How to calc true odds

[ilovepizza82]

Hi,

Could anyone help me to calc true odds having this type of data ?
Any formula, hints, anything ?

LEGEND:
trp1 - the trap dog was in the last race
trp2 - the trap dog was in the race before the last etc.

fin1 - finnish position in the last race

notrial1 - run time for the dog in the last race

dist1 - distance in the last race

[Derek27]

Hi,

Could anyone help me to calc true odds having this type of data ?
Any formula, hints, anything ?

LEGEND:
trp1 - the trap dog was in the last race
trp2 - the trap dog was in the race before the last etc.

fin1 - finnish position in the last race

notrial1 - run time for the dog in the last race

dist1 - distance in the last race

It's not possible to calculate the "true odds" for any sporting event. If it was it would be impossible to win as nobody would offer the true odds.

[wearthefoxhat]

Hi,

Could anyone help me to calc true odds having this type of data ?
Any formula, hints, anything ?

LEGEND:
trp1 - the trap dog was in the last race
trp2 - the trap dog was in the race before the last etc.

fin1 - finnish position in the last race

notrial1 - run time for the dog in the last race

dist1 - distance in the last race

This may get you started. (Something I've used in the past)

Daily191023-NEW.xlsx

[ShaunWhite]

Could anyone help me to calc true odds having this type of data ?
Any formula, hints, anything ?

What you need is a large amount of historic price data.

You make the assumption that for the purposes of this, the market has been correct.

You take your stats and for each permutation you find all the similar dogs in your data and take their average price as being your model price for that combination of parameters.

If only a small sample match, and with so many permutations there often won't be many, then you can use fancy sounding techniques like Synthetic Minority Over-sampling Technique (SMOTE) to balance the data. But there's lots of ways to smooth your data if certain situations aren't well represented.

You now have a model price for each combination/permutation of parameters, or at least points you can extrapolate from.

For each race, you run every selection through your model to get your target price. Then back or lay whenever you see prices at better than your price + your margin and/or continually offer at your price +margin.

.... That's the very simplified version, but it's the basic idea that underpins most price modelling from fundamentals.

[ShaunWhite]

It's not possible to calculate the "true odds" for any sporting event. If it was it would be impossible to win as nobody would offer the true odds.

You know all this but to expand on it,
That's why models aren't shared. But as you suggest model prices are a range of prices which are normally distributed around a centre price, not just a single point. And models vary. So there's always some push pull from people who are prepared to stray further from the centre price and there always will be.

It's not just that it can't be done, it's that you wouldn't know even if you had done it. The true odds of a single event are unknowable, but with some probabilities more likely than others. If that wasn't the case then 100/1 horses would never win.

[MemphisFlash]

Hi,

Could anyone help me to calc true odds having this type of data ?
Any formula, hints, anything ?

LEGEND:
trp1 - the trap dog was in the last race
trp2 - the trap dog was in the race before the last etc.

fin1 - finnish position in the last race

notrial1 - run time for the dog in the last race

dist1 - distance in the last race

This may get you started. (Something I've used in the past)


Daily191023-NEW.xlsx

Where are you getting all that data from?

[wearthefoxhat]

Hi,

Could anyone help me to calc true odds having this type of data ?
Any formula, hints, anything ?

LEGEND:
trp1 - the trap dog was in the last race
trp2 - the trap dog was in the race before the last etc.

fin1 - finnish position in the last race

notrial1 - run time for the dog in the last race

dist1 - distance in the last race

This may get you started. (Something I've used in the past)


Daily191023-NEW.xlsx

Where are you getting all that data from?

The data was from the original sheet posted by the OP, I just added the extra tab showing a quick/easy way of calculating a rating/price using distance and times.

True Odds are never truly known, however, if your model takes into account variables that are weighted by importance (and tested), produce a rating and a tissue price, that could amount to true odds and you back/lay as many in the race over 300+ races a day.

[ShaunWhite]

The simple version..... (but simple obviously doesn't mean accurate otherwise there wouldn't be complex methods)

Q. In horse or greyhound racing, how do you calculated the odds for each selection based on a rating figure.

ChatGPT
A. Calculating odds from ratings is an intricate process, often involving a blend of mathematical models and subjective judgment. However, a simple and common approach to convert ratings into odds is the "probabilistic method". Here's a step-by-step guide for this method using horse or greyhound racing as an example:

Calculate the Sum of Ratings:
Sum the ratings of all the horses or greyhounds in the race.

For example, if you have 3 horses with ratings of 100, 80, and 60, the sum would be 240.

Calculate the Probability for Each Horse:
Divide each horse's individual rating by the sum of the ratings.

For the horse rated 100 in our example:
Probability = Horse's Rating / Sum of Ratings
= 100 / 240
= 0.4167 or 41.67%

Convert Probability to Decimal Odds:
Decimal odds = 1 / Probability

For the horse rated 100:
Decimal Odds = 1 / 0.4167
= 2.40

Convert Decimal Odds to Fractional Odds (if desired):
Subtract 1 from the decimal odds, and then express the result as a fraction.

For the horse rated 100:
Fractional Odds = (2.40 - 1)
= 1.40 or 7/5

So, the odds for the horse would be expressed as 7/5.

Repeat steps 2-4 for each horse or greyhound in the race.

