Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie
Hi all,
Vanilla are planning an update to the site on April 24th (next Wednesday). It is a major PHP8 update which is expected to boost performance across the site. The site will be down from 7pm and it is expected to take about an hour to complete. We appreciate your patience during the update.
Thanks all.

Bankroll Management or "How many buyins do I need for xxxx?"

  • 17-05-2007 2:40pm
    #1
    Registered Users Posts: 2,004 ✭✭✭


    Right, seems like every week there is some new post about bankroll management and questions like "how many buyins do I need for xxxx?" and the like. I checked the stickies and there doesn't seem to be anything comprehensive up there so I decided I'd write something now because I don't want to study.

    An important thing to realise is that there are no set rules you can follow when deciding on bankroll management. Hopefully here I can point out some of the things that should be taken into consideration though.

    The first thing you need to consider when trying to decide how big your bankroll needs to be in order to be sufficiently bankrolled to play a certain game is your skill level. The ammount of variance which a certain player will encounter when playing poker at a certain level is largely a function of his skill level relative to the other players in the game. Put simply that means that if you crush a certain game then you should, in the long run, be able to play with less variance than if you were playing in a game in which you are a breakeven player or marginal winner. Thus you can get a way with having a smaller bankroll.

    Another factor to consider is the general level of aggression in the game. A very aggressive game will induce a lot of variance for its players because they are being forced to make marginal decisions for large ammounts of money (read: large percentages of their stack) very often because their opponents are putting pressure on them every chance they get. Eventually you need to take a stand against this aggression in spots where you are not 100% sure that you are winning. Sometimes you are right and sometimes you are wrong, the fact is that these types of decisions all contribute to increasing variance and mean that you should try and have a bigger bankroll before tackling these types of games.

    This point also applied to short handed games like 6max and heads-up. The less players in the game the less likely it is that anyone has a premium hand. Therefore being very aggressive can often be a good strategy for winning in these games. Keep in mind though that this will push up your variance and the bankroll needed to play at a certain stake will increase the shorter the game gets.

    Also a note on the aggression paragraph. I feel that it can be difficult for low stakes player to get a good idea of what a truely aggressive game is. I often hear people saying that you will have higher variance if you play against really bad players, suce as those found in the SE or the Fitz. This is a complete fallacy. Bad players are easy to play against, you make a good hand, you bet, they call and you take their money. Yes they will suck out more often then good players, that is true. But that is only because they get their money in behind more often then good players and the suckouts make it appear like a high variance game.

    So the best advice I can give when trying to decide if you are bankrolled for a game is to consider the above factors. As a general guideline, if you are playing in a low aggression live game that you beat for a high rate then you can get away with a smaller bankroll like 15-20 buyins. If you are not so sure of your abilities then try and wait until you have a larger roll before stepping into the game, more like 20-25 buyins. In an agressive high stakes online game then these numbers need to be increased to account for the different game conditions. Personally I won't play 5/10 NLHE online without at least 50 buyins in my bankroll and I'd even consider that marginal enough. Why? Because it's a highly aggressive game in which I am not certain that I'm a winner.

    Summary:

    relative skill level up, bankroll needed down.
    aggression up, bankroll needed up.

    easy games: 15-25 buyins.
    meadium games: 25-35 buyins.
    hard games: >50 buyins.


«1

Comments

  • Registered Users Posts: 1,080 ✭✭✭HiCloy


    Nice post.

    Do you think if you multitable you need a higher bankroll?

    Say 4 1/2 online tables compared to 1. I don't but have heard it argued many times.


  • Registered Users Posts: 38,998 ✭✭✭✭Mellor


    I'd say you need a higher BR to multi table, as you will have a higher % of your BR in play at any single time


  • Closed Accounts Posts: 2,533 ✭✭✭ollyk1


    WP pok3rplaya and long overdue.

    Link to be added to the useful posts thread please mods.


  • Closed Accounts Posts: 1,639 ✭✭✭[nicK]


    Mellor wrote:
    I'd say you need a higher BR to multi table, as you will have a higher % of your BR in play at any single time

    this is generally only the case if your prone to tilting badly..


  • Registered Users Posts: 1,080 ✭✭✭HiCloy


    Mellor wrote:
    I'd say you need a higher BR to multi table, as you will have a higher % of your BR in play at any single time

    Yeah but your results on one table aren't correlated to those on others, unless you tilt really badly. So playing 4 tables for one hour should be the same variance wise as playing 1 table for 4 hours IMO


  • Advertisement
  • Registered Users Posts: 2,004 ✭✭✭pok3rplaya


    HiCloy wrote:
    Nice post.

