DERB - Page 12 - High Stakes Poker Forum - High Stakes Limit Texas Holdem

They aren't ignoring the hard numbers. The numbers tell you that there is a 95% chance that he is a winner. Lots of people play the 30/60, he seems to be a lottery winner, that 5% one. Even if you put his stats into 99% or 99.99% confidence intervals the argument is still valid.

Math is hard. The numbers don't say there's a 95% chance he's a winner. They say something else entirely.

When you start asserting that he's a lottery winner, the question then becomes:

"How do we know that James isn't the lottery winner?"

- Andrew

www.pokerstove.com

Because James can look in his database and see himself outplaying this guy hand after hand?

Because James' stats are generally in alignment with the vast majority of winning poker players?

Because James has undoubtedly maintained his success far longer than 100k hands?

Shall I continue?

If you like. Of course, the only difference is how big the lottery is. Maybe James is the one in a billion (or trilion, etc...) luck bucket?

- Andrew

www.pokerstove.com

05-11-2005 , 04:55 AM

#277

rigoletto

veteran

Join Date: Sep 2002 Posts: 2,349

Quote:

Math is hard. The numbers don't say there's a 95% chance he's a winner. They say something else entirely.

Could you expand on this? I would like to learn.

05-11-2005 , 05:00 AM

#278

NLSoldier

Carpal \'Tunnel

Join Date: Feb 2004 Posts: 24,252

Quote:

If you like. Of course, the only difference is how big the lottery is. Maybe James is the one in a billion (or trilion, etc...) luck bucket?

- Andrew

www.pokerstove.com

Lets add to James hand total the hand totals of 10 or 20 other top notch 2+2ers who share similar stats and winrates. Average of say 500,000 hands apiece, will give us 5 million hands.

I think you are going to need to come up with a number bigger than trillion.

05-11-2005 , 05:02 AM

#279

TimM

Carpal \'Tunnel

Join Date: Jan 2004 Posts: 9,705

Can we get his real SD/100. All the calculations in this thread have been based on estimates and assumptions for it.

05-11-2005 , 05:06 AM

#280

Andrew Prock

journeyman

Join Date: Sep 2002 Posts: 399

Quote:

Math is hard. The numbers don't say there's a 95% chance he's a winner. They say something else entirely.

Could you expand on this? I would like to learn.

There are essentially two numbers to consider, the win rate and variance. Of course, JV hasn't actually supplied the variance, so even the basic results are a bit tenuous. UprightCreature posted some very basic numbers which say that getting DERB's results over that stretch for a break even player is maybe a one in a million event.

Now consider the fact that DERB is not alone in these achievements. We aren't talking about some guy that just got lucky.

- Andrew

www.pokerstove.com

05-11-2005 , 05:09 AM

#281

Andrew Prock

journeyman

Join Date: Sep 2002 Posts: 399

Quote:

Maybe James is the one in a billion (or trilion, etc...) luck bucket?

Please note the well placed "etc" in my post.

- Andrew

www.pokerstove.com

05-11-2005 , 05:22 AM

#282

NLSoldier

Carpal \'Tunnel

Join Date: Feb 2004 Posts: 24,252

Quote:

Now consider the fact that DERB is not alone in these achievements. We aren't talking about some guy that just got lucky.

Usually before one considers a certain fact they are presented with some sort of evidence that what they are being told is indeed a fact and not some worthless BS being spouted by some clown on an internet forum.

05-11-2005 , 05:26 AM

#283

Andrew Prock

journeyman

Join Date: Sep 2002 Posts: 399

Quote:

Now consider the fact that DERB is not alone in these achievements. We aren't talking about some guy that just got lucky.

At last we agree! That's been my point all along. Of course, the difficult part here is determining exactly which BS is worthless, and which isn't.

- Andrew

www.pokerstove.com

05-11-2005 , 08:58 AM

#284

J_V

Pooh-Bah

Join Date: Sep 2002 Posts: 4,875

Fair enough. Finally a post that makes a sense.

05-11-2005 , 09:59 AM

#285

sxz18

journeyman

Join Date: Jun 2004 Posts: 213

I want to watch this guy play and "work his magic". What's his PP screename???

05-11-2005 , 10:07 AM

#286

RicktheRuler

grinder

Join Date: Jul 2004 Posts: 642

Please don't do this. Prock, you are trying to get these guys all worked up.

