Two Plus Two Publishing LLC Two Plus Two Publishing LLC
 

Go Back   Two Plus Two Poker Forums > >

Notices

Sports Betting Discussions related to wagering on sporting events.

Reply
 
Thread Tools Display Modes
Old 09-13-2017, 11:25 AM   #501
Iowa!
old hand
 
Iowa!'s Avatar
 
Join Date: Jun 2009
Posts: 1,289
Re: Mayweather v. McGregor

One of the many differences between having good models and doing calculations in your (limited) head is that good models will tell you when your edge is gone. At that point we simply pack up our **** and head to the beach.
Iowa! is offline   Reply With Quote
Old 09-13-2017, 02:41 PM   #502
andr3w321
adept
 
Join Date: Jul 2007
Location: Seattle
Posts: 1,113
Re: Mayweather v. McGregor

Quote:
Originally Posted by Derek123 View Post
I have a question about kelly betting and just thought Id post it here. Lets say I cap a bet at 60% and the book is laying -120. After a big sample of making similar type bets my win % is ~57%. So I have an edge but likely not as big as my initial calculation. I am unsure what to put in as my odds for determining the kelly stake. It seems to me like it should be 57% but I need a big sample to get to that number. Is it safe to use 60% but then bet like 1/4 Kelly?

It is a pretty common spot for me so any help would be great.
I would trust that the 57% # is more accurate. One of the reasons full kelly is super dumb is it doesn't allow for calculation errors like this, but 1/2 kelly does. When you bet the "60%" edge at 1/2 kelly at -120 that's 6% of your roll. What you're actually betting is 6% on a 57% edge at -120 which is 1.1 kelly so yes you probably want to lower your fractional kelly since you're increasing variance and reducing bankroll growth rate with 1.1 kelly.

Quote:
Originally Posted by Like View Post
Have any of you guys heard of Long Term Capital Management? I bet those guys thought they were geniuses and talked down to everyone too. How much did they lose in a few months? Like 5 billion?
These morons were not kelly staking. It's impossible to go broke betting kelly. They were using risk of ruin models. So assuming I make 10 bets with 10% of my bankroll each and each bet only has a 10% chance of going to zero I only have a minuscule chance of bustoville and I can just leverage myself 100x and everything should be fine right? The problem is if you're making correlated bets and don't know it like they were. When one bet goes to zero it's likely the others will go to zero too. So instead of a 0.1^10 chance of going bust it's more like 9%/year. 9%/year chance of going bust and you'll only last a few years like they did.
andr3w321 is offline   Reply With Quote
Old 09-13-2017, 05:37 PM   #503
Derek123
adept
 
Join Date: Jun 2005
Posts: 970
Re: Mayweather v. McGregor

Quote:
Originally Posted by andr3w321 View Post
I would trust that the 57% # is more accurate. One of the reasons full kelly is super dumb is it doesn't allow for calculation errors like this, but 1/2 kelly does. When you bet the "60%" edge at 1/2 kelly at -120 that's 6% of your roll. What you're actually betting is 6% on a 57% edge at -120 which is 1.1 kelly so yes you probably want to lower your fractional kelly since you're increasing variance and reducing bankroll growth rate with 1.1 kelly.
ok cool. thanks for clarifying
Derek123 is offline   Reply With Quote
Old 09-14-2017, 05:55 PM   #504
Soxxy
enthusiast
 
Soxxy's Avatar
 
Join Date: Jul 2017
Location: Vegas
Posts: 66
Re: Mayweather v. McGregor

All you have to do to keep your "edge" here is wait for the next time the best boxer of all time goes 12 rounds with a viral marketing hype job from a totally different sport that has never been in the ring before, profit.
Soxxy is offline   Reply With Quote
Old 09-14-2017, 06:36 PM   #505
Iowa!
old hand
 
Iowa!'s Avatar
 
Join Date: Jun 2009
Posts: 1,289
Re: Mayweather v. McGregor

Quote:
Originally Posted by Soxxy View Post
All you have to do to keep your "edge" here is wait for the next time the best boxer of all time goes 12 rounds with a viral marketing hype job from a totally different sport that has never been in the ring before, profit.
My monthly nut is 30k. What can I do in the meantime?
Iowa! is offline   Reply With Quote
Old 09-15-2017, 05:58 PM   #506
Fubster
Carpal \'Tunnel
 