    Do you think if you multitable you need a higher bankroll?

    Say 4 1/2 online tables compared to 1. I don't but have heard it argued many times.

    Thanks.

    I don't think it matters all that much really. Obviously you cant play 8 tables with a 10 buyin bankroll simply because if you get stacked twice you wont be able to rebuy. I've seen a bunch of maths done somewhere which worked out that you should have a larger roll for multitabling but it was only by a very small ammount. Just because you have a larger ammount of your bankroll in play at once doesn't mean that your chances of losing the whole lot at once go up.


  • Registered Users Posts: 2,563 ✭✭✭sikes


    Nice post.

    People should also consider the games they are playing. If they are playing full ring the buy in reuqirement is less than for 6 max, which in turn is less for HU.


  • Registered Users Posts: 2,004 ✭✭✭pok3rplaya


    sikes wrote:
    Nice post.

    People should also consider the games they are playing. If they are playing full ring the buy in reuqirement is less than for 6 max, which in turn is less for HU.

    Yes good point I forgot to include that. It is kinda covered by the aggression thing though but I soulda made a point out of it. I'll edit.


  • Registered Users Posts: 38,998 ✭✭✭✭Mellor


    pok3rplaya wrote:
    Thanks.

    I don't think it matters all that much really. Obviously you cant play 8 tables with a 10 buyin bankroll simply because if you get stacked twice you wont be able to rebuy. I've seen a bunch of maths done somewhere which worked out that you should have a larger roll for multitabling but it was only by a very small ammount. Just because you have a larger ammount of your bankroll in play at once doesn't mean that your chances of losing the whole lot at once go up.
    I wasn't suggesting that if you play 2 or 4 tables you need twice or four times the BR, but in situations were you have just enough to play 2/4.(say 25 buy-ins) And mutli-tabling 4 tables leaves you with 16, If you got stacked on two of these, if you rebuy, you are jumping in with only 23 buy-ins.
    The ammount you should increase it by if multi tabling is probably very small, and should increase depending on number of tables.
    I'd imagine that the number of tables would be enough.
    if you multi-table 4 tables, then increase the limit of 25 to 29. (if games are not very agressive at a paticular level, it could be reduce to half the no of multi tables, so 4 tables at a passive level, increase 25 to 27)


  • Closed Accounts Posts: 2,793 ✭✭✭bops


    nice post now go study or else you'll be trapped here forever :eek:


  • Advertisement
  • Registered Users Posts: 2,004 ✭✭✭pok3rplaya


    Mellor wrote:
    I wasn't suggesting that if you play 2 or 4 tables you need twice or four times the BR, but in situations were you have just enough to play 2/4.(say 25 buy-ins) And mutli-tabling 4 tables leaves you with 16, If you got stacked on two of these, if you rebuy, you are jumping in with only 23 buy-ins.
    The ammount you should increase it by if multi tabling is probably very small, and should increase depending on number of tables.
    I'd imagine that the number of tables would be enough.
    if you multi-table 4 tables, then increase the limit of 25 to 29. (if games are not very agressive at a paticular level, it could be reduce to half the no of multi tables, so 4 tables at a passive level, increase 25 to 27)

    But if you play at 1 table instead of 4 with your 25 buyins, you'll still end off down at 23 buyins except it'll take you 4 hours instead of 1 hour or whatever.


  • Registered Users Posts: 762 ✭✭✭Duff Man Jr.


    What do people think for sng's, 25 buy ins should be plenty imo. But on 2+2 some people say 100 buy ins. One guy who had 900 10+1's played and a 3k bankroll was wondering if he should move up.


  • Registered Users Posts: 10,894 ✭✭✭✭phantom_lord


    I don't think you should adjust for multi-tabling, the varience is the same as playing one table, you're just getting through the hands quicker.

    anyway, if there is a difference for multi-tabling, whoever came up with the 20-30 buy-in rule most likely took it into account, since pretty everyone multi-tables.


  • Registered Users Posts: 2,004 ✭✭✭pok3rplaya


    What do people think for sng's, 25 buy ins should be plenty imo. But on 2+2 some people say 100 buy ins. One guy who had 900 10+1's played and a 3k bankroll was wondering if he should move up.

    I've very little experience with SnGs tbh though 100 does sound about right to me and I'm sure the 2+2 guys have a good idea what they're on about.

