MTT Bankroll thoughts

[A_Junglen]

There's some good advice here on developing some kind of system to pay yourself. However, bankroll management is a very perosnal subject. There are many variables, enough so that I don't believe there to be a "correct" way to manage ones bankroll.

Variables include

- Intent: Do you have a job, or is poker your primary income?

- Borrowing Power

- Ability to reload

- Bills and/or expenses

- Backup money available?


And many more I'm sure. I've posted something similar to this before, but bankroll management is a common subject and therefore I repeat myself.

There will always be guidelines, but in the end it's up to you.

[BennyMac]

I got to thinking that there should be more of a theoretical basis for the 50-100+ buy-in rule of thumb. I found this old thread that had a formula intended mainly for cash games:

http://archiveserver.twoplustwo.com/show...rt=all&vc=1

The post with the formulas comes third.

Here's an associated calculator:

http://www.reviewpokerrooms.com/poke...uirements.html



Can this be reasonably applied to tournaments? I tried, but the resultiing bankroll requirements seem impossibly large. Could it be that there is some problem coming from the fact that all of the outliers are on the positive side? The most you can lose is 1 buy-in, but you can win big. This is missed by simply using standard deviations.

I applied it to my large (>50 people) tournaments at $10+1 or more. I've played 177 with an average of over 1000 players and shown an average profit of 3.1 buy-ins and a standard deviation of 28.1 buy-ins. Using this formula and a bankroll of 100 buy-ins gives me a 46% chance of going broke. That doesn't strike me as reasonable.

Any statisticians out there who can speak to this? A computer program simulating 1000's of tournaments would be interesting as well.

[Yuv]

[ QUOTE ]
SUGGESTION:

Don't lose 2/3 of your bankroll in one night playing 5/10 and 10/20 O8B

WEEEEEEEEEEEEEEeeeeeeeeeeeeee

HELLLOOOO $4/180s!!

[/ QUOTE ]

That, and don't go on life tilt and blow 10k in 1k HU matches, or you'll find yourself chopping 375 FPP sats [img]/images/graemlins/smile.gif[/img]

[bonds25]

[ QUOTE ]
[ QUOTE ]
SUGGESTION:

Don't lose 2/3 of your bankroll in one night playing 5/10 and 10/20 O8B

WEEEEEEEEEEEEEEeeeeeeeeeeeeee

HELLLOOOO $4/180s!!

[/ QUOTE ]

That, and don't go on life tilt and blow 10k in 1k HU matches, or you'll find yourself chopping 375 FPP sats [img]/images/graemlins/smile.gif[/img]

[/ QUOTE ]

haha, laughing with you not at you.

[BennyMac]

I wrote a little program to calculate my risk of ruin at given bankrolls. I used my past performance as the distribution of outcomes and counted going broke within the first the 1,000 tournaments played as a bankruptcy. I then ran 100,000 of these 1,000 tournament trials and considered my risk of ruin as the percentage of the 100,000 trials where I went bankrupt.

With 50 entries, my risk of ruin was roughly 6%.

With 100 entries, my risk of ruin was roughly 0.4%.

I think overall (but not lately) I've been fairly lucky so I also ran it while removing my big scores from the distribution.

100 entries:

Without top 1: 1.3%
Without top 2: 7.4%
Without top 3: 41.0%

I think I'm going to play whatever level I have < 5% chance of going bust at. Move down as soon as it hits 5%.

If there's any interest, PM me and I can run your numbers too.

[danielg]

Benny mac,
im not the smartest person going round when it comes to math, anychance you can put that into english for me :P

Or you can email it to me @ dgreenwood18@hotmail.com


Danielg

[0evg0]

[ QUOTE ]
[ QUOTE ]
SUGGESTION:

Don't lose 2/3 of your bankroll in one night playing 5/10 and 10/20 O8B

WEEEEEEEEEEEEEEeeeeeeeeeeeeee

HELLLOOOO $4/180s!!

