new idea on prices. why do we bother?

[duffman]

I think sgtsunshine that people know just as much as the bookies know however long before kick-off it is and thats what really matters. And I also think any decent attempt at a system would keep it simple and store what other peoples opinions were on certain factors (so you wouldnt need to know all factors yourself). Not sure what you mean by quantifying things, I thought my whole previous post tried to explain that odds are created using subjective opinions...do you not agree?

So I agree with muppet that if you used the interpretation of a large crowd of people it would more accurate than that of a small team of odds compilers.

The argument 'value betting' v 'strike rate' betting is pretty much argued black and blue elsewhere....but even if you did believe in 'strike rate' betting you could store results instead of odds and use a system like this to find.."okay, every time Chelsea played when most people thought they would struggle to score goals"...they drew 0-0....or something like that...then you could bet on a draw or 0-0 draw...

[duffman]

"Federer was 1.12 to beat Ferrero, and today he was 1.26 to beat Ferrero in Hamburg. Both Masters series, both on clay. Federer has a little wobble in between and his odds shift quite a lot."

Exactly my idea! So the bookies have the made the change in price under the same circumstances because they believe Federer is 'off form'..nothing else, just their opinion....So if its your opinion that Federer is not in bad form, then you could use a system to find what the price should be (1.12) and then the 1.26 would the be a value bet!

And lads ye cant just write this thing off by saying 'keep things simple'..we're trying to find the holy grail here ha

[muppet77]

well said. a different approach is needed. most punters + bookies use some sort of elo style rating system. why should we?!

[si1965]

I think you also have to acknowledge that there is a distinction between opening price and the eventual starting price - the former being the domain of the compilers where, when errors occur, the shrewd punter (e.g. Shipsupstreets comes to mind) can exploit early price offerings.
The latter price is all due to market pressure and while it can be an adjustment due to an error in the initial price or some new information, it may equally have nothing to do with any of the factors the compilers utilised in the first place e.g. unfounded rumours that abound in the 'next manager' market.

[merlin]

... just as a matter of interest by way of keeping things simple, I had

7 correct homes from 8 selections on my Aussie ratings today Martin,

proofed on the Non European thread and using no values whatsoever.

Early days but encouraging.

[muppet77]

very good, but how do you know that there was no value?

[merlin]

.... with a strike rate so high who cares, one can adopt a number of

different permutations without bothering it's value or not, I would imagine

that should you land a multifold it would not matter one little bit.

[muppet77]

i agree a winner's a winner, but you said that your simple aussie system was not a value one. how do you know this?
keep up the good work.

[PAULM03]

Really interesting read this, keep it up

Originally Posted by merlin

... just as a matter of interest by way of keeping things simple, I had

7 correct homes from 8 selections on my Aussie ratings today Martin,

proofed on the Non European thread and using no values whatsoever.

Early days but encouraging.

Maybe you have a slight misunderstanding of value Merlin. The method of finding 'value' is not important, but by definition if you are a punter in consistent profit then you are finding value.

For instance someone backing the horses who picks his selections purely on instinct and what he reads in the racing post could be just as succesful as someone who takes a completely different approach (like slapdash for instance). Both are finding value (they wouldn't be in profit were thay not) just in completely different ways.

[merlin]

.... I agree with you Paul, it's exactly down to how one defines 'value'.

The value I see is that a particular rating for one team is indeed significant

enough to see that team as a winner. Yet I image there are others here

who view that quite differently.They see one team as being undervalued

by the bookies, therefore their true definition is that as the greater odds

on offer represent value as that team/teams have a greater chance of

winning in relation to their calculations.

[muppet77]

The value I see is that a particular rating for one team is indeed significant enough to see that team as a winner.

so would you be happy with a high strike rate or a high profit?

[merlin]

.... personally speaking Martin it very much once again depends on the

individual's betting patterns.If you are the patient type and only bet

singles, that's where your type of value comes in. But I would very much

prefer a high strike rate, knowing full well that the majority of selections

had been 'landed' and with systematic permutation good profits can be

obtained.

[muppet77]

no way! really?

so if i offered you a team to win at 1.01, who looked immortal, you would back them if they were selected by your sytem as having a high chance of winning, even if their true odds were 1.05?

[traderpaul]

Muppett77, there is no such thing as true odds. If there were there would be no reason for anyone to bet. If Merlins method of defining value says that he should back a 1.01 price then he backs it! Your calculation of value may be 1.05 and you would lay the 1.01. Your calculation of value has no bearing on Merlin (at least it shouldn't have!)

[merlin]

.... thanks traderpaul for a further valid viewpoint.To recap on your

hypothetical question and offer above Martin, my short swift answer on

that would be no! Common sense and judgement are two main ingredients

which ought to be applied with any bet.If my ratings indicated that any

team with odds of 1.01 whether they were stonewall certainties or not,

I would not even bother to entertain or consider as worthy of putting

my hard earned cash on! To entertain odds of 1.01 would be a definite

'no no' as far as I am concerned.

[traderpaul]

Actually I shouldn't have said there would be no reason for anyone to bet if we knew what true odds were. We just wouldn't bet for the potential to profit. People do bet for other reasons they value; entertainment, camaraderie, other touchy feely stuff! People buy lottery tickets, back their "favourite" team, etc.

[muppet77]

obviously 'true odds' exist - it is just that nobody knows them in advance.

[traderpaul]

Yes and leprechauns exist too. Its just that you need to put a dozen pints under your belt before you see one!

[muppet77]

ok, imagine you were building a statistical model of a football game, and knew EVERY factor involved - all the usual home form away form, scoring ability....all the way down to who is playing, motivation, how the weather will change, if that shot that hits the bar will go one way or the other......then you would know the true odds.

this thread is just about testing the theory that says that collective masses are more knowledgeable than a few experts.

i thought that it would be interesting to see how a different approach may work. most people using systems on here are crunching numbers to one degree or another. why? the bookie will be way ahead, and then some with their overround.

that's what we are trying to tackle.

[merlin]

I thought that it would be interesting to see how a different approach may work. most people using systems on here are crunching numbers to one degree or another. why? the bookie will be way ahead, and then some with their overround.

I'm afraid I don't agree with this comment at all Martin speaking entirely from my point of view.As far as true odds go, will we ever be sure that we
really know all the facts likely to affect a result,there will always be something unforseen cropping up like the awarding of a penalty kick or the turning down of one when 50,000+ swear blindly that the ref was wrong!Maybe you can get near,but never 100% sure.