- This topic has 41 replies, 5 voices, and was last updated 16 years, 7 months ago by
.
-
Posts
-
Friday 31 August
[b:156uwau6]Sandown
2.10 Lecanvey 102
2.45 Canadian Danehill 97 won 7/1[/color:156uwau6]
3.15 Black Dahlia 96
3.50 Bold Choice 104
4.20 Fabrian 102
4.55 Cassiara 100
5.25 Venir Rouge 100Hamilton
4.45 Stoic Leader 95Chester
2.20 Sue’s Surprise 99 won 8/13[/color:156uwau6]
2.55 White Deer 106
3.25 American Art 105
4.00 Kylayne 111
4.30 Smokey Oakey 105Stakes 12.00pts
Returns 9.62
Loss 2.38ptsRunning Total Stakes 196.00pts Loss 30.67pts[/b:156uwau6][/color:156uwau6]
The main lesson has been that my original way of rating horses using a base figure derived from the RPR top-rated, is flawed. Instead, each horse’s best Topspeed and recent RPR should be related to the horse’s adjusted RPR for today(the task) and not the previously used base rating.
.. without wanting to come across as being cheeky .. I really can’t work out why you aren’t producing your own ratings instead of using the RPR. It’s not hard and if you look to keeping things simple it doesn’t have to be a mammoth task either. You would need to learn how to do a bit of modelling, but I do mine in excel and access, so it’s not beyond the pale.
I have a similar method to what you have described above, create a rating for each horse relative to itself and normalise it for today’s race.
I don’t make any further adjustments to the ratings because they are profitable. It seems that anything that will improve our strike rate will reduce our profitability disproportionately. There was a function in RSB which let you examine ‘jockey in race by strike rank’ .. they are overbet, as are horses that won last time out.
Good luck today .. Sandown ..
Dave,
You can do your own ratings on the RP website and while I agree that doing your own is better than using RPR for various reasons, it is hard work – I’ve tried it. It is very interesting and absorbing – too absorbing, in fact. Even if I had more computer skills or the help of a programmer, I’m not sure I would want to spend much more time looking at form, ratings etc., than I do at present.
I might end up going slightly mad like poor Mark Coton(former Pricewise), who turned to religion to escape from the relentless study of racing form and the disappointments that inevitably accompany such a pursuit – despite the occasional highs.
I like my daily fix, but I can take it or leave it as a hobbyist. If I was doing it for a living, things might be different.
That’s a fair point, you could easily waste your life away messing around with all of this .. it took me ages to get my speed ratings working but I did most of it at work. If I had been at home they would never have got finished between going for walks, watching football, shopping, DIY and gardening.
I suppose there is a fine line between being an enthusiastic hobbyist and an obsessive.
On a personal note, my main task this year is to up my stakes a bit and work on getting better prices which is probably going to be as hard as working out the ratings themselves.
Dave, I’m very much in the anti spam camp.
Just over a year ago I sold the rights to my ratings (for 5 years) and they are available commercially. Since then I have developed the database a lot. The current ratings are purely for my use. A few years ago I did operate a profitable tipping service but this had to be suspended due to ill health. All going well I will return to this when I move to warmer climes about this time next year.
Artemis, I agree with the reasoning on placing less emphasis on jockeyship. A similar case could also be made for the draw?
Fair play to you Wallace .. I nip on and have a look at the site every now and then, I was pretty impressed with the sheet you gave me at Kelso, I wouldn’t put anyone off taking them up for a trial.
It’s not my scene though, I like doing my own thing .. plus if you subscribe to something then you feel compelled to be involved every day and it’s not really practical for me.
That’s the end of this particular thread. The quick ratings method has taught me one important thing about the ratings method which I will make use of in the future. See earlier post.
There was a loss of about 15 percent on 196 bets across a range of abilities, which is about what I expected. You can’t make a profit on ratings alone for all races; you have to selective in terms of quality and also look at additional positive factors. Even then, you will be very unlikely to beat SP these days.
I will be returning in the near future with a revised RP ratings method on a new thread.
- You must be logged in to reply to this topic.