You are viewing jhub3000

Previous Entry | Next Entry


Poker players like to analyze how lucky they are. Some are delusional players who play mediocre w/ an 8% ROI over 1K games & a selective memory that causes them to whine like little bitches at the tables. Others use luck analyzers such as all in EV on HEM to analyze how lucky they are. No matter which type you are, hopefully this entry will help you understand your luck better.

For players without a luck analyzing program the best tool you can use is probably sharkscope. If a player plays multiple buy-ins I think the best to tell if someone is likely running good or bad is to compare their average profit to their avg. stake x avg. ROI. For example, if somebody has an average stake of $100 w/ an average ROI of 2% then their average profit should probably be somewhere between $2 (0.02x100) & $2.40 (since avg. ROI is rounded to nearest whole # 0.024 x 100). Therefore, if their avg. profit is below that range there's a good chance they're running below expectation & if they're above that range they're likely running above expectation, likely due to running good at their highest stakes. Every year you see people towards the top of leaderboards whose avg. profit is 2 to 3x or even more than their avg. ROI x avg. stake. That generally doesn't last over the years. The delusional players will tell you that it's because they focus more at their highest stakes (who doesn't?) or they play better against good players but anyone w/ common sense can realize that's probably not the case. We also need to use common sense & sample size. If your avg. ROI is 6% & the next best reg at your stake has an avg. ROI of 3%, you're likely luckboxing it up.

Now let's talk about those who use luck analyzer. First off, let's get straight what a luck analyzer does. It only factors in 1 form of luck. If a player is all in for a 100 chip pot as a 70% favorite then his chip equity in the pot is 70 chips. If he wins the pot then he ran 30 chips above expectation & if he loses he ran 70 chips below expectation. A program such as HEM will then factor in chip stacks & the prize structure to turn these chips into a $ number & adjust your profits & ROI. It doesn't factor things such as card distribution (constantly calling someone at the top of their range, constantly pushing into the top of someone's range, etc...), which can have just as big of an effect on luck, at least in super turbos. I believe it also factors in the luck of the other players at the table, but I am not sure.

Now that we know what a luck analyzer does, let's look at some #s. I cannot confirm this but some math genius figured that all in EV is like multiplying your sample by 3.9. So if you have 1,000 games played & look at your adjusted ROI, it's like looking at your ROI over 3,900 games. Here's a few important sample size #s to remember:

After 500 games you're 50% likely to be within 5% of your true ROI
After 1,000 games you're 67% likely to be within 5% of your true ROI
After 3,000 games you're 90% likely to be within 5% of your true ROI
After 25,000 games you're 90% likely to be within 2% of your true ROI

Let's say you play 15,000 $50 STs. If we multiply that by 3.9 it's 58,500 games. Aside from the fact that your game & the games you play in may have changed a lot during that sample, 58,500 games should get you pretty damn close to your true ROI.

Now let's say you play 15,000 STs from $35 to $200 w/ an average stake of $72.67. your adjusted ROI #s at each buy-in are as follows:
2,000 $200s: -3.3%
2,000 $100s: -2.3%
7,000 $50s: 2.5%
4,000 $35s: 3%

This should workout to an overall adjusted ROI of 1.22%. However, despite having the same # of games as the 15,000 $50 ST sample, this grouping of games is likely to give you a far less accurate adjusted ROI because it is composed of smaller samples of games from different stakes. Looking at that group of adjusted ROIs & using common sense will tell you that. From $35 to $50 that's an avg. stake of $44.55 w/ an adjusted ROI of 2.68%. With an avg. stake of $72.67, a player shouldn't expect that 2.68% ROI to take too much of a dip, especially a 1.46% ROI dip, or an ROI that is less than 46% as high. This is why you need to look at your adjusted ROI at each buy-in level instead of just looking at your overall ROI, & use common sense. A more realistic estimate of that players' ROIs at $100s & $200s would be something like 2.2% & 1.3% respectively. With these new #s the overall ROI now jumps to 2.43%.

My point is that luck analyzers can be a useful tool IF YOU UNDERSTAND THEM. If you don't understand them they're likely to just cause more headaches for you.

As for 2011, the FTP STs have been ridiculously flooded w/ regs at every buy-in level. The games are definitely as tough as they've ever been. I don't see the worst pros being able to stick around at this rate, unless they live in a country with a very cheap cost of living.

