True odds calculation

[firlandsfarm]

If a two horse race is priced at

No 1 - Evens
No 2 - Evens

Then that is true odds? A 100% book??

However, we all know this never happens.

I'm reading 'true odds' as being the true chance of a selection winning in which case collectively they would always form a 100% book. But a 100% book does not suggest the odds for each individual selection is 'true'. Your example would only be true if the 2 runners had a perfectly equal chance of winning ... that's what's rarely so. A 100% book may represent true odds but it may not. An over/under round book may represent the true odds for some of the selections but never for all.

We normally refer to a 100% book as being an 'efficient book'. So does that mean to be an 'efficient book' simply requires the sum of the probabilities of the selections equals 1 and the 'trueness' of those odds for each selection is irrelevant?

[wearthefoxhat]

If a two horse race is priced at

No 1 - Evens
No 2 - Evens

Then that is true odds? A 100% book??

However, we all know this never happens.

I'm reading 'true odds' as being the true chance of a selection winning in which case collectively they would always form a 100% book. But a 100% book does not suggest the odds for each individual selection is 'true'. Your example would only be true if the 2 runners had a perfectly equal chance of winning ... that's what's rarely so. A 100% book may represent true odds but it may not. An over/under round book may represent the true odds for some of the selections but never for all.

We normally refer to a 100% book as being an 'efficient book'. So does that mean to be an 'efficient book' simply requires the sum of the probabilities of the selection equals 1 and the 'trueness' of those odds for each selection is irrelevant?

The 2 @ Evens (100% book) is the starting point. The "true odds" are then dependent on other attributes that each selection has achieved and if there is potential to improve, or even regress. This is where the more quality data/information will give you an edge.

One factor to also consider is how the book is put together by the odds compiler. Nowadays, the top end of the market is more compressed, therefore reducing value, and the 100% book + overround is spread out to the remaining runners in the race.

This, however, then creates an opportunity for the bigger priced runners that now may offer value. The good news is, thanks to odds-checker, and freely available data/information (some paid for), there are pockets of value not accounted for by the bookies/exchanges.

Also, horse racing by it's nature, does not offer a complete picture, as there is so much information we are not privy to, unlike something like tic-tac-toe.

[wearthefoxhat]

Just to add to the last post, here's an example of "ingredients" that can go into baking the "true odds" cake.

https://www.sbo.net/horse-racing/factor ... rformance/

[Realrocknrolla]

If a two horse race is priced at

No 1 - Evens
No 2 - Evens

Then that is true odds? A 100% book??

However, we all know this never happens.

I'm reading 'true odds' as being the true chance of a selection winning in which case collectively they would always form a 100% book. But a 100% book does not suggest the odds for each individual selection is 'true'. Your example would only be true if the 2 runners had a perfectly equal chance of winning ... that's what's rarely so. A 100% book may represent true odds but it may not. An over/under round book may represent the true odds for some of the selections but never for all.

We normally refer to a 100% book as being an 'efficient book'. So does that mean to be an 'efficient book' simply requires the sum of the probabilities of the selection equals 1 and the 'trueness' of those odds for each selection is irrelevant?

The 2 @ Evens (100% book) is the starting point. The "true odds" are then dependent on other attributes that each selection has achieved and if there is potential to improve, or even regress. This is where the more quality data/information will give you an edge.

One factor to also consider is how the book is put together by the odds compiler. Nowadays, the top end of the market is more compressed, therefore reducing value, and the 100% book + overround is spread out to the remaining runners in the race.

This, however, then creates an opportunity for the bigger priced runners that now may offer value. The good news is, thanks to odds-checker, and freely available data/information (some paid for), there are pockets of value not accounted for by the bookies/exchanges.

Also, horse racing by it's nature, does not offer a complete picture, as there is so much information we are not privy to, unlike something like tic-tac-toe.

Least I found the start

is (fried)

[andy28]

As Peter says pre off is driven by opinion, the edge comes because some put more effort into forming their opinion. .

[ShaunWhite]

As Peter says pre off is driven by opinion, the edge comes because some put more effort into forming their opinion. .

All markets are driven by supply and demand.

[wearthefoxhat]

Least I found the start

is (fried)

TBH it's easy to get lost down the rabbit hole of data/information.

I tend to streamline what I feel has a greater preponderance and weight it accordingly.

Who can forget Agnes Haddock....(winner of the Scoop 6 back in the day for £2)

agness.png

[firlandsfarm]

As Peter says pre off is driven by opinion, the edge comes because some put more effort into forming their opinion. .

All markets are driven by supply and demand.

There's not much difference between opinion and supply and demand ... supply and demand is the application of opinion.

[firlandsfarm]

One aspect of seeking the holy grail of true odds (probability) hasn't been discussed and I don't have the solution, I'm just laying it on the table. Having gathered all the information together either as exclusive facts or bundled into 'ratings' you then have to convert that information into probabilities (odds). I have been searching for years to find the formula for that final step. I don't think there is one! So it all comes down to the interpretation of the data, not the data itself. You can piece together the best, most informative, most relevant set of data and screw it up at the last step!