Fubster's Avatar
 
Join Date: Apr 2007
Posts: 8,126
Re: Mayweather v. McGregor

Sit down with your spouse or significant other and make a household budget
Fubster is offline   Reply With Quote
Old 09-16-2017, 05:54 AM   #507
Mihkel05
Pooh-Bah
 
Join Date: Jul 2015
Posts: 3,748
Re: Mayweather v. McGregor

Quote:
Originally Posted by Derek123 View Post
ok cool. thanks for clarifying
I disagree with almost everything he wrote to you btw. I would venture that virtually everyone who is an expert (n00b/Iowa/etc) would say the same thing. Betting Kelly fractions doesn't magically make you immune to estimation errors, it makes you less sensitive to over betting errors. (Especially >2x Kelly where +EV turns -EG.) Obviously someone who reads ****ing William Poundstone for advice on a mathematical optimization isn't gonna have any concept of what is actually going on, much less be able to articulate the finer points in any competent fashion.

I generally disagree with literally everything he writes on Kelly since he barely understands the topic tbf.

Quote:
Originally Posted by Fubster View Post
Sit down with your spouse or significant other and make a household budget
Do you think he just came to 30k via the same way most people ITT came to a bet size on Mayweather?
Mihkel05 is online now   Reply With Quote
Old 09-16-2017, 12:48 PM   #508
Fubster
Carpal \'Tunnel
 
Fubster's Avatar
 
Join Date: Apr 2007
Posts: 8,126
Re: Mayweather v. McGregor

It's important to have a household budget to begin saving for retirement.
Fubster is offline   Reply With Quote
Old 09-16-2017, 01:13 PM   #509
Sabaneta
adept
 
Join Date: May 2009
Posts: 951
Re: Mayweather v. McGregor

Quote:
Originally Posted by Fubster View Post
It's important to have a household budget to begin saving for retirement.
Retirement fund = bankroll
Sabaneta is offline   Reply With Quote
Old 09-16-2017, 06:18 PM   #510
MisterRodriguez
enthusiast
 
Join Date: May 2017
Posts: 58
Re: Mayweather v. McGregor

I was going to buy this week Fortune Formula.You think its not a good read,and it might offer misleading information?
MisterRodriguez is offline   Reply With Quote
Old 09-16-2017, 06:57 PM   #511
WateryBoil
Pooh-Bah
 
WateryBoil's Avatar
 
Join Date: Jul 2014
Location: And mine are long and sharp my Lord
Posts: 4,094
Re: Mayweather v. McGregor

wut does that even mean?
WateryBoil is offline   Reply With Quote
Old 09-16-2017, 07:29 PM   #512
Mihkel05
Pooh-Bah
 
Join Date: Jul 2015
Posts: 3,748
Re: Mayweather v. McGregor

Its a good read. It won't make you an expert on KC. See all the posts that have been written. It doesn't even cover some of the most basic issues with how to apply Kelly, much less some of the mathematical issues.

I'd gladly debate anyone, anywhere on Kelly expert vs the book and me just shooting off the cuff.
Mihkel05 is online now   Reply With Quote
Old 09-18-2017, 09:37 PM   #513
Derek123
adept
 
Join Date: Jun 2005
Posts: 970
Re: Mayweather v. McGregor

Quote:
Originally Posted by Mihkel05 View Post
I disagree with almost everything he wrote to you btw. I would venture that virtually everyone who is an expert (n00b/Iowa/etc) would say the same thing. Betting Kelly fractions doesn't magically make you immune to estimation errors, it makes you less sensitive to over betting errors. (Especially >2x Kelly where +EV turns -EG.) Obviously someone who reads ****ing William Poundstone for advice on a mathematical optimization isn't gonna have any concept of what is actually going on, much less be able to articulate the finer points in any competent fashion.

I generally disagree with literally everything he writes on Kelly since he barely understands the topic tbf.
Any thoughts on my post that he responded to?
Derek123 is offline   Reply With Quote
Old Today, 09:00 AM   #514
Mihkel05
Pooh-Bah
 
Join Date: Jul 2015
Posts: 3,748
Re: Mayweather v. McGregor

Your problem is that your modeling is systematically inaccurate. Kelly isn't appropriate when you're always overshooting 3%. (Maybe people boil in a margin to account for this. I don't because I'm a wizard or something.)
Mihkel05 is online now   Reply With Quote

Reply
      

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Forum Jump


All times are GMT -4. The time now is 09:39 AM.


Powered by vBulletin®
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © 2008-2010, Two Plus Two Interactive
 
 
Poker Players - Streaming Live Online