• Hi Guest Just in case you were not aware I wanted to highlight that you can now get a free 7 day trial of Horseracebase here.
    We have a lot of members who are existing users of Horseracebase so help is always available if needed, as well as dedicated section of the fourm here.
    Best Wishes
    AR

Coder/scraper needed

I am struggling as to why horse A, who has had the same DSLR profile as Horse B is expected to perform in the same way as Horse B next time out?
 
Thedeadheat said:
I know for a fact. As I have done work.

Horses that run every 25-35 days regulars have a lower strike rate than normal. I believe this because they way they are trained.
Click to expand...

The following is over the last 3 years. You will see that the BFSP A/E remains around 1.0 irrespective of the DSLR. This is indicating that the horses are winning as often their odds are indicating they should.
1537967759328.png
 
ArkRoyal said:
The following is over the last 3 years. You will see that the BFSP A/E remains around 1.0 irrespective of the DSLR. This is indicating that the horses are winning as often their odds are indicating they should.
View attachment 65933
Click to expand...

These stats are just previous run only??

Horses that are running regularly around the 35 days mark don’t win as often as they should. Not just previous but accumulative.
 
To show my point.

And as I stated previously about 0 form.

These are my picks today

3.55R- Blue Gardenia
4.30R- Torrid
4.40G- Master Carpenter
7.45W- Caesars Commet

I have taken 0 consideration in form. Purely on the bases that I believe these horses have shown me something today that’s they will perform better than they have recently.

For the purpose of this experiment I use 0 form and I will use 0 form.

I am trying to gather data for a horses profile.

Hopefully I find someone who can code. The problem I have is if I use freelancer or upwelling I can’t get my ideas across. I ideally need someone with horse racing knowledge other wise I’m just fighting with someone who barely speaks English.

I’ve posted on these sites for scrapers etc but it seems all Bangladesh coders. I have no problem with that but this is quite a complex task and would require great understanding skills of English.

I am not the greatest communicator myself.
 
Last edited:
You’d have to have your bands predefined as say
B1 = 1-7
B2 = 8-14
B3 = 15-21
B4 = 22-28
B5 = 29-36
B6 = 37-44
B7 = 45+
Otherwise you’re likely to end up with a practically infinite number of possibilities.
So if Horse H1 has the profile in last 5 runs of B1,B4,B5,B2,B7
Are you then wanting to find all instances over the last x years where other horses have had the same days off profile and then how they've performed? Or just the horses with the same trainer?
Like ArkRoyal I can’t see what value the former would give for unconnected horses.
However the latter would seem to mean you’re looking for the days off profile for it’s trainer?
 
pawras said:
You’d have to have your bands predefined as say
B1 = 1-7
B2 = 8-14
B3 = 15-21
B4 = 22-28
B5 = 29-36
B6 = 37-44
B7 = 45+
Otherwise you’re likely to end up with a practically infinite number of possibilities.
So if Horse H1 has the profile in last 5 runs of B1,B4,B5,B2,B7
Are you then wanting to find all instances over the last x years where other horses have had the same days off profile and then how they've performed? Or just the horses with the same trainer?
Like ArkRoyal I can’t see what value the former would give for unconnected horses.
However the latter would seem to mean you’re looking for the days off profile for it’s trainer?
Click to expand...

You have explained that in one post in which I have been trying to do.

That is exactly it ideally overall for all horses and for trainers.

I’d like to know the same for distance too and with going conditions.
 
So if you’re trying to find days off profiles for trainers, from the figures you’ve quoted that’s 59000 to test per trainer.
In the last 12 months there’s been about 1400 different trainers running horses in GB/IRE, you could cut that down to just those who’ve had a winner in the last 12 months and that gets about 850.
Even with the 850 you’re talking 50,000,0000
 
If I stick with all results. I think 25 years of data would give around 250,000 races. ( winners) and something like 1.5 million runners...

That’s the sort of data I want to look at.
 
Not sure in what context you mean by "all results"?

Not sure about others by my DB goes back to the start of 2000.
 
So for every run of every horse over the last x years you want to look at the days off bands for it's last 5 runs and find out whether that is winning days off profile for it's trainer?

Sorry my DB is SQL Server, too big for Excel or Access and isn't for sale as it's my own bespoke design/build etc over a period of years.
 
pawras said:
So for every run of every horse over the last x years you want to look at the days off bands for it's last 5 runs and find out whether that is winning days off profile for it's trainer?

Sorry my DB is SQL Server, too big for Excel or Access and isn't for sale as it's my own bespoke design/build etc over a period of years.
Click to expand...

Yes.

But not just trainer for all horses. Last 5, 4,3,2,1

Id like to see how all horses faired. So I could click sort for highest strike rate. And it would show me which profile has the highest strike rate. Or the lowest strike rate.

I could then check it for distance. Is there certain distances that suite a certain profile.

I just have no idea how I would organise this in a spreadsheet.
 
As before I don't see what value looking at every days off profile combination for all unconnected horses will give. Those for a particular trainer maybe but not the former.
 
You must log in or register to reply here.
Back
Top