cashing % in large tournaments

[levi]

just started playing the $22 and $33 buy-ins at party and after 25 of these, with fields ranging between 1500 and 2000 without a cash (top 130) I'm wondering if anyone has an idea of how many tourneys a good player could fail to cash in. Or, what kind of percentage cashing rate in the long run is realistic.

[PokerGoblin]

Cashing in those MTT's are not all that difficult, Half the field is gone after the first hour and by the second break it's down to about 200 or so. There are a lot of reckless idiots early on, try to maximize your gains in the pots you play.

I've only played in a few at party, I just started last week. I finished 21st out of 1200 in the last one, that's the best I have done. If all you want is your buy-in back, just make to the second break, then play only the top 5 starting hands, and you should be able to survive enough orbits to make the cutoff. Just make sure you have a big enough stack to survive the rising levels.

PG

[Potowame]

I think 25 straight out of the money is a problem. Can't really tell what it is, but I would think you need to post some bust out hands or hands where you lost a Large % of your chips, for review.

[Tom Bayes]

Well, if you are playijg MTTs that are paying between 6-10% of the field (let's say 8% for example), you should be cashing at a rate that is higher than the theoretical "average" player who will cash 8% of the time. I have a notion (don't have the data to support it) that a good MTT player will cash at a rate from 1.5 to 2.5 times the "average" player-how much more will be a function of that player's style and the caliber of opposition. A tight, conservative player should cash more often than an looser, more aggressive player, althoug the more aggressive player could have a higher ROI and a higher net profit due to the fact that they will make more final tables, where the real money is at. One top 3 finish will pay for a LOT of early bust-outs. To take an example from the WPT, a player named Casey Kastle has cashed in like 7 or 8 of the WPT events but has never made the TV table, while Gus Hansen has won 3 events and plays with the "I'll finish first or last" mentality. Obviously Gus has made a lot more $$$ off the WPT than Kastle.

In your case, if we assume you are "average" and should cash 8% of the time and are willing to make the necessary mathematical assumption to use the binomial distribution (each trial is indepedent and the probability of success is known and constant), then the probability of having X=0 successes/cashes out of n=25 trials/tourneys is .27. So an average player would have a 1 in 4 chance of having a dry spell like yours. A more proficient player with a 12% cash rate would go 0-for-25 about 4% of the time and a player with a 20% cash rate would go 0-for-25 less than 1% of the time.

I'm not going to do it, but one could find confidence intervals for these calculations. They would be pretty wide because the sample is still pretty small. So while it's not out of the realm of possibility that you have just been unlucky, you should consider the fact that you might very well be only an average or even below average player at the $20/$30 level. If so, your choices are to improve your game or drop down to lower buy-in tournaments.

[SossMan]

very good post, Tom.

[levi]

wow....great response. Thanks for the input. My experience prior to this with tourneys has been 100-200 entries at Vegas, and with these 1500-2000 entry tourneys I'm always busting out around #200-400, so I figured my mid game needs some work.

[getridofbush]

[ QUOTE ]
In your case, if we assume you are "average" and should cash 8% of the time and are willing to make the necessary mathematical assumption to use the binomial distribution (each trial is indepedent and the probability of success is known and constant), then the probability of having X=0 successes/cashes out of n=25 trials/tourneys is .27. So an average player would have a 1 in 4 chance of having a dry spell like yours.

[/ QUOTE ]

Not to be argumentative, but here's what stata tells me:

. di Binomial(25,1,.08)
.87563571

That is, if you have a true cash rate of 8%, in 25 trials you should have 1 or more success (cash) with probability .87563571, so the probability of a slump like this is actually less than 12.5%.

[ QUOTE ]
A more proficient player with a 12% cash rate would go 0-for-25 about 4% of the time and a player with a 20% cash rate would go 0-for-25 less than 1% of the time.

[/ QUOTE ]

Both of these numbers are correct.

[ QUOTE ]
I'm not going to do it, but one could find confidence intervals for these calculations.

[/ QUOTE ]

Suppose we want to create a CI for the number of cashes one would expect a guy with cash rate R in the set {.08,.12,.2} to have in 25 trials.

Let C be the (random) number of cashes in 25 trials, and assume what Tom did about constant cash rate, independent trials, etc. We can answer the CI question for intervals of significance level 90% by choosing interesting values of r and finding c_lo and c_hi such that

Pr(C=c_lo|R=r) = 0.95 and Pr(C=c_hi|R=r) = 0.05.

Thus

Pr(C=0|R=.08) = 1, Pr(C=1|R=.08) = .876 (see above), and
Pr(C=5|R=.08) = 0.045.

So a 90% CI for number of cashes given cash rate of .08 is
something like (0.5, 5).

Now consider r=0.12 (your average player):

Pr(C=1|R=.12) = 0.959
Pr(C=6|R=.12) = 0.071, Pr(C=7|R=.12) = 0.024.

Thus a 90% CI for the number of cashes in 25 trials for the average player would be sth like (1,6.5).

Lastly, for the expert with r=.2, we have

Pr(C=9|R=0.2) = .047
Pr(C=3|R=0.2) = .902, Pr(C=2|R=0.2) = 0.973,

so our 90% CI is sth like (2.7, 9) for the expert.

[Note that I have essentially used linear interpolation for the cases in which our probability values are bracketed by two integer values of c.]

This was interesting to think about---thanks for your post, Tom.

[MLG]

post some hands, or some questions. if you want to improve your game you are sure at the right place.

[rjamesd]

wow, that is some impressive math. (I always suspected that democrats were the smartest, but that seals the deal.)

So what I got out of the discussion so far is that I play WAY too conservatively. I play $5-10 online tourneys with 300-500 players. I've played maybe 100. I'm not full of excrement when I say that I cash about half the time. However the best I have ever placed is 6th and I think I've made the final table 4 times. Consequently my bankroll is inching up, but ever so slowly.

I think it would be too much trouble to descibe my whole tournament strategy and playing style but I'd say I'm typical tight aggressive. I start with premium hands, rarely limp in, and bet aggressively when I hit, unless I got the nuts and few possibilities of getting sucked out on. I also rarely bet less than top pair with a good kicker (except in the late stages and/or with less than 2 opponents.)

can anywone think of some things that I might look at to tweak my game? i.e.:

look to limp in with more multiway type hands (suited connectors, small pp, etc)?

steal more? I usually wait until the late stages to steal blinds and rarely try to steal on the turn or the river.

Put a picture af Daniel Negraneau or Gus Hansen on my computer?

[kickindickie]

I find limping with a variety of hands in the first few rounds (and later on if you have built a stack)of a tournament can be an excellent way of building your stack. Obviously, how frequently depends on your opponents and how aggressive the table seems to be. IMO, the implied odds you get when you hit "garbage" outweigh the small blinds you lose in the event that the pot is raised. Also, you have to be willing to release hands when the price is not right. For example, say you limp and get in a pot with a 67...and the flop comes 45A...you have to be willing to give that up if someone represents an ace (or any strong hand for that matter ) with a moderate bet. When you find players who are dumb enough to let you draw cheap... it's beautiful, especially with the added bonus you get from taking down a hand with garbage. Avoid hands such as ace rag that can get you into trouble...as you can often be beat even if you hit. Hope this helps.