Bankroll Management

[kolchak]

If there is one thing I'm complacent at, it's having a structure towards my bankroll, I'm a bit adhoc when it comes to this, and I often do what I feel comfortable at the time with. Has anyone got some tips for a guy like me?

 

[Daytripper]

If there is one thing I'm complacent at, it's having a structure towards my bankroll, I'm a bit adhoc when it comes to this, and I often do what I feel comfortable at the time with. Has anyone got some tips for a guy like me?

I structure it as 10 * 100 BB's per level in my roll.

So to play 10c/25c, I need at least $250 in my account.
To play 25c/50c, I need at least $500 in my account.
To play 50c/$1, I need at least $1000 in my account.

If you must drop a level, then do so - there is no shame in doing so.

You will play much better with good BR management.

 

[paxmaniac]

Two important principles: 1) Play a level you can beat, and 2) Never put money on the table that you can't afford to lose.

Many players will be much nittier than Daytripper's guidelines, but I think playing up to 10% of your BR online is ok (as long as you can beat the game) since there is always a smaller game to step down to. Be more conservative if you multi-table, since you could lose 3 or 4 buyins almost at once.

 

[mdc]

A really simple way to do it is;
- Stick to games you believe you can beat
- Drop down a level if your BR < 8 x buy-in.

 

[The Cryptkeeper]

Most literature suggests having a minimum of 50 buy ins when playing SNG's exclusively which is what I play.

That means a bankroll of $1100 if you play at the $22 level although I am pretty serious with my bankroll management and like to keep about 100 buy ins at any one time. As stated above though, there is absolutely no shame in moving down a level if you need to.

 

[paxmaniac]

Most literature suggests having a minimum of 50 buy ins when playing SNG's exclusively which is what I play.

Most "literature" assumes you play the same level until you bust when making risk of ruin calculations. If you have lower stakes to drop down to, you can easily play above 5% of your BR assuming you can beat the game and that you strictly apply rules on dropping levels when your BR drops.

Rules like 5% buyin are much more important when there are not many lower games to drop to, for example when you play live only.

 

[The Cryptkeeper]

Most "literature" assumes you play the same level until you bust when making risk of ruin calculations. If you have lower stakes to drop down to, you can easily play above 5% of your BR assuming you can beat the game and that you strictly apply rules on dropping levels when your BR drops.

Rules like 5% buyin are much more important when there are not many lower games to drop to, for example when you play live only.

Correct.....I should have added the above caveat.

 

[Reg Hickey]

I structure it as 10 * 100 BB's per level in my roll.

So to play 10c/25c, I need at least $250 in my account.
To play 25c/50c, I need at least $500 in my account.
To play 50c/$1, I need at least $1000 in my account.

If you must drop a level, then do so - there is no shame in doing so.

You will play much better with good BR management.

This is way too aggressive IMO.

In November-December I had a 16 buy-in downswing at 20NL - if I had been playing under your guidelines I probably would have had to drop at least two levels (or go broke), and I'd probably still be back at 5NL now. As it was I didn't have to drop down at all.

If you don't think a downswing like that can happen to you, think again. It happens to the best and worst players alike.

Especially if you are relatively new to the game, 20 buy-ins (a buy-in being 100BBs) should be the minimum. Personally I look for 20-25 - in fact right now my BR is $1150 and I haven't moved up to 50NL yet (although I'm about to).

I know very good micro/small stakes players who don't move up until they have 30 or even more buy-ins.

 

[mdc]

This is way too aggressive IMO.

In November-December I had a 16 buy-in downswing at 20NL - if I had been playing under your guidelines I probably would have had to drop at least two levels (or go broke), and I'd probably still be back at 5NL now. As it was I didn't have to drop down at all.

The idea is, once you get below 8 or so buy-ins, you drop down a level...then you're back at 16 buy-ins for that lower level.

If you don't think a downswing like that can happen to you, think again. It happens to the best and worst players alike.

Complete drivel. The better the player, the less likely he is to go on a big downswing. That's just maths.

With all due respect, if you're dropping 16 BI's at 20NL, you've got a ton of holes in your game that need fixing.

