Morning News

Good morning...

This will be very brief today as I've got loads to tell you about the Speed Ratings and also...

I've got loads of final tuning to do on them.

Yes, we've got Speed Ratings;

Sometimes they give incredible results and,

Sometimes they just give average results.

I need to do one last bit of analysis to fine tune them so that we get consistent results all the time.

Anyway, I'll get to that, firstly let's get on and take a quick look at yesterday's results in more detail....

Yesterday's Results

Yesterday's Results, show a strike rate (for our top rated horse) of 25%.

The Overall Profit Index.

If you had blindly backed every top three rated horse in all of the races you would have had 57.56 units returned to 59 units staked. The Overall Profit Index was, thus, 97.56%

Top Rated Horse - If you had blindly backed the top rated horse in all of the races you would have had 14.71 units returned to 20 units staked. The Top Rated Profit Index was, thus, 73.55%

2nd Rated Horse - If you had blindly backed the second rated horse in all of the races you would have had 20.35 units returned to 19 units staked. The Second Rated Profit Index was, thus, 107.11%

3nd Rated Horse - If you had blindly backed the third rated horse in all of the races you would have had 22.5 units returned to 20 units staked. The Third Rated Profit Index was, thus, 112.5%

Forecasts

There were 2 forecasts yesterday.

Wincanton 14:40:00 - The CSF paid £11.07

Wincanton 16:40:00 - The CSF paid £4.68

Reverse Forecasts

There were 3 reverse forecasts yesterday.

Bangor-on-Dee 14:55:00 - The CSF paid £13.64

Kempton (AW) 19:15:00 - The CSF paid £20.07

Kempton (AW) 19:45:00 - The CSF paid £26.45

Tricasts/Trifectas

There were 1 tricasts/trifectas yesterday.

Wincanton 16:40:00 - The Trifecta paid £6.30

Reverse Tricasts/Trifectas

There were 3 reverse tricasts/trifectas yesterday.

Wincanton 14:10:00 - The Trifecta paid £10.30

Bangor-on-Dee 15:55:00 - The Trifecta paid £43.70

Winners

From the 20 rated races there were 13 winners (from the top 3 rated horses.) 2 of those were at nice prices.

Course	Time	Horse	ISP	BSP
Wincanton	15:40	Cappielow Park (3)	13.00	17.5
Kempton (AW)	19:15	The Gay Cavalier (6)	6.50	7.0

The Portfolio

There was a Banker's Bet (but not a Chancer's Bet) yesterday.

Bankers Bet = Bumper Drops! - where R4 Status = 1 and where Race Type = National Hunt - Flat and where RC0 (Good Race/Bad Race) = 1 and Horse has dropped 1 Class

Selection = Back to ISP - Bangor-on-Dee - 16:55:00 - Preseli Rock (5)

Races (over last year): 18 Wins: 5 Strike Rate: 27.78% ROI: 102.23% AV SP: 3.68

Preseli Rock came second by a head.

On the Portfolio Page there is now a link to the Banker's Bet Proofing Page and, also to the Chancer's Bet proofing page.

You may subscribe to my Portfolio. To do so, you need to go to your Members Page and change option 4 on the Your Profile bit.

If you haven't subscribed, the button will say Yes - just click on it and you'll be signed up.

Live Ratings

There was only one non-runner yesterday.

Bangor 16:55 - Our second rated horse was a non-runner. The Live Ratings brought in Nomadic Lad as the new third rated horse. Nomadic Lad came fourth.

Speed Ratings

So, as of yesterday, I've got Speed Ratings.

Yesterday I was a pretty happy bunny but...

And there's always a but, isn't there?

I still have to do some work on them.

Yesterday I took 12 races at random and checked the performance of the top rated (as far as the Speed Ratings were concerned) horses and, you know what...

Every bluddy one of them won!

However, when yesterday's results started coming in, it was apparent that we were doing better on R4 (our current ratings) than we were from the new Speed Ratings.

I'll explain this in a minute but, that's actually not too bad!

It's just Maths.

And Maths is just something that boys study when they are 14 years old and have spots and can't get a girlfriend (and, more importantly, can't get into the school cricket team!)

Why else would anyone want to study Maths?

