The Best Ratings In The World?

If Carlsberg did ratings…

It’s a very (very!) bold claim, I know, but in today’s post I want to introduce you to geegeez’ new speed ratings, and to try to explain and/or validate the headline statement.

As well as that, I’ll introduce the man behind the ratings, and I’ve come up with a related competition which is both fun to enter, and has awesome prizes.

First up, then, let me introduce you to the speed ratings, and the man behind them.

The job of creating a ratings set, if those ratings are to serve any valuable purpose whatsoever, is a gargantuan one. I have personally long harboured romantic notions about burning the midnight oil over many moons, refining and perfecting a set of winning ratings.

The reality is that life – business, family, horse racing – gets in the way and, in any case, there is no guarantee that after all the labour the numbers would stack up. A sensible alternative was to track down an existing ratings set with three key factors in its favour:

1. Proven at winner-finding over a period of time

2. Under the radar

3. Profitable

There is a fourth element which relates to cost effectiveness and, therefore, to being able to secure the figures for geegeez subscribers. Happily, that box gets a tick as well.

The problem with the major ratings – Timeform, Racing Post Rating, and the like – is that there is a ‘determinism’ to them. That is, they are factored into the betting markets.

These days, with odds compilers having ready access to that info ahead of time, it’s unclear whether the driver is cause or effect… but the key is that, while these ratings are good at finding winners, they are not good at making profit, because the odds are stifled by the volume of backers and by the cautious nature of the layers to the ‘number’ horses.

That’s why it was fundamental for me to locate a ratings set that was under the radar. This normally means finding a very smart ‘mad professor’ type who is largely unconcerned by the world of marketing; and agreeing a licensing deal, which works well for both parties, with said very clever person.

Introducing the man behind the numbers…

And that’s precisely the arrangement I’ve struck with Dr Peter May, a man some of you may already be familiar with. He’s a serially published author, he wrote for Raceform Update and others (including attheraces.co.uk currently), and he is as scientific an aggregator of speed data as I know. And he has a cracking little website where he sells all sorts of interesting data here.

His methods eschew established practices such as the use of standard times, in favour of an iterative algorithmic approach based on neural networks. Neural what?! You can read more about neural networks here if you’d like to but, in the same way you don’t need to know how a car runs to drive one, it’s not necessary to understand the machinations under the bonnet to derive utility from these figures.

Dr May, or Peter to his friends, has a degree in Mathematical and Computer Science; a Masters in Expert Systems; and, a PhD in Artificial Intelligence. He is responsible for authoring the only racing book I ever started but couldn’t finish, Forecasting Methods for Horse Racing.

I have read a number of Peter’s other books, and I found them much more accessible, largely by dint of the fact that they had almost no complex algebraic theorems published between their covers!

So that’s the man, but what of his ratings?

Ask fifty people who have produced ratings, or fifty people who have used ratings, how to use a set of numbers output from a ratings algorithm, and you will generally get a different response every time. One response you will probably not get is, “back the top rated blindly in all races”.

That’s because such an approach has so little finesse to it, such market agnosticism, that it cannot possibly be profitable. Except that with Dr May’s ratings, it was.

From the start of 2012 to the end of 2013, Peter’s top rated selections made a profit of 670 points at Betfair SP, blindly backing the top rated pick in the over 13,000 races on the flat in that time.

You will NEVER see Timeform publish the profit/loss figures for their top-rated, nor Racing Post Rating. That’s because they habitually lose a stack. Take a look at the below table, which makes for eye-opening and instructive reading:

Favourite top, RP Ratings bottom

That is the current ‘Press Challenge’ table, as at 27th August 2014, taken from the Racing Post.

We can see that ‘the favourite’ outperforms every single daily newspaper tipster. And look, there at the very bottom of the table… is the Racing Post Ratings’ performance.

Backing the jolly blindly would have lost you 7.43% of your dough – a slow and mildly pleasurable punting death.

Backing Racing Post Ratings top-rated would have been a much painful and rapid descent off your wagering mortal coil.

I use the above for illustration only. Because, as I said, it is not the intention of anyone that produces a rating set for its users to back top rated picks blindly. But, if you can do that and make a profit – or even get anywhere close to making a profit – then surely the opportunities for more selective use of such prophetic data are rife.

How best to use these brilliant ratings…?

That of course is the million dollar question. And the answer is… you decide.

Now before you get up and walk out, let me explain.

The man behind the numbers has produced a sort of ‘system builder’ tool. More accurately, it’s an analysis tool, which allows users to try out various scenarios on the top rated selections under various conditions.

Dr May’s software tools allow us to look at different age groups and/or different race types, and run a whole array of scenario against them very simply. The video below shows how easy it is to do.

And that gave me an idea for a competition. In the video below, I’ve come up with a scenario which looks like a profitable angle to use in conjunction with Peter’s top rated selections. So what I want you to do is to experiment with the profiler tool and come up with your own ideas for profitable angles.

The best idea – judged by me and my helpers here at geegeez – will win a lifetime subscription to Geegeez Gold… which also means a subscription to Dr May’s excellent speed ratings.

And one runner up – coming up with either a flat or a National Hunt angle – will win a year’s subscription to Geegeez Gold, worth £197. Not bad, eh?

Here’s how to enter…

1. Watch the video below

2. Install the software on your machine [Note, I’m sorry but I cannot answer technical queries either about the installation or the usage of the software. There is help at the link shown in the video].

DOWNLOAD THE SOFTWARE HERE

3. Play around with some scenarios. Then, when you’ve found an interesting looking angle – either back or lay, flat or National Hunt…

4. Leave a comment below. Include your name, the code (i.e. flat or NH), and your angle in the comment. Please copy the angle details into the comment, and include the wins, runs, win% and Profit/£ figures, as outlined in the video. We will naturally be checking the ones we like best, so no making it up! 😉

Competition closes next Friday, 5th September. Here’s a full video walkthrough of how to use the tool and enter the competition. 

Good luck!

The PJM Speed Ratings will be live on geegeez.co.uk, for Gold subscribers, from Monday 1st September. If you’re not currently a subscriber, you can take a ten day free trial here.

I’m very excited to be able to share Peter’s ratings as an addition to what many people are saying are the most comprehensive and easy-to-use set of form tools and race cards available in Britain. And, with a ten day trial, you’re welcome to judge that perception for yourself. 🙂

Matt