Especially if you are relatively new to the game, 20 buy-ins (a buy-in being 100BBs) should be the minimum. Personally I look for 20-25 - in fact right now my BR is $1150 and I haven't moved up to 50NL yet (although I'm about to).

I'll agree with the 20 buy-ins for n00bs, simply because they often don't have the discipline to drop down when required, and tilt too much.

If that's not an issue though, a lower minimum is fine.

I know very good micro/small stakes players who don't move up until they have 30 or even more buy-ins.

That's probably the main reason they're still stuck playing micro limits. I mean, I know a couple of guys who've built up over 3K playing nothing but 25NL and 50NL...doesn't make it the right strategy.

 

[The Passenger]

i agree with you RH. My ieals of good bankroll management are probably a bit nitty but i think it's better to be on the safe side. I'm now starting to get the discipline to stick to my ideals.

It also depends on your playing stay - a LAG naturally has more variance than a TAG player. Also whether you multitable, particularly if you tend to tilt a bit too much. If you multitable and can keep your emotions in check it's not too bad, but if you can't, then having a couple of bad beats on one tables can lead to poor play on 4-6 tables and you can run through a big chunk of your bankroll very quickly.

 

[The Passenger]

Complete drivel. The better the player, the less likely he is to go on a big downswing. That's just maths.

whilst that's true, it can and it definetaly happens. Even durrrr had a massive rough patch.

http://www.cardplayer.com/magazine/article/18095

Million-Dollar Downswing

When there are ups, there are, inevitably, a few downs. Dwan experienced the largest downswing of his career just before the 2007 World Series of Poker. After running well for a few weeks at the $200-$400 level, his bankroll hovered around the $3 million mark. Then, in the span of four months, he lost more than $2 million.

"That was really rough to deal with. I was thinking there was no way I was going to make that money back anywhere other than poker. It would take like 10 years if I was really successful somewhere else. It was some bad play and a ton of bad luck."

With two-thirds of a bankroll gone, it would be easy for many 20-year-olds to buckle under the pressure. However, Dwan proved once again why he's not just an ordinary player. With wisdom beyond his years, he rolled up his sleeves and put in the work. He dropped down to $25-$50 and steadily rebuilt.

"I didn't have much of a choice. It was either that or throw the rest of my money in a bank account and start working at McDonalds. Grinding it out sounded more fun."

 

[mdc]

whilst that's true, it can and it definetaly happens. Even durrrr had a massive rough patch.

http://www.cardplayer.com/magazine/article/18095

Million-Dollar Downswing

When there are ups, there are, inevitably, a few downs. Dwan experienced the largest downswing of his career just before the 2007 World Series of Poker. After running well for a few weeks at the $200-$400 level, his bankroll hovered around the $3 million mark. Then, in the span of four months, he lost more than $2 million.

"That was really rough to deal with. I was thinking there was no way I was going to make that money back anywhere other than poker. It would take like 10 years if I was really successful somewhere else. It was some bad play and a ton of bad luck."

With two-thirds of a bankroll gone, it would be easy for many 20-year-olds to buckle under the pressure. However, Dwan proved once again why he's not just an ordinary player. With wisdom beyond his years, he rolled up his sleeves and put in the work. He dropped down to $25-$50 and steadily rebuilt.

"I didn't have much of a choice. It was either that or throw the rest of my money in a bank account and start working at McDonalds. Grinding it out sounded more fun."

The difference is, the best high-stakes players operate at very small margins - everybody's good there! Massive downswings are understandable.

At 20NL, more than half the players are complete fish. If you're a decent player, your margin there is large.

 

[paxmaniac]

This is way too aggressive IMO.

Different strokes...

In November-December I had a 16 buy-in downswing at 20NL - if I had been playing under your guidelines I probably would have had to drop at least two levels (or go broke), and I'd probably still be back at 5NL now. As it was I didn't have to drop down at all.

On the flipside however, if you have aggressive BR management and hit a heater, you will jump up through the levels rapidly. It's an appropriate strategy for someone with very limited BR (say starting at 20NL), who thinks they can probably beat 200NL or 400NL and needs to move up as quickly as possible. There is absolutely no doubt that aggressive BR management will get you there faster on average, with still very minimal risk of ruin.