It's important to note that this method provides a purely mathematical representation of odds based on ratings. In the real world, other factors like public perception, bookmaker margins, and external influences might alter the actual odds offered on selections. Furthermore, ratings themselves can be influenced by numerous factors like recent performance, track conditions, jockey/trainer statistics, and more.

[wearthefoxhat]

Here some considerations (17 on the list so far) that could go towards producing a rating.

Weighting each one and ranking them is one way to get a rating for each runner, produce a rating, calculate "true odds", back/lay all runners that are over/under the ratings on all tracks, day in day out, jobs a goodun.

GreysRtg.png

[ilovepizza82]

Hi,

Could anyone help me to calc true odds having this type of data ?
Any formula, hints, anything ?

LEGEND:
trp1 - the trap dog was in the last race
trp2 - the trap dog was in the race before the last etc.

fin1 - finnish position in the last race

notrial1 - run time for the dog in the last race

dist1 - distance in the last race

This may get you started. (Something I've used in the past)


Daily191023-NEW.xlsx

Thank you Mr. Fox, it looks decent.

[ilovepizza82]

Could anyone help me to calc true odds having this type of data ?
Any formula, hints, anything ?

What you need is a large amount of historic price data.

You make the assumption that for the purposes of this, the market has been correct.

You take your stats and for each permutation you find all the similar dogs in your data and take their average price as being your model price for that combination of parameters.

If only a small sample match, and with so many permutations there often won't be many, then you can use fancy sounding techniques like Synthetic Minority Over-sampling Technique (SMOTE) to balance the data. But there's lots of ways to smooth your data if certain situations aren't well represented.

You now have a model price for each combination/permutation of parameters, or at least points you can extrapolate from.

For each race, you run every selection through your model to get your target price. Then back or lay whenever you see prices at better than your price + your margin and/or continually offer at your price +margin.

.... That's the very simplified version, but it's the basic idea that underpins most price modelling from fundamentals.

Yea...i think im gonna need a litte more time to digest what you wrote...

[ilovepizza82]

The simple version..... (but simple obviously doesn't mean accurate otherwise there wouldn't be complex methods)

Q. In horse or greyhound racing, how do you calculated the odds for each selection based on a rating figure.

ChatGPT
A. Calculating odds from ratings is an intricate process, often involving a blend of mathematical models and subjective judgment. However, a simple and common approach to convert ratings into odds is the "probabilistic method". Here's a step-by-step guide for this method using horse or greyhound racing as an example:

Calculate the Sum of Ratings:
Sum the ratings of all the horses or greyhounds in the race.

For example, if you have 3 horses with ratings of 100, 80, and 60, the sum would be 240.

Calculate the Probability for Each Horse:
Divide each horse's individual rating by the sum of the ratings.

For the horse rated 100 in our example:
Probability = Horse's Rating / Sum of Ratings
= 100 / 240
= 0.4167 or 41.67%

Convert Probability to Decimal Odds:
Decimal odds = 1 / Probability

For the horse rated 100:
Decimal Odds = 1 / 0.4167
= 2.40

Convert Decimal Odds to Fractional Odds (if desired):
Subtract 1 from the decimal odds, and then express the result as a fraction.

For the horse rated 100:
Fractional Odds = (2.40 - 1)
= 1.40 or 7/5

So, the odds for the horse would be expressed as 7/5.

Repeat steps 2-4 for each horse or greyhound in the race.

It's important to note that this method provides a purely mathematical representation of odds based on ratings. In the real world, other factors like public perception, bookmaker margins, and external influences might alter the actual odds offered on selections. Furthermore, ratings themselves can be influenced by numerous factors like recent performance, track conditions, jockey/trainer statistics, and more.

Wow. This loooks quite easy to do. Ill try that one as well. Thanks !

[Archery1969]

What you could do is:

- Every dog in the race has run >= 5 times
- Get the CHESTER rating for each dog/race in the past and take an average.
- Repeat for each dog in the race
- Then do what Shaun suggested for working out true ish odds.

If the BF back price is > than your odds + margin then back it and if BF lay price < than your odds - margin then lay it.

Repeat for every qualifying race over a month and see what your P/L is.

[ShaunWhite]

Wow. This loooks quite easy to do. Ill try that one as well. Thanks !

Easy doesn't mean correct. Remember a decent edge might be 2 or 3% so that's how good your evaluation needs to be.

Also remember if you're using fundamentals to create your book you're back to square one if you change sports. However if you use fair price calcs based on market data instead then you have a good chance your algo will work on multiple sports, that's my situation. One calc that works unchanged on several sports based on market microdynamics which will always be present, and so far have ticked along for about 5yrs now.

There's lots of ways to do it. None of it is easy though otherwise all these experienced guys giving you advice would be sitting on a beach not at their desk.

Have you considered ML? It's 2024 soon. What's your situation with data for developing your model and backtesting it?

[ShaunWhite]

Yea...i think im gonna need a litte more time to digest what you wrote...

That's just the basic overview. The reality is tricky. I know one or two full timers with outstanding red brick qualifications who've spent years on it.... But make a fortune, 400+ on UK dogs and then more on the bigger stuff. You won't do that, very few will, but if you educate yourself then you can certainly do well.