Forums

General Betting

Welcome to Live View – Take the tour to learn more
Start Tour
There is currently 1 person viewing this thread.
Rowan86
29 Aug 10 13:48
Joined:
Date Joined: 13 Jan 05
| Topic/replies: 4,169 | Blogger: Rowan86's blog
I would expect 95% of people on here to understand the value concept with Betfair being a place for sharp minds and all that, but for what it's worth (probably not much), here's my proof of it:

Value Concept:

Where

d = decimal odds
pr = Decimal probability of bet winning

d represents value if 1/d > pr

Proof Of Value Concept:

n = number of bets made
x = amount staked on each bet
p = total payout
d = decimal odds
pr = decimal probability of bet winning

Then,

nx = Total amount wagered

and

P = pr*d*nx

Let x = arbitrary value where x >0,

Where 1/d > pr,

as n tends to infinty, (p - nx) tends to infinity.

Where 1/d < pr,

as n tends to infinty, (p - nx) tends to minus infinity.
Pause Switch to Standard View Proof Of Value Concept
Show More
Loading...
Report kenilworth August 29, 2010 3:39 PM BST
Rowan86     


I would expect 95% of people on here to understand the value concept


If that were the case, Betfair would not exist.
Report Rowan86 August 29, 2010 3:48 PM BST
Maybe they're just bad at quantifying value?
Report kenilworth August 29, 2010 3:54 PM BST
How can 95% profit meaningfully from 5% ?
Report Rowan86 August 29, 2010 4:01 PM BST
Well, just because they understand that concept, doesn't mean they're going to agree on the prices.

Is your point that if 95% at least understood the concept, that wouldn't allow enough value to be exploited?
Report Lori August 29, 2010 4:49 PM BST
I think many more people understand the concept than profit from it.

You have to be exceptionally daft not to get it, but you can be not daft at all and still lose money.
Report Rowan86 August 29, 2010 4:51 PM BST
I agree Lori.
Report scotbet August 29, 2010 10:55 PM BST
spot on lori
Report FatalStatistics August 30, 2010 12:01 AM BST
I understand and agree Rowan, however i have found that probability is reliably inconsistent ! and therefore calculating pr becomes mind bogglingly difficult. particularly in correct scores where i have chosen to serve the term of my natural life.
Good luck all though.
Report Rowan86 August 30, 2010 12:40 AM BST
Thanks for your post fatal statistics. Happy

I agree that the problem is obviously in trying to quantify pr. Interesting getting your perspective.
Report Rowan86 August 30, 2010 12:40 AM BST
How's it going btw Scotbet? Thanks for posting on this thread. Nice to see you.
Report swift-tuttle August 30, 2010 6:49 AM BST
First thing I look for in these calculations is the probability of winning value. It's used in Kelly.

Is pr calculated? NO
Is pr used in a calculation? YES
In that case, where does the value pr come from? OUT OF THE THEORIST'S HEAD


So the whole thing is not worth the paper it is written on.
Report Optimizer August 30, 2010 7:36 AM BST
You have got the inequalities wrong. Also, where does pr come from ;-)
Report Optimizer August 30, 2010 7:39 AM BST
Whoa, the posts go from newer to older..
Report ijs August 30, 2010 10:15 AM BST
if urealy think this utter garbage is going to work ur a bigger mug than u sound....pathetic
Report Rowan86 August 30, 2010 1:30 PM BST
Damn. I did get the inequalities wrong. FFS.

CORRECTION:

Value Concept:

Where

d = decimal odds
pr = Decimal probability of bet winning

d represents value if pr > 1/d

Proof Of Value Concept:

n = number of bets made
x = amount staked on each bet
p = total payout
d = decimal odds
pr = decimal probability of bet winning

Then,

nx = Total amount wagered

and

P = pr*d*nx

Let x = arbitrary value where x >0,

Where pr > 1/d,

as n tends to infinty, (p - nx) tends to infinity.