Especially if you are relatively new to the game, 20 buy-ins (a buy-in being 100BBs) should be the minimum. Personally I look for 20-25 - in fact right now my BR is $1150 and I haven't moved up to 50NL yet (although I'm about to).

The real reason (in my view) to have a very nitty BR strategy (say 30 BI or more) is that you are not certain you are beating the level. Perhaps you are still learning and improving. My own BR is nitty (for NL at least) for this reason - I am new to it, and learning.

 

[paxmaniac]

Another reason to continue playing down is because you might simply have a better absolute win rate. I play a bit of micro PLO occasionally (despite knowing very little about Omaha). On one site I play, I'm sure that my absolute win rate is higher at 10EUR tables than at 20EUR just because the players are so bad.

 

[Paul`]

If you are playing multiple tables (ie 15+) big downswings can happen very quickly... And of course big upswings.

But it does depend on your play and the risks you want to take....

 

[paxmaniac]

If you are playing multiple tables (ie 15+) big downswings can happen very quickly... And of course big upswings.

Multi-tabling is another kettle of fish entirely. One approach would be to assume that any money on active tables is not part of your BR and then apply your rules as normal (in deciding whether to buy-in and at what level).

 

[Reg Hickey]

mdc, I'm not going to reply to your post point-by-point because I couldn't be bothered and you probably wouldn't get it anyway.

Frankly, if you don't believe that good players can have downswings then you have no idea what you're talking about.

Generally speaking, the number and size of your downswings will depend on your style and how good you are. A very good LAG for example will experience much more volatility that an equally good TAG, and maybe even more than a mediocre TAG who is not as good a player but just plays tight and stays out of trouble.

But regardless of playing style, it is clearly possible for any player to lose money over a given number of hands. It is less statistically probable for some than it is for others, but that doesn't make it impossible. And given the sheer number of players and hands played online, it actually happens relatively often, and to players of all levels of skill and experience.

Tom Dwan has already been mentioned.

I also know of an experienced winning 100NL/200NL player who went on a 40 buy-in downswing last year.

Another player I know of who in the past has made over $1m profit from online poker within a calendar year, and is a proven winner all the way from micros to high stakes, recently said in his blog that he has lost more than 4bb/100 over his last 30k hands.

For a more general mathematical perspective, I suggest you have a look at these, which discuss the maths behind why even a solid winning player can easily lose money over a period of time:

http://archives1.twoplustwo.com/showflat.php?Cat=0&Number=8457727&page=0&fpart=1&vc=1&nt=15

http://archives1.twoplustwo.com/showthreaded.php?Cat=0&Number=4634922&an=&page=0&vc=1

Now my own downswing might well be due to a lot of holes in my game - I don't profess to be a gun. However, I have a PT database of my last 75K hands and even with that downswing in there, I am a solid 5ptBB/100 winner overall from 2NL to 25NL, which I would guess compares fairly favourably with 90% of the micro-stakes fish. Sure, I could still be a losing player at these levels, but that sample size makes it pretty statisically improbable, wouldn't you say?

 

[Reg Hickey]

On the flipside however, if you have aggressive BR management and hit a heater, you will jump up through the levels rapidly. It's an appropriate strategy for someone with very limited BR (say starting at 20NL), who thinks they can probably beat 200NL or 400NL and needs to move up as quickly as possible. There is absolutely no doubt that aggressive BR management will get you there faster on average, with still very minimal risk of ruin.

It might get you there faster on average, but it also means much higher variance in dollar terms, which means that the likelihood of running out of your relatively small pile of money on one of the dips is also much higher.

And if you want to keep most of your money I wouldn't play under-rolled in the hope of hitting a heater. Thats a recipe for disaster.

Look, at the end of the day, if you don't mind losing your roll because the amount is not important and you can just re-deposit anyway, then play with 10 buy-ins. Its entirely up to you. We all know the concept of leverage - you will magnify your wins and your losses, so as long as you keep beating the game then you will keep going up.