05-11-2005 , 10:08 AM

#287

RicktheRuler

grinder

Join Date: Jul 2004 Posts: 642

I am going to take a shot in the dark here. I think it is........... DERB !!!!!!!!!

05-11-2005 , 10:22 AM

#288

PokerMike

journeyman

Join Date: Nov 2004 Posts: 226

99.99% confidence;

SD/100 Lower Bound Upper Bound
20 0.04 5.14
21 -0.09 5.26
22 -0.22 5.39
23 -0.34 5.52
24 -0.47 5.65
25 -0.6 5.77

I think with his numbers the lotto argument is very valid, even if you tweak his SD and confidence a few points either way.

05-11-2005 , 10:24 AM

#289

I Play 2 Ski

grinder

Join Date: Sep 2004 Posts: 678

Quote:

I am going to take a shot in the dark here. I think it is........... DERB !!!!!!!!!

dood your like psychic

05-11-2005 , 03:03 PM

#290

Justin A

Carpal \'Tunnel

Join Date: May 2004 Posts: 9,010

Quote:

Even considering how high his BB/100 is?

I just learned about confidence intervals in a class. I should be able to figure this out. Im gonna try anyways.

edit-ok nm its already been done, i think this is what you are talking about anyways.

Confidence intervals don't correctly predict the likelihood of this player being a winner, since this player was chosen because of the abnormality of his winrate.

05-11-2005 , 03:49 PM

#291

J_V

Pooh-Bah

Join Date: Sep 2002 Posts: 4,875

http://www.myfreeweb.nl/vida_guerra/vida_guerra05.htm

http://www.myfreeweb.nl/vida_guerra/vida_guerra16.htm

http://www.myfreeweb.nl/vida_guerra/vida_guerra35.htm

http://www.myfreeweb.nl/vida_guerra/vida_guerra34.htm

NSFW for all whatever that means.

DERB has been catching all day today. What a lucksack!

05-11-2005 , 03:56 PM

#292

J_V

Pooh-Bah

Join Date: Sep 2002 Posts: 4,875

Quote:

Confidence intervals don't correctly predict the likelihood of this player being a winner, since this player was chosen because of the abnormality of his winrate.

Finally, someone who gets it.

05-11-2005 , 04:17 PM

#293

Catt

veteran

Join Date: Jan 2005 Posts: 3,245

Quote:

PS - calling Andrew stupid is silly, he clearly is not and makes your position look weak.

Hmm, you must have missed this thread... DERB

I offer no opinion on DERB, the arguments being presented on whether DERB is a lottery winner or a quirky but winning player, or the persons involved in the discussion.

I only offer the opinion that this is the best post on 2+2 in quite some time.

05-11-2005 , 04:19 PM

#294

Lawrence Ng

veteran

Join Date: Dec 2003 Posts: 3,212

Quote:

And what you don't understand is that I'm saying I understand how unlikely it is, according to the math, that he is not actually playing a winning style. That is the whole point of the discussion.

James I'm still surprised you don't believe there some form of cheating here with DERB's style.

Lawrence

05-11-2005 , 04:46 PM

#295

sxz18

journeyman

Join Date: Jun 2004 Posts: 213

Um, his party SN isn't DERB. I know this for two reasons.

1. If it was DERB, everyone in this forum wouldn't be all secretive about who this person was. I mean look at what they say.."Are we talking about the same player?? I've seen him too! Blah blah blah". If his name was DERB, they'd just say "Yeah, I see DERB win all the time"

2. I tried adding DERB to my buddylist on PP and the ID doesn't exist.

I win.

What's his PP sn?

05-11-2005 , 05:08 PM

#296

mikelow

veteran

Join Date: Sep 2002 Posts: 2,277

You can't kill it indeed.

Is it just possible his approach could be a winning one?

05-11-2005 , 05:23 PM

#297

Justin A

Carpal \'Tunnel

Join Date: May 2004 Posts: 9,010

Quote:

It's great you went through all that deductive reasoning, but maybe you should have read the original post in the thread where he says DERB stands for "don't ever run bad."

05-11-2005 , 05:48 PM

#298

rigoletto

veteran

Join Date: Sep 2002 Posts: 2,349

Quote:

Confidence intervals don't correctly predict the likelihood of this player being a winner, since this player was chosen because of the abnormality of his winrate.

We are not talking confidence interval between a population of players, but confidence interval for a 100K sample of hands where each member is 100 hands!!!!!!!!!!!!