    Perhaps someone who is the shizzle at SnGs could make another post in this thread and maybe someone about Omaha too.


  • Registered Users Posts: 7,537 ✭✭✭Ste05


    Very nice post pok3rplaya, this is how we improve the Poker forum, by actually adding content instead of moaning about it, Fair play.

    A few comments:

    I agree with most of this, but have one comment in relation to this point:
    pok3rplaya wrote:
    Put simply that means that if you crush a certain game then you should, in the long run, be able to play with less variance than if you were playing in a game in which you are a breakeven player or marginal winner. Thus you can get a way with having a smaller bankroll.
    My problem with this is it's sometimes difficult to know if a player is actually crushing a game or just going through a period of positive variance, to be able to properly assess if someone is actually crushing a level he needs a decent sample size, and in order to get the appropriate sample size he needs an adequate bankroll to get it. It also allows someone to delude themselves that they are crushing a level and therefore they don't need a full BR, as far as I'm concerned the less subjective people can be with their BR management the better. I think if you are playing at any level you should always have at least 20-25 Buy-In's and more when you are moving up and less if for some reason you have to move down.

    It's difficult to properly articulate but basically, say if you are ordinarily playing 1/2 you should really have a BR of $4,000+ and then when you move up to 2/4 you should begin taking shots when you reach say this same threshold, (i.e. a BR of $8,000+), maybe drop a 1/2 table from your usual amount of tables and add in a 2/4, then add in another as your BR grows, and gradually you begin to get a sample size and when you have amassed say 30 Buy-In's from your shot taking and original 1/2 play, you take the full step up, then as your sample size increases you can see your results and if over a decent sample you are profitable at the level you can then relax your BR requirements back to 20-25 and repeat as you move up.

    I don't think there's any need to talk about moving back down as you'll usually know you can crush the game and if you need to move down you can play on a tighter BR once you're not tilting.


  • Registered Users Posts: 2,004 ✭✭✭pok3rplaya


    Yes ok thats a fair point Ste. Basically what I would say is that trying to amass a decent sample size to see how big of a winner you are in a particular game is pretty much impossible. I believe that even samples of 100 thousand hands can easily be off by +-2ptBB/100.

    So, the only other thing we can do is take an objective look at the game we are in at the moment and try to find out as much information as possible about the level above and decide if we're good enough to move up and then go for it. When moving up I'll often allocate a certain ammount of money to testing (say 4 buyins for the level I'm moving into) and move back down if I lose that. Regroup at the lower level and try again. Of course before I do that I'll be sure to have 30 buyins for the higher level so that if I feel like I can beat it easily but still run bad and lose, i'll still have enough of a bankroll to stay in the game. But yeah, it's best to be very conservative when moving into unknown territory.


  • Registered Users Posts: 7,537 ✭✭✭Ste05


    True, I think I was more trying to stop people deluding themselves that after 4,000 hands and a PTBB/100 of 8, that they are crushing a level and will continue on forever, then when their next 4,000 hands comes in at -4 PTBB/100 but an overall PTBB/100 of +4, they'll still think that well I was crushing the level and so this must just be a down swing.

    That's why when I deal with BR management I like there to be 0 amount of subjectivity allowed and it should all be based on objective quantifiable mathematics (not math!!).

    I also agree that we'll probably never get our true win rate and +/-2 PTBB/100 is probably about right over a significant sample, but if after 30,000 hands we have a PTBB/100 of over 5, then I think we can safely say we are profitable if not crushing the game.


  • Registered Users Posts: 2,004 ✭✭✭pok3rplaya


    Well just for some perspective

    400NL party, I played 75,000 hands with a ptBB/100 of over 7 and thought I was killing it easy, like best player in the world style. Now I'm 120,000 hands in and I've dropped to 2.3ptBB/100.

    I've also read cts (high stakes 2+2er and arguably one of the best in the world) say that he's had 12ptBB/100, 100k hand stretches followed by being breakeven for 100k hands.


  • Registered Users Posts: 7,537 ✭✭✭Ste05


    Yeah, I agree I think stats can only ever used for indicating patterns and the likes, you could go on for as long as you want and it still might not be a true win rate, but will basically always just be an indicator, the longer term we look at the closer to true it becomes but might never actually show a True Win rate.

    That's why I prefer to just keep everything objective with regards to BR management and keep a strict Buy-In amount for all levels depending on how risk averse you are and how close to the sun you wish to fly (risk of ruin)


  • Registered Users Posts: 2,886 ✭✭✭Marq


    plus your true win rate should always be changing as the conditions of the game do not stay static, i.e. you improve (or disimprove) as a player and the standard of player in the game gets better/weaker.

