attempt this in the first place. Remember no one starts playing 6 tables at $55. I would imagine that pretty much everyone but the already wealthy start playing at the $5 games and work their way up.
Should I quit my job today, or after 500 more SNGs?
Well, I think the existence of this blog answers this question. But what is interesting, is that a lot of people prefer SNGs to cash games for income. Where it would seem that cash games, if you are skilled enough, is where the real money is at. At this point in time, I definitely prefer SNGs, but that is just because I’m a pussy, and can’t stomach the big losses in cash games. Anyways, if you are getting 20%+ ROI on your game, you can theoretically do that all day and make a good profit. The problem becomes stamina and boredom. The poster keeps on the fact that no matter how good you think you are, you are going to lose eventually. Certainly a grim perspective, but its always good to have a reality check, especially from someone who at least has some data to back it up.
How can I tell if I'm really good at this game, or if I'm just a luckbox on a turbo-steamer?
The poster goes into huge detail here asking what reasons you think you are good at poker. He goes a little overboard implying that unless you’ve graduated from MIT with 40 years of poker study, played 5,000,000 hands, you are probably just lucky. Personally, I don’t figure myself to be an amazing poker player. I have way too many weak leaks in my game, and I know I can be way too passive and “foldy”. I can also outtrick myself too often, and end up pushing my stack in on bluffs for no good reason. However, I do know that I am better than the average player. I’ve done a lot of research, I’ve played a lot and I think about poker all the time. I have to be better than the “average” player. I’m not the best, far from it, but I am decent, which if you are honest and analytical will your results; you can come to this conclusion yourself.
So that’s that. A pretty good post by him if you ask me, despite the multiple references to the doomsday scenario. On one hand this post reassured me that at the moment, I’m doing very well. On the other hand, it kept my head level by reminding me that it can all go down in flames, so you must remain sharp and improving. I think it is pretty safe to say that I’ll be sticking with my SNG strategy until I’m either bored, burnt out or go broke, as it seems to be a relatively safe way to make money. Off to the tables I go, wish me luck, it seems I need it.
posted by RikkiDee at 5:00 PM
3 Comments:
Excellent post, it helped me a lot
Very Interesting!
Just some insight to your stats. I had over 10000 $215 sng logs between 2004-1006 and had a 47.1 % cashout rate. Admittantly, I never tracked actual roi% vs my bankroll, but I do remember cashing out the max every day ($500) and always having between 25k-50k+ in my account. Ahhh, the good old days now we are barred in the USA :(
Post a Comment
<< Home