All I'm saying is that no matter how good you are you will eventually have a downswing due to variance, and if you are playing under-rolled when that happens you are likely to lose most or all of your money. And unlike heaters, it only needs to happen once and you're back to square one.

 

[paxmaniac]

If you move down quickly, then yes. But I know good players who have played under-rolled and lost a large proportion of their roll very quickly. It only takes 3-4 bad hands and your roll might be slashed in half.

Good aggressive BR management means you *must* move down quickly if required.

And if you want to keep most of your money I wouldn't play under-rolled in the hope of hitting a heater. Thats a recipe for disaster.

Heaters are just as inevitable as coolers - it's the other side of the same coin. If you are conservative, you will lose less and win less on average. If you are aggressive you win more and lose more. If you are a clear winning player at the level, then an aggressive strategy will result is a better average net result. If you prefer to minimize variance rather than maximize profit, that's your call.

Look, at the end of the day, if you don't mind losing your roll because the amount is not important and you can just re-deposit anyway, then play with 10 buy-ins.

You're missing the point. If you manage properly, you can play an aggressive strategy and *still* have minimal chance of busting. If you can rebuy after busting your roll, then it isn't really your roll.

 

[mdc]

Frankly, if you don't believe that good players can have downswings then you have no idea what you're talking about.

Find me where I said that. You said good players and bad players alike, which is rubbish.

But regardless of playing style, it is clearly possible for any player to lose money over a given number of hands. It is less statistically probable for some than it is for others, but that doesn't make it impossible. And given the sheer number of players and hands played online, it actually happens relatively often, and to players of all levels of skill and experience.

This is the key point that you are glossing over.

A -5BB/100 player will almost certainly have a 20 BI downswing within a 1 month period. A 5BB/100 player at small stakes will almost never have a 20 BI downswing.

To group both under the same umbrella is silly.

I also know of an experienced winning 100NL/200NL player who went on a 40 buy-in downswing last year.

Yes, and I know a guy who won second division in powerball. Both are extremely unlikely events.

For a more general mathematical perspective, I suggest you have a look at these, which discuss the maths behind why even a solid winning player can easily lose money over a period of time:

Again, let's be careful with our choice of words.

Now my own downswing might well be due to a lot of holes in my game - I don't profess to be a gun. However, I have a PT database of my last 75K hands and even with that downswing in there, I am a solid 5ptBB/100 winner overall from 2NL to 25NL, which I would guess compares fairly favourably with 90% of the micro-stakes fish. Sure, I could still be a losing player at these levels, but that sample size makes it pretty statisically improbable, wouldn't you say?

Let me make this clear.

If you are a +5BB player, and you went on a 16BI down-swing, there are a couple of explanations:
- a) You were playing your A game the whole time and are simply a victim of variance
- b) During your losing streak, you simply weren't the same +5BB player.

Statistically, b) is a lot more likely than a). And in my experience, b) is almost always the correct explanation.

Yes, tail-end variance exists. But it is important to recognize it's a very, very unlikely event, and plan your BR accordingly.

And as I said in my previous post...if you just drop down to a lower level once you reach <8 BI's, your risk-of-ruin converges to almost zero regardless of how bad you're running.

 

[Paul`]

It's good to see some debate on bankroll management. It is one of... if not the most important things to consider (especially online) when playing poker.

I have had the ups and downs.. and after 2 years or so I am now taking bankroll management very seriously...

 

[The Passenger]

It's good to see some debate on bankroll management. It is one of... if not the most important things to consider (especially online) when playing poker.

I have had the ups and downs.. and after 2 years or so I am now taking bankroll management very seriously...

i mentioned on 2+2 that the two of the most underrated skills in poker are bankroll mangement and tilt control (which both tie in together somewhat)...

and i think the reason they are underrated and often overlooked is because they are money savings skills. they're not money making skills.

and poker players, to generalise a fair bit, like to have a bit of gamble. and gamblers by nature, don't like to save money....

i read an interesting thing from daniel negreannu yesterday though where he said just about every high stakes player has needed to take a punt at one stage or another in their career.

basically, most of them have moved a level taking the profit ONLY from their current level, and seeing what happens... if it works then great, if it doesn't go back to grinding away at their original level and then go up a level the normal way (i.e. whenyou've built up enough b/roll). Unfortunately if it doesn't work you've lost the profit that you grinded away.....

he also mentioned that most of the guy's at bobby's room in the bellagio are probably a bit under-rolled for the game but willing to give it a shot.