The bad thing is that I have a bit more work to do (but that's only bad for me.)

Just when I thought that it was it was all over, there's still some more maths for me to do!

You'll be happy to know that my spots have finally cleared up but, I still can't get into a cricket team!

An Explanation...

I realised that the 12 races that I had taken at random were from the start of the database.

Thus, the horses might have only run one or two races from which I've used their speed data.

Now, I have (perhaps naively) decided to base a horse's speed rating upon the results from the its last 5 races.

There is some logic there; Focus Ratings was initially based upon Clive Holt's Fine Form.

I had this silly idea that I could do better than him as I have about a trillion times more computing power than he had back in 1976.

And that's quite literally; I have 3 computers, 2 servers in a data center and an Amazon AWS cloud computing account - that all adds up to a lot more than Clive Holt's biro and the back of a fag packet.

Even if Clive Holt had a pocket calculator (they started being available about then), I'm still way ahead of him in computing power.

Now, Clive Holt used the results from the last five races to generate his ratings and I have mathematically proven that the wins and places from the last five races are the most consistent metric for finding horse form. That won't get you loads of winners but it does describe the overall average form of a horse better than anything else.

So, for the Speed Ratings, I have used the recent speed of a horse over the last five races as my base figure.

What yesterday showed me is that that might have been wrong.

Perhaps I should have used just the last three races, or even the last two.

Today I shall test my Speed Ratings based upon the speed of the horse over the last race, the last two races, the last three races etc. so that I can determine which way is the best to show the likely performance of the horse in today's race.

Ho hum...

More maths (and I feel a spot coming on!)

Conclusion...

Now, it's been a long hard process getting as far as we have (in terms of speed ratings) and yesterday, when I actually had numbers in front of me, I felt about two inches taller as the weight of the world had been lifted off my shoulders.

However, I do need to ask you to wait a little bit longer as I do want to get it right.

The speed figures that we have today have logic and are of some use.

I hope that the speed figures that we have tomorrow have more logic and become indispensable.

To see the image full size, just click on it. Don't forget to use your Back button to return here.

I may well be doing a post at some point today to explain where I am and to show you examples of the new speed ratings in the three formats:

1). All horses in all races (including the previously unrated races),

2). The top three horses in all races (including the previously unrated races) and...

3). The top three horses but not including the previously unrated races.

If not, then I'll certainly be updating you tomorrow morning.

You see, I could just publish what I have and leave it as that; certainly I am really rather quite keen to get this behind me, finish the Winter Upgrades and then get on with my life.

However, it's important that I get it right.

And so I have to do the last bit of analysis to ensure that that happens.

A Caveat...

Now, to be honest, I've never been a big fan of Speed Ratings.

Until we get sectional timings at all our race courses they are, in my opinion, just a guess (although, maybe, an educated one) at the horse's abilities.

And, no matter how much work that I have put in...

My Speed Ratings will be just the same; just an educated guess.

They will only be available for AW and Flat races and they may well prove to only be of use for the shorter races.

Only time will tell.

Although, those 12 winners from 12 races that I found yesterday do make me rather excited.

So, I'm not promising the Holy Grail here but, what I can offer is logically constructed Speed Ratings, for free, in addition to the current form and bayesian based ratings.

It may well be that by combining the two ratings for certain races (sprints, for example) we end up getting more winners; I certainly hope so.

It may well be, that in October, I start waffling on about R5 - a combination of our current R4 and the new Speed Ratings.

Who knows?

All I can say is that it's been an interesting project and I believe that the results will be of some worth.

I've missed a few Wine O'Clocks and had a few sleepless nights (more than a few, if the truth be told) due to the work that I have been doing but...

Feel sure that, once it's all finished,

I'll be calling the Winebulance to help me celebrate a Wine O'Clock like the no other!

Today

Today's early test ratings show 34 rated races from five meetings.

All of today's races are rated.

Now, I'd love to spend time doing some analysis on today's races but I'm really rather quite desperate to get on and finish the speed ratings.

Have a great day's racing!

Once again, if you've got anything to say about Focus Ratings, please wander over to the Focus Ratings Review at Tipsters Review and have your say.

The income that we make from sign ups from that site keeps me in Red Wine and Smelly Cheese!

As always...

My kindest regards