The BTC Horse Racing Thread

[Angus Macleod]

@Finn-Kristensen @John-Folan Some effort you are putting into this gents following this thread with great interest

[John Folan]

No selections. Following on from the thoughts on draw bias I analysed something I am testing on the AW. I hadn’t kept draw stats but distance played a massive role. Looking into it further the courses with decent course bias performed better over some distances than others. The results on Timeform and the draw bias was the obvious factor over those distances. Bit of a slog but I am onto something. It literally will be horses for courses at this rate.

[John Folan]

@martin-futter said in The BTC Horse Racing Thread:

@john-folan said in The BTC Horse Racing Thread:

None today. One point profit from the month(two selections lol)

profitable month

My thoughts. Hopefully more selections next month

[Martin]

@john-folan said in The BTC Horse Racing Thread:

None today. One point profit from the month(two selections lol)

profitable month

[John Folan]

None today. One point profit from the month(two selections lol)

[A Former User]

Nice one @John-Folan

[John Folan]

@john-folan said in The BTC Horse Racing Thread:

A selection! 3.5-9 odds range and check the going just before the off and avoid if heavy

Smashed home

[John Folan]

A selection! 3.5-9 odds range and check the going just before the off and avoid if heavy

[John Folan]

@john-folan said in The BTC Horse Racing Thread:

@finn-kristensen said in The BTC Horse Racing Thread:

Wolverhampton doesn't seem to have a low Draw bias, so not an angle to find there from that perspective.

No it’s deffo course shape that affects the draw. I’ve been looking today. Lingfield has no course bias either.

I’ll compile a list of courses and distances that have bias and see what we have to work with.

[John Folan]

@finn-kristensen said in The BTC Horse Racing Thread:

Wolverhampton doesn't seem to have a low Draw bias, so not an angle to find there from that perspective.

No it’s deffo course shape that affects the draw. I’ve been looking today. Lingfield has no course bias either.

[Finn Kristensen]

Wolverhampton doesn't seem to have a low Draw bias, so not an angle to find there from that perspective.

[Finn Kristensen]

If we narrow the selections to Top 3 in the BSP ranks or only accept selections with odds at 9 or lower at BSP, we lose a large portion of the profits (drops to around 25-30 points) while the strike rate increases to just below 40%.

[Finn Kristensen]

@john-folan Southwell looks interesting.

I've run it for 5-12 runner races and these are the results:

A very clear positive bias for Stall 1 there.

[John Folan]

@finn-kristensen said in The BTC Horse Racing Thread:

@john-folan It can do a lot of things

I just checked the results if we only accept horses with BSP between 1 - 9.0. This is what I get:

Stall 2 keeps showing as the best, however, stall 10 also keeps reappearing as profitable. I am not familiar with Doncaster's 5F track so I don't know if there is anything that speaks for high draws as well?

I think flat stats says the outside stall is decent too. Some courses have different bias’s to others based on its shape. Let me see what i can dig up about each race course for the flat/aw.

[Finn Kristensen]

@john-folan It can do a lot of things

I just checked the results if we only accept horses with BSP between 1 - 9.0. This is what I get:

Stall 2 keeps showing as the best, however, stall 10 also keeps reappearing as profitable. I am not familiar with Doncaster's 5F track so I don't know if there is anything that speaks for high draws as well?

[John Folan]

@john-folan said in The BTC Horse Racing Thread:

@finn-kristensen said in The BTC Horse Racing Thread:

@john-folan I just included odds rank, so we only use horses ranked 1-3 at BSP. These are the results:

Interesting. What else can your software do? Filter wise?

Shows there is deffo bias. Just a case of how to approach it

[John Folan]

@finn-kristensen said in The BTC Horse Racing Thread:

@john-folan I just included odds rank, so we only use horses ranked 1-3 at BSP. These are the results:

Interesting. What else can your software do? Filter wise?

[Finn Kristensen]

@john-folan I just included odds rank, so we only use horses ranked 1-3 at BSP. These are the results:

[John Folan]

@finn-kristensen said in The BTC Horse Racing Thread:

@john-folan I just made an analysis on Doncaster. Can't really confirm your figures there tbh.

For the 5F races (AW) I get the following:

So, nothing really tells us that there is a clear advantage for the low draws. However, Stall 2 seems to be the best for whatever reason. I did an extra analysis where I expanded the number of runs to include all races with 5-14 runners - the results are pretty much the same. Stall 2 seems to be the best, but the win rate remains at around 24%. The corresponding Place figures are just below 40% for the Draw at 2.

I again have to stress that the numbers we look at here are at the low end, so it would be problematic to draw too clear conclusions based on these findings.

Needs odd filters. I’m interested in an aw backing system. Draw is the obvious filter that’s needed. I’m going to play a bit. Thanks for the input. I like the way that software you use sets it out. It probably needs runners in the top three in the betting to really notice the benefit. A crap horse is. Crap horse no matter what stall.

[Finn Kristensen]

@john-folan I just made an analysis on Doncaster. Can't really confirm your figures there tbh.

For the 5F races (AW) I get the following:

So, nothing really tells us that there is a clear advantage for the low draws. However, Stall 2 seems to be the best for whatever reason. I did an extra analysis where I expanded the number of runs to include all races with 5-14 runners - the results are pretty much the same. Stall 2 seems to be the best, but the win rate remains at around 24%. The corresponding Place figures are just below 40% for the Draw at 2.

I again have to stress that the numbers we look at here are at the low end, so it would be problematic to draw too clear conclusions based on these findings.