Since voss is such a cool guy that he tweets he's been bugging me to plug his twitter. If you're the type of person who has been dying to know when voss takes a shit or eats a big mac, just click the link below & your dreams may come true:

http://twitter.com/voss1313

GL at the tables,
Hub

Comments

( 20 comments — Leave a comment )
(Anonymous)
Feb. 3rd, 2011 07:47 pm (UTC)
wow, nice work!
I just tried to use your average stake * average ROI/100 formula on about 4k games sample from 2 sites. Also I'm almost everywhere (in MTTs, MTT-SNGs and STTs) below expectation ranging from 5-40%, its nice possibility to know, how you're running in certain game types and when.

I'll do it with more detailed stats from my HEM and not from SharkScope next time, just curious what results I'll get then, thanks!
(Anonymous)
Feb. 4th, 2011 01:47 am (UTC)
hem ev stat
well hello there, just curious if you would be willing to put up a hem graph of your red line and maybe talk about normal swings/abnormal swings? Also how much thought do you put into the red line for these super turbos when alot of the things are almost unavoidable due to the nature of stack sizes, etc

thanks for your time mate looking forward to your reply
jhub3000
Feb. 4th, 2011 02:03 am (UTC)
Re: hem ev stat
I've actually tried posting a graph before but got some error on HEM. Maybe I'll try again sometime.

In STs I think most players will experience drastic swings like 200 buy-ins & 10K game breakeven stretches.

Ya as far as the red line for STs, it's more likely to be closer to your true ROI than your unadjusted ROI is, but there's certainly plenty of things it doesn't account for. If you're running 2% above expectation in all in EV, you could very well be running 2% below EV in card distribution, which make the adjusted stat pretty worthless. One form of luck I have never heard anybody talk about is seating assignment. Seating assignment can definitely effect your EV. If you have a shove happy reg to your right or someone who calls wide to your left more often than usual then you're running below EV in that department. There's just so many little things all in EV doesn't account for. Basically, kind of like I said in the entry, it's a nice tool, but use common sense as well & don't beat yourself up over it.

Just let you know if you have any specific questions.

GL,
Hub
(Anonymous)
Feb. 4th, 2011 03:54 am (UTC)
Re: hem ev stat
brilliant response mate you bring up some great points that are always overlooked imo. just a few more Q's

Card distribution ev you mentioned, is there any way to quantify it? as in running 2% below EV in card dist? I guess what Im trying to say is there any way to check/validate that or is it just a guess? Im assuming you reckon a guess.

When I mentioned swings i meant ev/red line bi swings for the super turbos not actual $, unless your answer was directed to this then I apologise mate!

Regarding the graph its rather simple mate just right click on the graph and save image as .jpg and host it on any site or even your blog maybe.

Cheers and good luck to you as well
jhub3000
Feb. 5th, 2011 02:14 pm (UTC)
Re: hem ev stat
There's definitely no way to quantify card distribution & I don't expect there to ever be. It seems like it would be too complicated. You'd never be able to remember either because you'd probably just have selective memory. If you have played a lot of games in your career you might have a better idea of what kind of ROI to expect. You could then look at your unadjusted & adjusted #s & estimate. A spot where you should be able to get a good idea is comparing ROIs buy-ins across levels, as in this blog entry. In the entry the player has an adjusted ROI of 2.5% at $50s over 7,000 games & 3% at $35s over 4,000 games, which averages out to an adjusted of 2.68% at an avg. stake of $44.55. Those sample sizes aren't perfect but they'll probably get you within 1% of your true ROI. We then see he has a negative adjusted ROI at $100s & $200s over 2,000 games each. Those are smaller samples & 2,000 games adjusted will probably only get you within 4.3% or so of your true ROI. Given these 2 sets of sample sizes, math & common sense lead us to believe that the player will actually have a positive ROI in $100s & $200s. We can then conclude that he was likely running below EV in card distribution at those buy-ins.

As far as EV/red line swings, I wouldn't be able to answer that right now since I'm not at home. However, I can tell you that playing such a wide range of buy-ins I have had numerous stretches of 2K games where my adjusted ROI/profit was negative. There was also a time when I was somewhere around $60-$100k below EV on my career. I have no idea what I am now because my databases are all messed up w/ some games in multiple databases & no cumulative database (I had to start new databases because HEM wasn't recognizing some structures correctly). I know h1nt saw a graph where I think I was at least $60K below EV though, so he can verify that. However, I never really looked at my adjusted #s at various buy-ins all too much over that span, so I can't really estimate card distribution effects. I am positive I ran way below EV in 2009 though, which should be common sense looking at my other years. If you have any more specific ?s about red line swings just let me know.