    Even something like a small increase in the amount of money in the game as a whole will make a difference to your winrate: say for example if everyone had to buy in for the max and keep topped up to the max, and the player pool remained constant (hypothetical, but used to illustrate the point).


  • Advertisement
  • Registered Users Posts: 2,437 ✭✭✭luckylucky


    Yeah nice post, I think you're buy-in criteria is pretty much spot on.
    What do people think for sng's, 25 buy ins should be plenty imo. But on 2+2 some people say 100 buy ins. One guy who had 900 10+1's played and a 3k bankroll was wondering if he should move up.


    I play a mix of cash games and sngs. I'll be bluntly honest, I don't really pay huge heed to what my bankroll requirements are for sngs, most of the time I do have at least 40 buy-ins available, anytime I start running too bad in them, I just stop playing them and grind it out in cash instead.

    Anyway getitng back on topic, if sng is your only outlet and/or you are strict with your bankroll... I'm surprised at people saying 100 buy-ins, this seems kind of high imo, not that I claim to be an expert, still I would have thought you could have a lot less than 100 buy-ins. After all you can always move down levels. Still though your 25 buy-ins might be a tad low though but not crazy low. Sngs can be highly streaky. There have been times when I've cashed in like 9 out of 10 in a row and other times where I'm like 1 cash in 20.

    Maybe 30 to 40 sounds about right.

    Have you got the link to the 2+ 2 thread(s) where people have said 100 buy-ins. As for the guy who has a $3k bankroll playing $10 games - ffs is all I can say to that.

    On Omaha - the last time I checked on the variance - it was about 50% higher than that of Hold Em. So that's the bad news, the good news though think the $1-$2 and $2-$4 Omaha games are softer than the Hold Em ones. So I would just increase the opr's buy-ins by 50%.

    On the subject of Omaha - If you buy-in shortstacked into full ring games i.e 20 big blinds - anyone have any idea of how many shortstacked buy-ins you should have given that variance is going to be even higher(relative to your buy-in size!)?


  • Registered Users Posts: 762 ✭✭✭Duff Man Jr.


    luckylucky wrote:
    Have you got the link to the 2+ 2 thread(s) where people have said 100 buy-ins. As for the guy who has a $3k bankroll playing $10 games - ffs is all I can say to that.
    100 buy-ins,
    http://forumserver.twoplustwo.com/showflat.php?Cat=0&Number=10393958&an=0&page=0#Post10393958

    The 3k Bankroll one,
    http://forumserver.twoplustwo.com/showflat.php?Cat=0&Number=10393221&an=0&page=0#Post10393221


  • Registered Users Posts: 2,437 ✭✭✭luckylucky


    100 buy-ins,
    http://forumserver.twoplustwo.com/showflat.php?Cat=0&Number=10393958&an=0&page=0#Post10393958

    Yeah the 2+2 Sticky recommends 100 buy-ins as well. So guess go with that then if you want to be very strict with your sng bankroll

    In the specific 2+2 post some guy says "my br management rule: if the amount of money i lose in a session/downswing does not make me puke, then br is fine. "

    i'll go with that one :D


  • Closed Accounts Posts: 2,771 ✭✭✭TommyGunne


    Could maybe using the Kelly criterion at half or quarter stakes be of any benefit. The only thing is you gotta estimate how likely you are to win/cash the tournament/SnG or as the case may be estimate your profit in cash games. This is easiest to do over a long history of hands. I havent heard of this being used in relation to poker before, as it is usually related to sports betting but the same principles should apply.
    If anyone knows much about this however they can correct me, as im no expert(just been introduced to this method.) However this should get a personalised and accurate figure for your bankroll.
    If you want to check it out on wikipedia...
    http://en.wikipedia.org/wiki/Kelly_criterion


  • Closed Accounts Posts: 2,771 ✭✭✭TommyGunne


    Just an idea, thought it would fit in this thread about bankrolls...


  • Registered Users Posts: 2,437 ✭✭✭luckylucky


    TommyGunne wrote:
    Could maybe using the Kelly criterion at half or quarter stakes be of any benefit. The only thing is you gotta estimate how likely you are to win/cash the tournament/SnG or as the case may be estimate your profit in cash games. This is easiest to do over a long history of hands. I havent heard of this being used in relation to poker before, as it is usually related to sports betting but the same principles should apply.
    If anyone knows much about this however they can correct me, as im no expert(just been introduced to this method.) However this should get a personalised and accurate figure for your bankroll.
    If you want to check it out on wikipedia...
    http://en.wikipedia.org/wiki/Kelly_criterion

    I used to use this way back in my soccer betting days.