[/ QUOTE ]

That, and don't go on life tilt and blow 10k in 1k HU matches, or you'll find yourself chopping 375 FPP sats [img]/images/graemlins/smile.gif[/img]

[/ QUOTE ]

I've chopped a 375 FPP sat before. It was a new low. I think I would rather just go perma-busto than do that again. I seriously expected support to show up and take an extra 5 minutes to do the chop just because they were laughing too much.

Catch me in the $3 LO8 and a $4/180 at the moment.

And my BR is now officially 1/6 what it was 3 days ago. I'll keep you guys posted.

[BennyMac]

[ QUOTE ]
Benny mac,
im not the smartest person going round when it comes to math, anychance you can put that into english for me :P
Danielg

[/ QUOTE ]

I can try. The math isn't complicated so any confusion is likely my fault.

The tournaments that I'm interested are those with more than 50 entrants. In my case, I'm also eliminating the lowest stake tournaments (anything less than 10+1 unless it's a rebuy.) This leaves me with 178 tournaments.

I use my performance in these 178 tournaments to predict my future performance. They're mostly 10 dollar tournaments, but I normalize so that all entrys are equal to 1 and all payouts are divided by the entry fee. Since I finished out of the money 148 times, there is a 148/178 chance a given tournament will show a profit of -1. My biggest profit for a tournament was 327.81 entries. Each tournament I simulate, I have a 1/178 chance of getting a profit of 327.81.

I specify my bankroll in advance as a function of entry fees. Let's say I have 50 entry fees worth. I simulate 1,000 tournaments and if at any time my bankroll hits 0, I stop and call that a bankruptcy. I do this 100,000 times and count the number of bankruptcies. I call the percentage of times that I go bankrupt my "risk of ruin". In my case, it's roughly 5.9% when I have a bankroll of 50 entries. If my bankroll was only 10 entries, there would be a greater than 50% chance that I would go bankrupt.

Does that clarify things?

[PokerLearner]

BennyMac, so every time you play an additional tournament you update your risk of ruin (i.e. if you finish out of the money in your 179th tournament, do you re-calculate your risks)?
Why not include the lower buy-ins, too? After all, there is a point where you are going to drop down to these buy-ins, and you want to know your risk of ruin in them, too, right?
Also, because I travel around the world a lot I get to play at weird times, and have noticed a huge effect of time of day I play (e.g. afternoon East-Coast Time vs. late evening vs. weekend). Do you think it would be worthwhile to take this into account?
Your way of calculating bankroll requirement sounds very interesting. Do you just calculate this in Excel, or what software do you use to run these simulations?

[BennyMac]

[ QUOTE ]
BennyMac, so every time you play an additional tournament you update your risk of ruin (i.e. if you finish out of the money in your 179th tournament, do you re-calculate your risks)?

[/ QUOTE ]

Well, I just wrote the simulator yesterday, but I plan on adding my most recent tournaments every time I want to recalculate my risk of ruin. I won't be doing that after every tournament though.

[ QUOTE ]

Why not include the lower buy-ins, too? After all, there is a point where you are going to drop down to these buy-ins, and you want to know your risk of ruin in them, too, right?


[/ QUOTE ]

I thought the lower buy-ins would bias the results in my favor. If I had to move down, then I would definitely include them.

[ QUOTE ]

Also, because I travel around the world a lot I get to play at weird times, and have noticed a huge effect of time of day I play (e.g. afternoon East-Coast Time vs. late evening vs. weekend). Do you think it would be worthwhile to take this into account?


[/ QUOTE ]

You would know better than I would. If it affects your play greatly then I would say you should. You might want to play at different buy-ins at different times. Personally, I haven't kept track of times.

[ QUOTE ]

Your way of calculating bankroll requirement sounds very interesting. Do you just calculate this in Excel, or what software do you use to run these simulations?

[/ QUOTE ]

I wrote it in C. It's pretty ugly, but gets the job done.