GL,
Hub
donkeyherder
Feb. 13th, 2011 07:17 am (UTC)
Re: hem ev stat
It's not a perfect tool considering hand distribution is situational, but you can use pokerstove to determine what % of the time you get various hands and breakdown stats using programs like poker tracker and pull the same hands say (AA, KK, QQ, JJ, AK, AQ) = 4.2% during any particular time frame or even filter it down to cash, tourneys, or sng's, and even down to exact games. It actually produces some pretty interesting stats and at least give better insight into your results. Gl, Donkeyherder.
jhub3000
Feb. 13th, 2011 07:32 am (UTC)
Re: hem ev stat
sounds like too much work for me :)

Obv this info is more helpful than not having it but I think it's the least important aspect of card distribution. I think card distribution is way too deep to analyze if you want to have any free time. For example, you might be getting QQ more often than you should, but that doesn't really matter if you're constantly shoving it into KK+, AK or and/or they fold way more than usual.

GL,
Hub
(Anonymous)
Feb. 4th, 2011 05:44 pm (UTC)
other players all-in
do you know if the HEM red line accounts for other players being all-in on bubbles? for example, when a player should bust and get you into the money but sucks out and stays on the table instead.
jhub3000
Feb. 4th, 2011 05:57 pm (UTC)
Re: other players all-in
I think it does but I'm not sure
jhub3000
Feb. 4th, 2011 08:39 pm (UTC)
Re: other players all-in
Actually after thinking this through I'm pretty sure it does account for this. If you look at another player under results it shows their adjusted ROI/profit so I can't imagine the program would calculate this but not make use of it for Hero's stats.

On a sidenote, other players adjusted #s will pretty much always be negative because it only factors in the hands that you were at the table. If you bust 6th & they take 1st HEM doesn't know this.

GL,
Hub
yudinetz
Feb. 6th, 2011 01:01 pm (UTC)
hi jhub!
im "gondyras" at FTP. what do u think about me? im very bad, dont i ?
could you give some adivces how to improve skill at STs ?
jhub3000
Feb. 6th, 2011 04:35 pm (UTC)
lol I don't coach. Games are tough enough as it is. You're good anyway. Trust me, I wouldn't say that if I didn't so. You're definitely a player I try to avoid if there's another good reg in the game w/ no fish.

Gl at the tables,
Hub
(Anonymous)
Feb. 10th, 2011 02:27 am (UTC)
9 man suer turbos
Howdy, enjoy your blog, do you plan on playing the 9man superturbos on Tilt this year? Imagine you would do well.
jhub3000
Feb. 10th, 2011 03:07 am (UTC)
Re: 9 man suer turbos
I don't have any plans for it at this time. With the current state of the game I don't see myself being able to just immediately jump into 9 mans & crush. I would probably have to do a significant amount of studying, which I just don't really have any ambition to do right now. If I was going to put the time into studying a different game it would probably be something other than 9 man STs. I'm doing just fine w/ 6 mans though.

GL,
Hub
(Anonymous)
Feb. 21st, 2011 03:42 pm (UTC)
hi
Do u have an affilate program in Cake , if yes how can i register there to became your player ?

P.S.
Have u ever been in Belarus ?
jhub3000
Feb. 21st, 2011 04:59 pm (UTC)
Re: hi
I do have an affiliate code w/ cake. Just type jhub as the signup code for 33% rakeback.

I have never been in Belarus.

GL,
Hub
(Anonymous)
Feb. 27th, 2011 07:24 pm (UTC)
Tax Accountant
Hey Jared, If possible could you please give me the name and or website/phone number of your tax accountant, I'm in Minnesota and would like to set up an appointment with him. Thanks Much.
jhub3000
Feb. 27th, 2011 09:33 pm (UTC)
Re: Tax Accountant
http://bergersontax.com/tax-form/contact-bergerson-tax-services/

Make sure to make an appointment w/ Jeff Bergerson. he has about 15 other poker clients. Schneids told me about him. Actually if you remember can you mention I referred you? I'm sure it will be good for some goodwill. :)

GL,
Hub
(Anonymous)
Feb. 27th, 2011 09:45 pm (UTC)
Re: Tax Accountant
I sure will. Hey man, I appreciate the prompt response. Good luck with the rest of the year!
jhub3000
Feb. 28th, 2011 02:00 am (UTC)
Re: Tax Accountant
Thanks. U2
( 20 comments — Leave a comment )