Just turned pro got a couple questions?

[pifhluk]

[ QUOTE ]


Regulars say "Your dumb. You dont understand sample size. Youre on a heater. Its going to get worse."

Very few of these regulars back up any of there statistical assertions with any sort of math.

[/ QUOTE ]

Math does not tell you [censored] about how you are going to react when you lose every 80/20 every 70/30 every 50/50 and every 5,3,2 outer hits against you. Downswings can be absolutely brutal and until a player has played through one and succeeded their results are pretty much meaningless.

[IFoldPktOnes]

I did some math for you, heres the results:


I plugged a ROI of 35% over a 500 sample into the ROI simulator and then conditioned on the fact that a ROI over 25% isn't possible (I have no idea if thats true, I justed picked a number).

I didn't condition on the content/nature of OP's posts.

[dktoller]

[ QUOTE ]
any sort of math

[/ QUOTE ]
Let's put aside the speculation and look at toerazor's numbers. First, some assumptions. 1. 514 games all 9-man SNGs with 10% rake. 2. ITM of 43% 3. Equal placements of 1st/2nd/3rd. (This equates to the reported 32% ROI).

Now we can establish a confidence interval that makes a quantitative statement about his true ROI. With 95% confidence we can state that the true ROI is at least 19%. That dog can hunt! (Coincidentally, if he loses his next 100 games... his ROI will be 19%!)

Naturally to 'go pro' you need more than just poker skills, but that's not my point. In this case 500 games is more than enough to draw some strong conclusions about long-term ROI.

[Calculation details: Assume normal distribution with sample std dev based on four types of placements: 1/2/3/none, weighted 5/3/2/0. Variance = 3.38, SSD = 1.84. 95% one-side confidence interval, alpha = 1.645. Subtract 10% rake.]

Now there is an expectation (think Bayesian stats) of attainable rates, and I think it fair to argue his true ROI is likely less than the current 32%. (Like a ball player who hits .460 in May is rarely finishes the season above .400). But at the same time he is not playing turbos (less variance and higher ROI) and is not doing extensive multi-tabling (which often is assumed when people discuss attainable rates).

[dktoller]

Curious, that's very different from the stats I came up with. Obviously due to the fact of 'conditioning' against the upper bound of 25%. I'm in agreement with the reasoning, but how exactly does that work?

This puts his most-likely ROI around 17-18%.

[Fiasco]

[ QUOTE ]
[ QUOTE ]
Very few of these regulars back up any of there statistical assertions with any sort of math.

[/ QUOTE ]

[ QUOTE ]
Now Im not familiar with the exact math involved for approximating longterm expectation on ROI

[/ QUOTE ]

Sorry but this made me laugh.

[/ QUOTE ]


OK, thats legit. Im just saying I want somebody to do the math. Im open to the posssibiility that Im wrong.

[Fiasco]

[ QUOTE ]
[ QUOTE ]
any sort of math

[/ QUOTE ]
Let's put aside the speculation and look at toerazor's numbers. First, some assumptions. 1. 514 games all 9-man SNGs with 10% rake. 2. ITM of 43% 3. Equal placements of 1st/2nd/3rd. (This equates to the reported 32% ROI).

Now we can establish a confidence interval that makes a quantitative statement about his true ROI. With 95% confidence we can state that the true ROI is at least 19%. That dog can hunt! (Coincidentally, if he loses his next 100 games... his ROI will be 19%!)

Naturally to 'go pro' you need more than just poker skills, but that's not my point. In this case 500 games is more than enough to draw some strong conclusions about long-term ROI.

[Calculation details: Assume normal distribution with sample std dev based on four types of placements: 1/2/3/none, weighted 5/3/2/0. Variance = 3.38, SSD = 1.84. 95% one-side confidence interval, alpha = 1.645. Subtract 10% rake.]

Now there is an expectation (think Bayesian stats) of attainable rates, and I think it fair to argue his true ROI is likely less than the current 32%. (Like a ball player who hits .460 in May is rarely finishes the season above .400). But at the same time he is not playing turbos (less variance and higher ROI) and is not doing extensive multi-tabling (which often is assumed when people discuss attainable rates).

[/ QUOTE ]

Friggin awesome, thank you. Over 19% ROI with a 95% confidence! Thats pretty damn good.

And if Im not mistaken that would mean that the upper bound of that 95% confidence interval is 45% ROI.

tyvm dktoller, this is exactly what this thread needed.

[Fiasco]

[ QUOTE ]
I did some math for you, heres the results:


I plugged a ROI of 35% over a 500 sample into the ROI simulator and then conditioned on the fact that a ROI over 25% isn't possible (I have no idea if thats true, I justed picked a number).

I didn't condition on the content/nature of OP's posts.

[/ QUOTE ]

cute.

[Toerazor]

Thanks alot for your input guys. I do realize that I probably will not run at 30% ROI in the long run. But I do believe that I can sustain an ROI aroun 25% percent in the $30 SNG. Once I move up to the $50 SNG i realize if I stay at 20% ROI that will be great. Another factor that I dont know if its true or not is that I play on Full Tilt. I used to play on pokerstars under same name and thought the competition was much harder. I was also learning the game then and had not played nearly as much as I have now. I also started playing cash games couple nights a week about a year and a half ago which I believe helped out my game alot. Living in Las Vegas and talking to a couple people who play for a living helped out also. For my bankroll management which I am very good at I believe. I dont really spew off money ever and if I lose $150 in a day I will stop. I have only done a couple times which is nice. When I am running good and making money I play as long as possible. When I was talkiing to a pro in Vegas one of the most important things he told me about playing poker for a living was limiting your losses. Controlling your losses is one of the important things I constantly think about. I feel that helps me alot. I also keep 100 buy ins for each level in my bankroll including rake. I will move up to the $50 SNG when i reach $5500 bucks. Hopefully things continue to do well. Have a good one. Peace

[saddu]

according to startracker the best player with an average by-in between 24-50$ after 2000 games (non turbo but im not sure) has an average roi of 14%, althought i beleve u can substain
a 20-25% if u play only 4-6 tables at time and playing so grat as you've done so far good job. (sry english)

[pdiggz]

[ QUOTE ]

[Calculation details: Assume normal distribution with sample std dev based on four types of placements: 1/2/3/none, weighted 5/3/2/0.

[/ QUOTE ]

Assuming a normal distribution here is isn't correct, no offense. His place distribution will look nothing like that. Especially when you say previously.

[ QUOTE ]
Equal placements of 1st/2nd/3rd.

[/ QUOTE ]