RATINGS

[lemmyhead]

How often do you read from ratings sellers that their ratings 'are not to be used blindly, they are a guide to the likely winner'? Well my thinking is that I already have that. It's the BF market. As we know, the BF market is 100% accurate, i.e. horses @2.0 will win 50% of races. So my thinking is, if you can get ratings that are 100%+ accurate you back/lay the difference. I do know this works. I had it sorted with Adrian Masseys' ratings. I made a few grand in the place lay market. Fantastic, then Massey closed. I've tried lots of ratings since. All to no avail.
I have seen all sorts of equations for converting ratings into odds. I ask myself why? Surely to simply convert the rating into a percentage should be an accurate reflection of the rating. I'll explain further to clear things up. Add the total of the ratings for all the horses in a race. Convert the individual rating to the total for the race to give each horse a percentage chance of winning. Am I being really stupid and missing something?
Nearly all the ratings I have tried come up with ridiculous odds. Has anybody any ideas? Can recommend a rating service?
Thanks.

[royline]

Hi, I have used horseracebase* for sometime now and find their ratings to be very good especially the top two in each race. Have a look they give you a free trial and it is not expensive if you go with them mention my name the lad that runs it is called Chris.
*Full link has been removed by moderator. Please google it if you wish to visit the service described.

[nig299]

Spurred on by your comments re Adrian Massey, I've spent the last few weeks downloading a shed load of data from the Betfair Directory Listing Site.

Some of the results are quite eye-opening. I always thought that novices and maidens were good races to lay the favourite. Taking a snapshot of some races between the 01/01/215 and 31/07/2015 tells me otherwise.

Maiden Stakes: Races 230 winning fav % 40.43
Nov H/Cap Hurdle: Races 32 winning fav% 40.63

I'll have to have a seious re-think about the way I approach these races. Altough the sample is very small, If anyone's interested, I've put it here:

https://www.dropbox.com/s/hy49nxa0m17ax ... .xlsx?dl=0