Rake Sprint!

[C8H10N4O2]

Haha...it says it on the front page, Tony.

"Over the next two weekends you can sprint your way to a share of $5,000 at Full Tilt Poker. We will host two exclusive 48 hour rake races, each with a prizepool of $2,500 so don't just sit in front of your TV this weekend, sprint your way to some extra cash! For more details please visit our promotion page."

I was also under the impression it was a dual-weekend thing lol... perhaps because this was on the front page.

[arjun13]

So Tony, what is the feedback from your side now that the rake sprint is over? Did you guys make more money despite giving away some on the sprint because we played more?

I think all of us would really love more of these races. Perhaps week-long ones are ideal, where standings can be updated. This would give us a better picture of where we are and how much more we need to play. It would motivate us to play more.

Please let us know.

[pajala]

Tony

So is the promotion running this weekend i notice it has been removed from the website?

[Mika Ollila]

Tony
So is the promotion running this weekend i notice it has been removed from the website?

The sprint is not on this weekend. Next one should be around middle of March.

[arjun13]

The sprint is not on this weekend. Next one should be around middle of March.

Why not have one every weekend? We will generate a lot more rake. Profit for us and for you.

And what is the verdict on week and month long rake races?

[grimReaper]

I was 8tabling Rush. 4 of nl50 and 4 of nl100. It went really well I thought. I was surprised I was able to stay focused over 2 days of that length.

The graphs don't include my first hour of play since it was 11pm Central Time Friday when I started playing, therefore PokerTracker doesn't count it as part of my weekend session.

Wow nice.

Interesting to see how much rake is taken. Assuming FTP's RNG is perfectly sound, by 40k hands is a huge "sample" from statistics viewpoint and the difference of expected and actual money won should converge to 0 if no rake, and the remaining should be rake (though it does go to 0 in the last 10k hands). Just shows how great rakeback is.

[travz21]

Wow nice.

Interesting to see how much rake is taken. Assuming FTP's RNG is perfectly sound, by 40k hands is a huge "sample" from statistics viewpoint and the difference of expected and actual money won should converge to 0 if no rake, and the remaining should be rake (though it does go to 0 in the last 10k hands). Just shows how great rakeback is.

Wat ?

[grimReaper]

Wat ?

Why is there such a large gap between expected and actual earnings?

[travz21]

Why is there such a large gap between expected and actual earnings?

Because when I got all in I was running bad? Do you ask yourself these questions and ever try to answer them on your own? And the gap isn't that big. A couple buyins below EV after 40k hands isn't much. It could have just been 1 pot that didn't go my way. Everything is more even once I get towards the end of my weekend session in terms of EV, though. I think I was only about 40 bucks below.

[grimReaper]

Because when I got all in I was running bad? Do you ask yourself these questions and ever try to answer them on your own?

You couldn't infer what I was really asking? Your the expected wins basically bounds your actual winnings for 45000 hands.

[C8H10N4O2]

You couldn't infer what I was really asking? Your the expected wins basically bounds your actual winnings for 45000 hands.

Wat ?

[travz21]

You couldn't infer what I was really asking? Your the expected wins basically bounds your actual winnings for 45000 hands.

I knew what you were asking. What I don't get is why you would ask such an obvious question.

[grimReaper]

I knew what you were asking. What I don't get is why you would ask such an obvious question.

I made a mistake. The (expected - actual)/total hands should converge to 0, I forgot about "divided by total hands."

Do you ever grind PLO? If so, is there a significant difference in profits?

[travz21]

I've played like 1k hands of .01/.02 PLO and got destroyed. I'm not sure what the higher spectrum of winrates are in PLO, but they'd have to be pretty high to counter the ability to 16table NL cash. I could only 8 table PLO and barely make it by.

[RAW_FORCE]

5ptbb+ winrates are pretty common at micro-low stakes 6-max PLO. As far as 40k hands being meaningful whatsoever...LOLOLOL. MAYBE vpip/pfr will have significantly converged at that point but thats about it.

[grimReaper]

As far as 40k hands being meaningful whatsoever...LOLOLOL.

Why not? Looking at the biggest gap at 35k, which is $500, and compare to the medium size gaps at around 10k-15k which are around $200, clearly the gap/total hands is approaching 0, especially if you consider rake. This is basic statistics.

[travz21]

Why not? Looking at the biggest gap at 35k, which is $500, and compare to the medium size gaps at around 10k-15k which are around $200, clearly the gap/total hands is approaching 0, especially if you consider rake. This is basic statistics.

You lost me again.

40k hands doesn't really mean much. Raw is right. 100k is the smallest sample that I actually start taking my stats to heart.

[C8H10N4O2]

Why not? Looking at the biggest gap at 35k, which is $500, and compare to the medium size gaps at around 10k-15k which are around $200, clearly the gap/total hands is approaching 0, especially if you consider rake. This is basic statistics.

I'm really confused why you keep saying "if you consider rake". You realize that in all the tracking software and stuff, rake is factored in, so there isn't any reason to "especially consider rake".

Also from my understanding PLO carries a lot higher variance than hold'em. A lot of people say you need 50k hands at a bare minimum for hold'em, so I would assume you would need even more than that for PLO.

[grimReaper]

You guys have the grinding experience, so I'll your word for it.

And actually the math confirms it. If you play .25/.50 NL or PL, for you to be +/- $0.01/hand 95% of the time from your "true" $winning/hand, whatever it may be, you must play n hands, where n:

0.01 = 1.96*SD(X)/sqrt(n)
where X is the random return of playing one hand.

so finding the implied SD(X) and seeing if it's consistent w/ reality:

SD(X) = sqrt(n)/196

if n=40000, SD(X) must be $1.02, if n=100000, SD(X) must be $1.61, both which seem too low to the true SD(X) when playing 0.25/.50. and of course you can make some guesses for SD(X) and find out how many hands you have to play to be within $0.01/hand to your true winnings/hand.