ODDS ON SELECTIONS TRIAL

[MCLARKE]

There have been a number of comments about never backing odds on horses. I am going to start a trial on these horses.

The basic premise is that these selections should have an edge because :-

1. The favourite-longshot bias. This is an observed phenomenon where bettors tens to overvalue longshots and undervalue favourites.

2. There is a tendency for some bettors to specifically not bet on odds on selections. If a horse is 11/10 they will have a bet but at 10/11 they will not. The bookmakers will be aware of this and therefore they will lengthen the odds on selections and shorten the odds against.

I will pick 1 horse a day where it is forecast odds on and the first show on BET365 at 5pm is odds on. If there is more than 1 selection I will take the one with the lowest forecast odds.

I will post the selection as my nap in the naps competition to avoid having too many threads in At The Races.

I will also keep a record of the price movements to attempt to identify the best time to back these selections.

I will update this thread once a week.

The 1st selections is Bresson in the 2.10 at Yarmouth tomorrow. Forecast odds are 2/11. Opening price at 5.01 on BET365 was 4/9. Interestingly this has already shortened to 2/9.

 

[MCLARKE]

I have purposefully excluded any form study at all in this exercise to try to prove that odds on selections are viable in their own right.

I am sure that any returns can be improved by adding additional criteria.

These include (for flat racing) :-

Exclude the top 3 in the weights. This increases returns by 9%

Irish bred 5%

1st or 2nd last time 3%

More than 6 runners 4%

Not drawn in top 3 3%

Good to Firm or Firm going 4%

Last ran more than 30 days ago 5%

Last ran less than 13 days ago 7%

Exclude 5/6 furlong races 2%

Forecast odds less than 4/5 2%

Male horses 2%

 

 

[Trotter]

Good luck with this !

I've often thought that you must be able to make a profit backing odds on shots but I reckon the stats suggest not !

It's also very dispiriting if a 1/5 shot loses and you know it's going to take 4 or 5 winners to get that stake money back ....... then if one of them loses you're up the creek needing 7 or 8 winners to get back to levels

Be interesting to see how you get on !

 

 

 

 

[Valiant Thor]

12 hours ago, MCLARKE said:

The 1st selections is Bresson in the 2.10 at Yarmouth

Why would that be ?
What betting forecast are you using for your datum?

Royal Parade is the lowest forecast fav (RP) and has the best Fav > 2nd Fav differential (which imo is a better indicator than just being fav)

Meet	time	Horse	F-Fav	2nd F-Fav		Fav	2nd Fav	Differential
Yarm	14.40	Royal Parade	1.33	6.00	>	75%	17%	351%
Brigh	14.20	Inspirited	1.44	5.50	>	69%	18%	282%
Yarm	14.10	Bresson	1.50	4.50	>	67%	22%	200%

ATB
VT

[Trotter]

Just to add to my post above regarding the stats suggesting that you can't make a profit on odds on shots ........ that of course refers to backing all odds-on shots

But of course no one would do that

On any given day there might be six odds on shots and four of them will win and two of them will lose ......... the art is in not backing the two that will lose !

Just like any selection process you have to pick which ones you think will win and avoid the ones you think might lose  .......... and then you have to be right most of the time. 

[MCLARKE]

I use the Race Advisor for my raw data. 

Horse	Forecast Odds
Bresson	1.62
Inspirited	1.73
Royal Parade	1.18

I believe they use the RP as their source data, not sure why they are different to the figures you quote. It doesn't make a lot of difference, I'm only using the data to quickly identify odds on horses.

I obviously made an error in selecting Bresson, for some reason I picked the 1st horse in the list, rather than the lowest odds. However it still fits the criteria of the process.

Interesting point about the fav to 2nd fav differential. This will remove those a lot of those situations where there are 2,3 or 4 runners in the race where there will often be an odds on favourite. The losses in these small field races are 3% higher than the average.