 

[Reg Hickey]

Good aggressive BR management means you *must* move down quickly if required.

***

Heaters are just as inevitable as coolers - it's the other side of the same coin. If you are conservative, you will lose less and win less on average. If you are aggressive you win more and lose more. If you are a clear winning player at the level, then an aggressive strategy will result is a better average net result. If you prefer to minimize variance rather than maximize profit, that's your call.

***

You're missing the point. If you manage properly, you can play an aggressive strategy and *still* have minimal chance of busting. If you can rebuy after busting your roll, then it isn't really your roll.

No, I'm not missing the point. I understand, but I disagree.

Yes, heaters are as inevitable as coolers, and if you move down on a cooler then you reduce your chance of busting. Fine.

The problem with this though is that the impact of a cooler of any given magnitude is far greater than a heater. A 10BI heater from 10BI gives you 20 BI, so you move up and you're at 10BI again, just at higher stakes. A 10BI cooler from 10BI means you're busto if you don't move down, and probably 2 levels lower if you do, depending when you do. Working your way back from that is going to be tough. And guess what? You're still under-rolled!

Unless of course you move down every time you lose a couple of hands, which is crazy.

By the way, I saw in another thread that you have recently lost a quarter of your roll playing limit, which is inherently low variance and can therefore withstand much more aggressive BR management. I can understand playing with a 1000BB roll in limit. But we are obviously talking NL here - imagine your loss if you were playing that?

 

[Reg Hickey]

Find me where I said that. You said good players and bad players alike, which is rubbish.

Perhaps you need a dictionary. That doesn't mean it will happen to them with the same regularity. It just means it will happen to them as well.

Let me make this clear.

If you are a +5BB player, and you went on a 16BI down-swing, there are a couple of explanations:
- a) You were playing your A game the whole time and are simply a victim of variance
- b) During your losing streak, you simply weren't the same +5BB player.

Statistically, b) is a lot more likely than a). And in my experience, b) is almost always the correct explanation.

Yes, tail-end variance exists. But it is important to recognize it's a very, very unlikely event, and plan your BR accordingly.

And as I said in my previous post...if you just drop down to a lower level once you reach <8 BI's, your risk-of-ruin converges to almost zero regardless of how bad you're running.

Sigh.

In actual fact, its not "very, very unlikely" at all. I don't have the maths in front of me, but lets get the parameters right here.

The chances of an average 4-5ptBB/100 winning player who plays 25/20 or so for 100K+ hands of 6-max a year having a 10BI+ downswing at some point over say, a 2 year period, are not that remote. Its not 95% likely, but its not 5% likely either. It will happen to quite a few of them.

This really is a pointless argument. If your 10BI strategy works for you, do it. It doesn't work for me, and it won't work for a hell of a lot of people who are relatively new to NL because their variance will be way too high and they'll be constantly moving between limits (down as much as up).

 

[mdc]

Perhaps you need a dictionary. That doesn't mean it will happen to them with the same regularity. It just means it will happen to them as well.

Since you insisted:

a⋅like   [uh-lahyk] Show IPA Pronunciation
–adverb
1. in the same manner or form; similarly: They treated all customers alike.
2. to the same degree; equally: All three were guilty alike.
–adjective
3. having resemblance or similarity; having or showing no marked or important difference: He thinks all politicians are alike.

In actual fact, its not "very, very unlikely" at all. I don't have the maths in front of me, but lets get the parameters right here.

The chances of an average 4-5ptBB/100 winning player who plays 25/20 or so for 100K+ hands of 6-max a year having a 10BI+ downswing at some point over say, a 2 year period, are not that remote. Its not 95% likely, but its not 5% likely either. It will happen to quite a few of them.

Shifting the posts a bit with those numbers.

Still, if you care to specify some particular parameters I can run the maths for you - that way we can start talking about what the probabilities actually are, rather than what they're not.