Satelite and Multi-Table Tournament Bankroll?

BIGGYSLICKS

Well I know there is no exact answer to this one but I am curious to read your thoughts on this topic. For both answers assume that we are dealing with a winning player who can beat his or her competition given the proper amount of tournaments or satelites.

Part 1: What do you think a proper multitable tournament bankroll is? I have read several assumptions but none of them take into consideration the varying amount of players in each individual tournament. Is there a # that can be assumed based on # of entrants? Explain, thanks.....

Part 2: What do you think a proper Satelite bankroll is? I mean is it not just one large tournament that you will continure to play if you move on to the main event? Given that there is no prize money going to you unless you both win or qualify ion your satelite and then cash in the main event, what considerations or bankroll requirements would you suggest for both single and multi-table tournaments of varying amounts of participants.

Thanks, I know this is not an easy one but I think it is one that many are not sure about..... I anxiously await your replies...


Biggy

Dave Scharf

With apologies.... I have no idea. I will give you some of my thoughts but understand that I am guessing and have no math to back me up.

(1) When you consider that a completely safe bankroll for a winning player in a ring game is about 300 big bets (i.e. $6,000 for a 10-20 game) it seems to me that a completely safe bankroll multi-table tournaments must be very big. Tournament are MUCH higher variance than ring games.

(2) Generally, 10% of the field make the money. Depending upon the tournament format, skilled players have an EV somewhere between 1.5 and 4 (they expect to multpily their buy-in by their EV). That means that when a winning player enters a tournament he EXPECTS that TODAY he will lose his buy-in.

My advice is treat tournament as a hobby. Invest money into tournaments that you can AFFORD to lose. Think of ring game play as your blue chip stocks... they move up and down a little bit day to day, but the overall trend is up. Touraments are your penny stocks. They usually go bankrupt and you lose your whole invesement but they are worth investing in because OCCASIONALLY one of them takes off and you make A LOT of money.

That and, for me, tournaments are more fun.

It's not as specific an answer as you wanted but I have no idea how to answer the question more specifically. I have forwarded your question to a Phd math friend of mine for his thoughts which I will post here if he has time for us. He's busy with a national research report right now.

SupaMuncher

Just off the top of my head, it would seem that a tournament would have a lower standard deviation than a cash game would.
My logic is this: You have the potential to win more (and lose MUCH more) money during a cash game. Especially with the bigger, longer multi-table tournaments, you can expect to log significant playtime at those tables. My understanding is that most players have a variance that is about 30 times their expected win rate (ie. if you expect to win $20/hr, you can expect at worst to lose -$580/hr and at best $620/hr...approximately). In a tournament, even with a $500 buy-in (assuming, say, 10 hours of table time), your maximum loss rate would be -$50/hr.

Keep in mind that there's a difference between standard deviation and variance...I believe that the sd for a tournament is lower than cash, while the variance is higher...could be wrong, though.

SupaMuncher

Sorry...thinking about it, it doesn't make any sense for a standard deviation to be higher for one, and a variance to be lower for the other, since they are directly correlated (some good my math degree is doing for me :P)

Anyways, the "drop" for tournaments is usually higher than the rake in cash games (as a percentage), so maybe that has role to play, as well? Maybe I'm just not making sense...

BIGGYSLICKS

That is a great analogy and it kind of put things into focus on the subject for me. I do enjoy tournaments soo much though that if there are some sort of guide lines I would like to employ them... especially considering the availability of NET tourneys every 2 seconds.
I have been following the extra money approach for a long time and was looking to be a little more secure in my bankroll decisions because my tournament play far exceeds my ring game play these days.

Thanks very much and I would be anxious to read your friends reveiw of the post.

Thanks

Biggy

Dave Scharf

As regard the variance/deviation of tourament play you have to compare apples to apples.

For instance, if you are playing $5 single table sit-n-go, what is that equivalent to? For sake of argument, assume that you could show and EV of 2 for these sit-n-gos (e.g. every time you play you EXPECT to make $10 for a profit of $5). Since they take about and hour you show $5/hour earnings. What is the equivalent game? My guess is that $1-2 limit hold'em. My anecdotal experience tells me that the my "win graph" will look a lot flatter playing $1-2 hold'em than it will playing $5 sit-n-go.

As an example, in Casino Regina poker classics I am not 16 straight tournaments without hitting the money! A loss of between $9K and $10K. I am pretty sure that if I has spent all those hours playing $15-30 it would be almost impossible to be down that much.

SupaMuncher

I see your point now. I had trapped myself into looking at a sample size of a single hour, when I should have realized that the longer you play, the lower your variance is.

To illustrate that concept in the extreme: if you look at a single $1-$2 hand that takes 2 minutes, you could potentially win a $20 pot, for an hourly win rate of $600/hr. This is clearly not going to be representative of your expected win rate, which would be somewhere between $2 and $4/hr, if you're a good player.

Conversely, if you played for a year without stopping, I would imagine that your average win rate over that period of time would be within a few cents of your expected win rate.

So, in conclusion, I'm wrong.