 

[MCLARKE]

7 minutes ago, Trotter said:

Just to add to my post above regarding the stats suggesting that you can't make a profit on odds on shots ........ that of course refers to backing all odds-on shots

But of course no one would do that

I suppose I am trying to prove just that, albeit selecting just 1 per day.

Backing them all at SP results in approximately a 4% loss, I'm hoping to turn this into a profit using BOG.

[MCLARKE]

5 hours ago, Trotter said:

It's also very dispiriting if a 1/5 shot loses and you know it's going to take 4 or 5 winners to get that stake money back ....... then if one of them loses you're up the creek needing 7 or 8 winners to get back to levels

On the other side of the coin, I like to watch my horses win and it will be quite comforting to watch them win most of their races. If you are confident that long term you will make a profit then the occasional loser is not an issue, it's a fact that 1/5 shots will lose once every 6 runs.

As with most things in racing you need the right temperament to cope with the inevitable losing runs. At least with this process those losing runs should be short.

[MCLARKE]

14 hours ago, Valiant Thor said:

Royal Parade is the lowest forecast fav (RP) and has the best Fav > 2nd Fav differential (which imo is a better indicator than just being fav)

I have analysed my base data and it does prove informative.

For a differential above 267% there is a profit of 10.96 from 499 selections (2.2%)

For a differential below 267% there is a loss of -75.62 from 913 selections (-8.0%)

 

[Valiant Thor]

5 hours ago, MCLARKE said:

I have analysed my base data and it does prove informative.

[Yalwen]

This is an interesting and fascinating post. I have been using the RaceAdvisor site for some time and believe that when combined with At The Races it’s the best free online information there is. I am puzzled, though. 

You obviously do extensive research and I try to do that too, what with spreadsheets and other daily betting chores but I am wondering how you calculate the ‘price differential’ to such precision as 267%. Would you be prepared to show us how as I don’t think I have come across it before.

Anyway, MCCLARKE good luck with your odds on venture and thanks for your post which I always find balanced and interesting. 
 

Yalwen

 

 

 

 

 

 

[Valiant Thor]

 

Double post ?

Edited August 5, 2022 by Valiant Thor

[Valiant Thor]

On 8/4/2022 at 9:58 AM, Valiant Thor said:

Royal Parade is the lowest forecast fav (RP) and has the best Fav > 2nd Fav differential (which imo is a better indicator than just being fav)

Meet	time	Horse	F-Fav	2nd F-Fav		Fav	2nd Fav	Differential
Yarm	14.40	Royal Parade	1.33	6.00	>	75%	17%	351%
Brigh	14.20	Inspirited	1.44	5.50	>	69%	18%	282%
Yarm	14.10	Bresson	1.50	4.50	>	67%	22%	200%

44 minutes ago, Yalwen said:

I am wondering how you calculate the ‘price differential’ to such precision as 267%. Would you be prepared to show us how as I don’t think I have come across it before.

Royal Parade (75% / 17% -1) * 100 = 342%

The reason why its 351% in the above post is due to not rounding in excel

ODDS	INVERSE
1.33	0.751879699
6.00	0.166666667

Edited August 5, 2022 by Valiant Thor

[MCLARKE]

Or more simply the odds of the 2nd favourite divided by the odds of the favourite less 1.

e.g. Royal Parade 6 / 1.33 - 1 = 3.51

 

[Yalwen]

Thank you both so much. Greatly appreciated, and the Dunces cap has now been removed from my head. It’s how we learn.

[MCLARKE]

6 minutes ago, Yalwen said:

Thank you both so much. Greatly appreciated, and the Dunces cap has now been removed from my head. It’s how we learn.

Very true. I hadn't really considered it until @Valiant Thormentioned it.

I'm not sure if it's more widely appropriate or just to odds on favourites. Something for me to look at this evening with a nice single malt !

