Ideal Progression Model for moving up Cash Game stakes?
Posted by Pran
Posted by
Pran
posted in
Low Stakes
Ideal Progression Model for moving up Cash Game stakes?
I want to gradually move up in stakes starting from the bottom. Here is my general plan:
- Win 50 buyins at $2 NL
- Take a 10 Buy in shot at $5 NL
- Win 50 buyins at 5 NL
4 Take a 10 Buy in shot at 10 NL - Win 50 buyins at 10 NL
Is this a sound strategy for moving up? I honestly just don't have the first clue as to how I should move up in stakes.
Loading 22 Comments...
Moving up is rarely a linear process. You may need to move back down at some point if things aren't going well.
In general, as long as you keep sufficient bankroll, you can move up and down as much as you want.
Usually anywhere from 40-100 buyins is enough, depending on your risk tolerance and your winrate. Play around with the numbers using the Poker Variance calculator to see what's right for you.
Seems reasonable really. Can consider using a 40 buy in threshold but 50 is perfectly fine also.
removed :)
Another aspect as it pertains to moving up stakes and BR management that I saw recently (I forgot where) was if you have a stable income from a job it's easier to shot take knowing you have income on its way. I think anywhere from 20-50 buy ins is reasonable. Would love to hear thoughts on this strategy.
If you have job income that's very reasonable and can go a little more aggressive when you know will have an edge on the limit above. Like shottake at 20bi for the level and move down at 15bi. It really depends on where you are with your game, if you are playing a level that's way below your skill level for example you can be more aggressive
It depends on your winrate and variance. However, for an average variance (about 90bb/100) and a 3bb/winrate you should move up stakes as soon as you have around 19BI for the next stake. yes, that sounds much more aggressive than what you hear everywhere but it's what the math says. 50BI is way way too much unless your winrate is barely above breakeven, you are expected to move up at 40% the pace of a 19BI strategy
I think you're talking about std. deviation when you say 90bb/100.
But that aside, what's the math behind moving up with 3bb/100 winrate and 19 buyins? The minimum bankroll for 5% risk of ruin is actually 40 buyins assuming these parameters.
I think a 19 BI approach may work if you're very disciplined and move back down again as soon as your bankroll goes down. Since you cut the stakes often enough, your "effective" bankroll is higher.
I'm not sure most people are disciplined enough to handle these kind of swings. I remember when I was forced to move back down during one of my downswings, it was quite tough to play the lower stakes.
So you cant reply to replies or its just my phone? Anyways, I moved up from 1 to 100NL employing an aggressive bankroll management and, yes I had to move down stakes countless times, and it's super easy. If you don't have the discipline to move down stakes when required then you really need to work on your mental side since it's so important to be emotionally stable and disciplined if you want to have long-term success in poker
As for the math, at 90bb/100 and 3bbwrt we can consider 100 hands of poker a game where we win 93bb half of the time and we lose 87bb the other half. Then we use a kelly calculator. It wil return that we need about 26BI (I forgot already) but since stakes are discrete we want to stick to playing between 19-38BI, that mean we move up when we have 19BI. I made a python script to calculate all that stuff too. And you can't use risk of ruin because, yes, you can move down stakes at any time so it's like you have more "lives"
Would say it depends heavily on how many tables your playing:
If it is all the bankroll you have (no side income)
1 tabling - 20 buy-ins is sufficient and you can take shots at higher tables having 15 buy-ins
4 tabling - 80 buy-ins is sufficient and you can take shots at higher tables 4 at a time with 60 buy-ins.
I would not be in a hurry if your win-rate is any decent you will be able to move up quickly.
Lets do the math if you have a very decent win rate 10 BB per 100 hands and are multi tabling (4 tables with each 100 hands per hour is 400 hands per hour) so you win rate per hour is 40 BB. So 5 hours of play is 200 BB is 1 buy-in for the next level. If you have good bankroll management you will already have 80 buy-ins of the current level is 40 buy-ins of the next one. Meaning that you only need 40 buy-ins more is 40x5 is 200 hours of play to move up. So in lets say 2 months you can move up a lvl. Now starting from NL1 and moving up every 2 months is 6 times in one year. You will end up playing 1x2x2x2x2x2x2 at NL64 lets say 50NL and after another year 50x64 = 3000NL. If you then still win 10 BB/100 you will win 40 BB per hour which is 1200$ per hour.
Number of tables doesn't affect bankroll required.
[Except insofar as you'll typically have a higher winrate if you play less tables.]
Well that cant be entirely true because if you play 20 tables that would mean that you have you play with your whole bankroll at once which isn't a good idea in my opinion.
When you multi table you in fact have more money at the table at once so if you tilt you will loose more money at once.
Maybe you don't need to take it into consideration as aggressive as I did but one needs to consider it.
At least that is my opinion :)
HAHAHAH I feel like we need a gigantic CITRUS PARIBUS at the end of that statement. But that's some good number crunching thanks mate :). Also the major reason why I have a progression model in the first place is that the skill level seems to increase at different levels of stakes.
Have enough BR to afford swings >50BB, and to not feel money pressured (money scared). So something around 40-50BB is enough for micro stakes.
However if sometimes you might have to move down and it shouldn't affect your mindset so badly.
So I can run the numbers but the basic idea is to do Kelly betting. But it is normal to do fractional Kelly as that betting slightly over Kelly is disastrous but under doesn’t hurt that badly and accounts for inaccurate estimation of win rates. The only thing is for true Kelly the instance your bank roll breaches a threshold you need to move down. It is true the number of tables does not make any difference provided you instantly move down.
It is also better from a mathematical point of view than take an x BI shot. Psychologically it may not be optimal. And there may be some discretisation effects that need to be accounted for. Eg there are not a continuous variety of stakes.
There is also the learning effect to take in to account. If one would hop up and down levels continuously the learning effect on how they play on a certain level will go away. The Kelly effect does not take that into account either. Its is not simply betting :)
It’s not and you can make adjustments, but I would start from the good baseline approximation and make appropriate refinements of the approximation from there.
Doesn't Kelly betting assume the same win rate throughout the whole betting. So another thing to recon with and there will be many others we forgot.
I would just have more trust in general wisdom in this situation then in some algorithm. Of course one can/should validate if the general wisdom is not to go to up to quickly according the Kelly principle but that is about it I think.
I guess my feeling is start with the model. Any bank roll management technique assumes a win rate. Hiding it under conventional wisdom just means that the assumptions are less clear. That’s why I recommended fractional Kelly. It is fairly robust to misestimations of win rate. But you should certainly see what values you get for different win rates. That is informative as well. You can also do the buffer trick. Go a bit past capital required for the next level before moving up. You end up in a place that’s not wildly different from what is bandied about but with a clearer idea of the assumptions and the principles behind what you are doing and how you have adjusted them for your situation. It also lets you understand why the answer to the question do I need 2x the roll if I am 2 tabling is no.
It does require iron discipline about moving down though and makes it clear why that iron discipline is needed. I can write a longer post on it at some point if people are interested.
It can also be adapted to changing win rates.
The other thing to take into account is that although an RIO'er is going to take an appropriate progression, work on their game, proper BRM, etc - There are guys sitting 50nl and 100nl who are as bad as the 2nl'ers but just have enough money in the bank that they couldn't conceive of playing with less than a $50 or $100 BI.
Be the first to add a comment