    There are formulas you can use for your own personalised poker bankroll criteria. I have them in a spreadsheet. I don't actually bother with them myself but I just did a bit of research on the internet out of curiosity. The formula I used was basically a derivative of something else I found.

    The less mathematically inclined might want to look away now...

    Say you are prepared for a 5% chance that you will lose your bankroll playing at a certain buy-in level what bankroll do you need? Take your e(v) i.e. what you would expect your bb/100 to be(i.e try to be realistic about it)
    Call this figure E
    what is your sd/100 - this figure like your bb/100 can be gotten from poker tracker
    For simplicity call this sd/100 V for variance

    then the Bankroll you need is (E*((1.64*V/(2*E)) ^2) - (1.64*V*(1.64*V/(2*E)) ^2)) * (-1)

    So that would mean if your ev = 5, your std = 40 then the amount of big bets you would need to kick off at a level where you accept a 5% risk that you will lose your bankroll is

    (5*((1.64*40/(2*5)) ^2) - (1.64*40*((1.64*40/(2*5)) ^2))) * (-1) = 2607.83616 big bets , for simplicity round this of to 2600

    So according to those kinda stats one would need to have a bankroll of 2600 * 4(big bet amount for a $1-$2 game) = 10400 for a $1-$2 game if you wanted just a 5% chance of going broke if you had a bb/100 of 5 and a std/100 = 40.

    That's over 50 buy-ins as you can see.
    This doesn't take account however being prepared to move up and down levels so for a player meeting that criteria I think 30-35 buyins is ok.

    btw the figure of 1.64 is for a 5% chance, for a 1% chance you would replace it with 2.36 and for a 10% replace it with 1.282.


  • Registered Users Posts: 14,810 ✭✭✭✭jimmii


    It depends what constitutes your bankroll. The amount you are willing to lose before you quit playing totally or the amount you currently have and don't want to have to deposit again. An amateur can also have a lower sized buy-in bankroll as they can replace there bankroll with other income sources where as most pro's only have one source of income so need to be more protective of it.


  • Registered Users Posts: 2,437 ✭✭✭luckylucky


    jimmii wrote:
    It depends what constitutes your bankroll. The amount you are willing to lose before you quit playing totally or the amount you currently have and don't want to have to deposit again. An amateur can also have a lower sized buy-in bankroll as they can replace there bankroll with other income sources where as most pro's only have one source of income so need to be more protective of it.

    very true. Even pros if they have to can always work at something else for a while if they go el busto.

    I think the most important thing is common sense and don't let yourself spiral out of control with the games you play in, easier said than done mind!

    In practical terms no matter what your bankroll, for a lot of players anyway ,you probably have a minimum level you will play at. For example if I went bust tomorrow and had to come back in 6 months time with say $2K I'd take a shot at $1-$2NL. If I lost it - it wouldn't be the end of the world, I'd prefer to take the shot rather than grinding it out at say .25/.5 nl.


  • Registered Users Posts: 1,461 ✭✭✭RedJoker


    BR requirements are player specific, they depend on

    your win rate, difficult to accurately work out because of the reasons already stated in this thread.

    your playing style, playing a high variance style and pushing all your edges, which is what you should be doing, will mean you need a larger bankroll.

    your required risk of ruin, how badly do you want to keep this br? If you plan to reload if you hit a bad streak then it doesn't really matter how many buy-ins you have. On the other hand, if this br is very important to you then you'll want a low risk of ruin.

    Another couple of points: If you plan on moving down when you get below a certain number of buy-ins then you should never go bust. Being able to move down fast enough to protect your br takes discipline.

    If you are a losing player than you don't need to worry about bankroll requirements, no number of buy-ins will protect your bankroll, concentrate on becoming a winning player first.


  • Advertisement
  • Registered Users Posts: 2,953 ✭✭✭dvdfan


    Taught id add this post ive been reading on 2+2 to this thread particulary read the link to Fim's post about bankroll management.

    http://forumserver.twoplustwo.com/showflat.php?Cat=0&Board=ssplnlpoker&Number=6570513&Searchpage=1&Main=6570513&Words=+munkey&topic=&Search=true#Post6570513


Advertisement