[ShaunWhite]

As Peter says pre off is driven by opinion, the edge comes because some put more effort into forming their opinion. .

All markets are driven by supply and demand.

There's not much difference between opinion and supply and demand ... supply and demand is the application of opinion.

Not much difference granted but a difference. Option alone only works as a constraint where the hive all have the same stake/effect. It would only take 1 rogue option with deep pockets to affect supply and demand which is what actually moves prices.

Besides, supply and demand is easier to use in an equation than 'opinion'

[napshnap]

One aspect of seeking the holy grail of true odds (probability) hasn't been discussed and I don't have the solution, I'm just laying it on the table. Having gathered all the information together either as exclusive facts or bundled into 'ratings' you then have to convert that information into probabilities (odds). I have been searching for years to find the formula for that final step. I don't think there is one! So it all comes down to the interpretation of the data, not the data itself. You can piece together the best, most informative, most relevant set of data and screw it up at the last step!

Really? What's so hard to gather data, then "streamline what you feel has a greater preponderance and weight it accordingly." (like wearthefoxhat suggests) then sum weights for each runner and finally calc odds from it?
Obviously the trickiest and most time consuming part of that "not so hard process" is the 2nd part, but the last one is a simple math.

I'm not saying that this will be "true odds" but at least it will be your odds, your starting point for experimentation!

[decomez6]

Country
Track
Going
Racetype just a few variables , how do you define 'TRUE' ?
and what about those horses with no formlines ?
Distance
Class
Time
Stall
Horse
Age
Pace
Weight
Jockey
Trainer
SP Fav
Industry SP
Betfair SP
IP Min
IP Max
Pred ISP
Place
Winning Distance
Runners

few horse racing variables.PNG

this is not an exhaustive list , but its part of the pricing matrix.
so its already reflected in the opening prices .

And also , if by TRUE' you mean the actual winning odds after the race is over , you fall into the GAMBLERS FALLACY trap !.

[wearthefoxhat]

One aspect of seeking the holy grail of true odds (probability) hasn't been discussed and I don't have the solution, I'm just laying it on the table. Having gathered all the information together either as exclusive facts or bundled into 'ratings' you then have to convert that information into probabilities (odds). I have been searching for years to find the formula for that final step. I don't think there is one! So it all comes down to the interpretation of the data, not the data itself. You can piece together the best, most informative, most relevant set of data and screw it up at the last step!

Really? What's so hard to gather data, then "streamline what you feel has a greater preponderance and weight it accordingly." (like wearthefoxhat suggests) then sum weights for each runner and finally calc odds from it?
Obviously the trickiest and most time consuming part of that "not so hard process" is the 2nd part, but the last one is a simple math.

I'm not saying that this will be "true odds" but at least it will be your odds, your starting point for experimentation!

Yeah, I'd say that sums it up.

Not saying the following is the answer, but there are some interesting aspects to it that can be converted to excel.

https://www.informracing.com/betting-tissue-tool-1/

[wearthefoxhat]

Country
Track
Going
Racetype just a few variables , how do you define 'TRUE' ?
and what about those horses with no formlines ?
Distance
Class
Time
Stall
Horse
Age
Pace
Weight
Jockey
Trainer
SP Fav
Industry SP
Betfair SP
IP Min
IP Max
Pred ISP
Place
Winning Distance
Runners

few horse racing variables.PNG


this is not an exhaustive list , but its part of the pricing matrix.
so its already reflected in the opening prices .

And also , if by TRUE' you mean the actual winning odds after the race is over , you fall into the GAMBLERS FALLACY trap !.

Big old list there Dec!

Some it will cross over and duplicate, but the key elements are all in there.

I'd say racetype selection, class of race, number of runners, are the initial filters that are sometimes not given much thought to.

Personally, I enjoy the whole process of data assessment/input and reviewing the results/market.

[firlandsfarm]

Really? What's so hard to gather data, then "streamline what you feel has a greater preponderance and weight it accordingly." (like wearthefoxhat suggests) then sum weights for each runner and finally calc odds from it?
Obviously the trickiest and most time consuming part of that "not so hard process" is the 2nd part, but the last one is a simple math.

napahnap I made no reference to the ease or difficulty of gathering the data, nor the 'streamlining' of it etc. because I was majoring on the step after that ... the conversion of that data into odds by calculation. Note, calculation ... not guessing, not assessing, not approximating for later fine tuning but hard +,-,/,x and other more complicated variations of those functions. I'm happy to stand corrected, could you publish the formula for calculating the odds . As I said I have never seen a formula that would stand up to examination by a mathematician and sorry wearthefoxhat and it may be that my understanding of your approach is out of date but the calculation I saw as being your approach some time ago was not mathematically sound for creating true odds for comparison.

I'm not saying that this will be "true odds" but at least it will be your odds, your starting point for experimentation!

But this thread is titled "True odds calculation" so I'm referring to a calculation of the true odds