[Valiant Thor]

2 minutes ago, MCLARKE said:

I'm not sure if it's more widely appropriate

It is

[MCLARKE]

On 8/4/2022 at 7:40 AM, Trotter said:

It's also very dispiriting if a 1/5 shot loses and you know it's going to take 4 or 5 winners to get that stake money back ....... then if one of them loses you're up the creek needing 7 or 8 winners to get back to levels

Hmmm, not the greatest of starts.

A 4/9 winner followed by 3 losers.

[MCLARKE]

On 8/5/2022 at 11:56 AM, MCLARKE said:

Very true. I hadn't really considered it until @Valiant Thormentioned it.

I'm not sure if it's more widely appropriate or just to odds on favourites. Something for me to look at this evening with a nice single malt !

I've had a trawl through the data of all the favourites and nothing really stands out.

A diff above 2.67 does show a small profit at BSP with an AE of 1.024.

[MCLARKE]

On 8/6/2022 at 3:37 PM, MCLARKE said:

Hmmm, not the greatest of starts.

A 4/9 winner followed by 3 losers.

Make that 4 losers in a row with Apache Spark not showing much of a spark !

[black rabbit]

hows the data on this looking ?

[MCLARKE]

Just the 1 runner since then, Indemnify today, backed at 1/3 and winning at 1/7 SP.

Early days and showing a big loss but encouragingly the 2 winners were backed at much better odds than SP.

 

[alexcaruso808]

10 hours ago, MCLARKE said:

Just the 1 runner since then, Indemnify today, backed at 1/3 and winning at 1/7 SP.

Early days and showing a big loss but encouragingly the 2 winners were backed at much better odds than SP.

 

Do you think that you will be able to catch up and make some profit with this?  What is your gut feeling?

[MCLARKE]

I would expect over a reasonable number of selections there will be a loss at SP of roughly 5%. The question is whether taking early prices I can beat SP sufficently to turn this into a profit. I think I can but time will tell.

The 2 winners so far have beaten SP by about 16%.

A bit disappointing to have a longish losing run at the start but even at those odds it would be expected to have a losing run of 4 once every 23 bets.

[Bang on]

12 hours ago, MCLARKE said:

Just the 1 runner since then, Indemnify today, backed at 1/3 and winning at 1/7 SP.

Early days and showing a big loss but encouragingly the 2 winners were backed at much better odds than SP.

 

There was a rule 4 - 10p in the £ as well !

I think with the bookmakers making the S.P.'s these days it will be a lot of effort for nothing.

[MCLARKE]

1 minute ago, Bang on said:

There was a rule 4 - 10p in the £ as well !

I think with the bookmakers making the S.P.'s these days it will be a lot of effort for nothing.

If I can make a return of 3%+ from it then it will be a welcome addition to my portfolio. The point is that a lot of people won't back at these odds so there is less incentive for the bookies to shorten them.

[Bang on]

1 minute ago, MCLARKE said:

If I can make a return of 3%+ from it then it will be a welcome addition to my portfolio. The point is that a lot of people won't back at these odds so there is less incentive for the bookies to shorten them.

A fair point, I myself rarely back below 9/4 these days. 

[MCLARKE]

Still showing a big loss, 4 winners from 9, -4.42 at SP, -3.23 at early price.

Encouragingly all 4 winners have beaten SP.

Today's selection, Presentandcounting in the 5.10 at Bangor had an early price of 4/6 and a SP of 1/10. That is an edge of 52%.

[harry_rag]

8 minutes ago, MCLARKE said:

Encouragingly all 4 winners have beaten SP.

Today's selection, Presentandcounting in the 5.10 at Bangor had an early price of 4/6 and a SP of 1/10. That is an edge of 52%.

There seems a contradiction in those sort of market moves being observed and the notion of "no-one" wanting to back odds on shots. Maybe a case of a lot of punters tending to leave them alone but some seriously shrewd money being piled on the "good" ones. So all you need to do is identify the ones that are going to shorten drastically before they actually do!

[MCLARKE]

9 minutes ago, harry_rag said:

So all you need to do is identify the ones that are going to shorten drastically before they actually do!

Any ideas gratefully accepted !