MCLARKE

I can only watch one horse in a race so I'll go with HAATEM who looks the best value. I would probably have £10 EW but would wait until the day of the race as there will probably be 4 places available somewhere and BOG would be available.

Right I've now done the boring part and collected and cleaned all my data and I've got 7 years worth to play with. Now the fun start begins (hopefully). I'll try and limit the results of the analysis to no more than 1 a day. I will use 2018 to 2022 as base data and 2023 as test data. Altogether in the 5 years I have nearly 300,000 horse performances so sample size shouldn't be a problem. The first analysis is to determine whether horses with "good" speed figures perform better (in terms of profitability) than horses with "poor" speed figures. At this stage I am only including runs in the same calendar year. On the next run the results are not good. The top 50% have an AE of 0.99. The next run is more encouraging. The top 50% have an AE of 1.01 with the top 1% having an AE of 1.09. In fact a positive AE is shown for runs 2 to 9. The top 50% is 1.01, the top 1% is 1.07. Not massive numbers but it gives me the confidence to limit my selections to the top 50% of ratings. My maximum rating is 5, my minimum is -36. I will change this scale so that 5 equates to 100 and -36 equates to 59.

Good, saves me having to rate him !

Yep, probably a good way of doing it at this stage. Another method which I'm thinking of is to calculate an average going allowance for the type of going and use that for foreign races.

I see it is now 3/1. Do they normally shorten quickly ? Is it still worth backing at 3/1 ?

To calculate the going allowance for the total card I would have to know the class adjustment for each race on the card.

I'm getting a bit lazy in my old age, I enjoy the analysis of my data but don't really enjoy manually inputting data which is what I would have to do for the foreign races.

As part of my analysis I will be looking at the different class of races and as you say it may be that I stay clear of the top class races. Although if it keeps me clear of the over-hyped Irish horses then maybe it's no bad thing.

I did read somewhere that they retospectively amended it to make it look better

Not really, I don't know how I would calculate the going allowance (I use the the class of the race (1 to 6) to calculate this for GB races).

I'll include it at SP

Another variable is that Sky do not offer BOG. Another feature is the "shape" of the betting, if the favourite is short priced then there is likely to be more benefit of betting each way

This will be one of the variables I will be looking at in my analysis

Indeed it does. However on the day Ballymount Boy achieved its rating of 102 it ran in a quicker time than Al Husn on the same card an hour later and that was rated 116. I assume RPR takes account of more than the speed of the race.