Where pr < 1/d,

as n tends to infinty, (p - nx) tends to minus infinity.
Report Rowan86 August 30, 2010 1:35 PM BST
Can't believe I did. Absolute SEETHING
Report Rowan86 August 30, 2010 1:35 PM BST
Can't believe I did that. Absolutely SEETHING.
Report Rowan86 August 30, 2010 1:40 PM BST
lol. Was waiting for a few people to ask 'how do we calculate pr?'
Report Rowan86 August 30, 2010 7:00 PM BST
swift-tuttle Joined: 22 Mar 04
Replies: 902 30 Aug 10 06:49   
First thing I look for in these calculations is the probability of winning value. It's used in Kelly.

Is pr calculated? NO
Is pr used in a calculation? YES
In that case, where does the value pr come from? OUT OF THE THEORIST'S HEAD


So the whole thing is not worth the paper it is written on.


Surely can't be any less meaningful than that pointless 'backing horses to win' thread you started. Laugh
Report swift-tuttle August 30, 2010 7:08 PM BST
Sorry, I seem to have struck a nerve there
Report Rowan86 August 30, 2010 7:15 PM BST
lol. Sorry. You are right, but that wasn't the point of the thread. I just thought I'd write a proof for a bit of fun. There are one or two people who don't understand or haven't thought about the value concept though.
Report swift-tuttle August 30, 2010 7:19 PM BST
I've thought about little else for the last few years now. All the recommendations tell you to go for Kelly but when I try to do that, I find it very difficult to put a value on the probability of the horse winning.
You're right to try and pin it down because I'm a firm believer in value being the key thing for successful punting.
Report Rowan86 August 30, 2010 7:23 PM BST
Yeah, it's a hard working out the pr of the horse winning, but that's the name of the game.
Report FINE AS FROG HAIR August 30, 2010 8:58 PM BST
Don't the market makers work out the pr ( or try to) first inorder to set their opening prices ?
How else do you create a book ?
Sorry if a stupid question.
Report Rowan86 August 30, 2010 9:20 PM BST
Yeah, you're right Fine As Frog Hair. They try and work out the pr. Deffo.
Report The Investor August 30, 2010 10:03 PM BST
I think it's the bookies that work that out more than the market makers. If you can do this with a high degree of accuracy, you have a licence to print money.
Report MIB34 August 30, 2010 10:19 PM BST
Dont  forget  the bookies  can see the order flow...


If  loads  of people start placing bets  on one horse,   they can adjust  there price.
- As well as adjust the prices  of all the horses  in the race.

(thik of the over/under  run_).
Report Rowan86 August 31, 2010 1:44 PM BST
Oh, when you said 'market makers', I thought you meant the bookies.
Report The Investor August 31, 2010 1:50 PM BST
I think there is some overlap, but market makers don't necessarily set the opening prices. The largest one certainly doesn't.
Report mythical prince August 31, 2010 2:01 PM BST
Rowan86 Joined: 13 Jan 05
Replies: 1151 29 Aug 10 13:48   


I would expect 95% of people on here to understand the value concept with Betfair being a place for sharp minds and all that, but for what it's worth (probably not much), here's my proof of it:

Value Concept:

Where

d = decimal odds
pr = Decimal probability of bet winning

d represents value if 1/d > pr

Proof Of Value Concept:

n = number of bets made
x = amount staked on each bet
p = total payout
d = decimal odds
pr = decimal probability of bet winning

Then,

nx = Total amount wagered

and

P = pr*d*nx

Let x = arbitrary value where x >0,

Where 1/d > pr,

as n tends to infinty, (p - nx) tends to infinity.

Where 1/d < pr,

as n tends to infinty, (p - nx) tends to minus infinity.

^ you forgot to add the random anomaly, when value backing in the 4.15 at windsor and getting chinned by a short head, seething with rage and lumping your whole bank on the next at wolves LaughLaugh
Report Rowan86 August 31, 2010 2:19 PM BST
LOOOOOOOOOOL Mythical. Laugh
Report Rowan86 August 31, 2010 2:20 PM BST
The inequalities were the wrong way round in that one you quoted though!
Post Your Reply
<CTRL+Enter> to submit
Please login to post a reply.

Wonder

Instance ID: 13539
www.betfair.com