Originally Posted by UselessBettor

I did a lot of analysis with the neural's on R+S.

I would suggest not bothering with them as they are not going to be profitable. There are niche cases where they work but they are hard to find.

Instead look at the ratings sheet which as EST, 12M, BL3, etc. These are much better if you combine them with some form analysis.

I use EST, 12M and BL3. I don't worry about LSR.

What you need to do is look at the ratings for the horse at the same distance, same track condition, same field size, (there are a few more but you can come up with what you like). This gives you a baseline of what you think is the base rating for this horse. it may be the same as EST or completely different. It should also give you a range of ratings that the horse is likely to produce. For example even though EST may be 30 you might find at this distance it usually produces a rating of 28-30 and on good tracks it produces a rating 29-34 and in similar field sizes it produces ratings of 26-29. This gives us a range of the horse producing a rating of 26 - 34. You can do this for all horses and now you have a range of ratings for each horse. You can then compare these and look across the last few races to determine if this horse is going to improve and run at the top of those ratings or whether it might struggle and run in the lower area of the ratings. This should give you a good idea of the chances to dutch to make a profit.

I hope that helps.

Originally Posted by entropy

Thanks for an interesting post on the R&S ratings UB!

I know zilch about the R&S ratings other than they are compiled using Don Scott methods.
I hope they found some new good tweaks because, although profitable in earlier days, they went downhill and DS reportedly died broke.

Scanning the site I could not find any records of how the ratings have panned out in practice.
Just a passing comment where they say the results of their Computer Racecards are "astonishing".

Did your analysis show this?

Am I missing out on being astonished?

My recent efforts with my own ratings could also be called "astonishing", astonishingly feeble!

Originally Posted by UselessBettor

They are based on DS Ratings but its just the initial base rating. Most ratings don't work and most people who follow the ones on R+S will lose money too. You can't just pick up a rating and hope it works. As I said you need to find the ratings this horse produced in similar conditions, similar distances, similar field sizes, etc. That takes work. Most punters won't do that work but anyone who takes the time to do it can use any set of "reasonable" ratings and come up with a derived set of ratings that can then be dutched.

I use lots of sources of data and R+S is one of them. I wouldn't say their ratings are astonishing, but I wouldn't say that about any of the ratings I get. But when I apply my process to the ratings, and then dutch the profitable situations I can easily profit.

Originally Posted by UselessBettor

......

Instead look at the ratings sheet which as EST, 12M, BL3, etc. These are much better if you combine them with some form analysis.

I use EST, 12M and BL3. I don't worry about LSR.

......

Originally Posted by Stix

I agree UB, those are the ratings I look at as well as the FR rating (Don Scott) in the worksheets to see if there is separation in the values. But horse profile is very important and of course the speed map is possibly the biggest piece of the puzzle - it is for me anyway.

Originally Posted by UselessBettor

Stix,

Do you prefer the R+S speedmap or the punters dot com speedmap ?

I don't have enough data on the punters dot com speed map to make any judgements.