Variance question for those who play a lot of MTT's

[Emperor]

I have finished in a few final tables online in my day, but I am pretty sure my ROI is negative thus far.

Although I play mainly ring games. I would like to start adding a nightly MTT to my schedule to see if it will break up the monotony and encourage me to play the 4-8 hrs the MTT is going. However I am curious about the variance involved.

Let say I play the nightly 15+1 at Party, it averages 2K entrants. Lets say skill wise I am better than 50% of the donks in the tourney, but 50% are better than me (for now).

How many MTT's before I can ballpark my ROI with any confidence? (for stats guys, lets say 66% confidence)

Will it be worth my time? or Do I need to be better than 90% of the players skill wise to make it worth my while? (lets say worth my while is a measly $10/hr average over 100?? tournaments)

[Emperor]

bump

[nycplayer]

I'm curious about this as well. I wonder what percentile you would have to be to be confident about making money longterm. I've come in first and second in several small tournaments, live and online (30-40 people) but never finish in the money on big MTTs.

[Jerrod Ankenman]

[ QUOTE ]
I have finished in a few final tables online in my day, but I am pretty sure my ROI is negative thus far.

Although I play mainly ring games. I would like to start adding a nightly MTT to my schedule to see if it will break up the monotony and encourage me to play the 4-8 hrs the MTT is going. However I am curious about the variance involved.

Let say I play the nightly 15+1 at Party, it averages 2K entrants. Lets say skill wise I am better than 50% of the donks in the tourney, but 50% are better than me (for now).

How many MTT's before I can ballpark my ROI with any confidence? (for stats guys, lets say 66% confidence)

Will it be worth my time? or Do I need to be better than 90% of the players skill wise to make it worth my while? (lets say worth my while is a measly $10/hr average over 100?? tournaments)

[/ QUOTE ]

Well, you can't use a normal approximation because the results are so skew. My off-the-cuff estimate is that you should probably have about half as many tournaments as there are players before you can have any reasonable estimate of your expectation, unless it's absurdly high (like seven buyins/tourney or something).

By the way, if you're better than 50% of the field and worse than 50% of the field, you're going to lose money. The reason is that 1) you pay juice and 2) the field self-selects the better players as the tournament goes on, so you'll be overall playing against a field that is better than you.

Jerrod Ankenman

[Delphin]

You will have to play many times the number of entrants in order to get a good estimate of your true win rate and ROI. The reason for this is that the majority of your winnings come from placing in the top few spots in the tournament, and your win rate depends very heavily on how often you do that.

How many tournaments do you have to play to figure out how likely you are to place first in the tourney? Probably at least enough that you have won 10 times or more in order to get a rough approximation of how often you will win. If you have placed 1st-10th about 10 times each, then you can probably calculate with reasonable accuracy your expectation for each tournament.

If you've won once, never placed 2nd or 3rd and have a couple other final table finishes, forget it. You can calculate a wild guess at your ROI, but that's all it will be.

Frequent SNG players agree that you have to play a couple hundred tourneys to get a really good estimate of your ROI, and this is in a 10 player tourney. I'm sure you can see that MTTs are much much more difficult to estimate ROI from.

[Emperor]

After not having played a tournament since february, I finished 5th in the 1340 entrant 15+1 on Party last night. I really doubt that I can expect a positive ROI though, as I don't think I am that much better than the rest of the field.