- This topic has 139 replies, 11 voices, and was last updated 19 years, 1 month ago by
.
- Posts
I have decided to start a thread for people who are interested in compiling ratings. Ratings is my preferred way to go with racing and I enjoy the research as much as anything else ..
These are my rating for the 2:00 Wolves today
Climate(IRE)235<br>Bridgewater Boys .. 221<br>Dudley Docker(IRE) .. 200<br>Prince Dayjur(USA) .. 187<br>Templet(USA) .. 156<br>Contra Mundum(USA) . 152<br>Hand Chime .. 122<br>Kingsmaite .. 120<br>Aventura(IRE) .. 119<br>Mademoiselle .. 119<br>Suffolk House .. 95<br>Crusoe(IRE) .. 87<br>Ligne D`Eau .. 66<br>(Stdev = 51)
.. comments are welcome !
Good luck for today Dave.
Nice to see a few more people posting in the "engine room".
Regards – M matron<br>:cool:
(Edited by Matron at 1:25 pm on Feb. 23, 2007)
Cheers M ..
Climate(IRE)235 .. 8th 9/4<br>Bridgewater Boys .. 221 .. 4th 5/1<br>Dudley Docker(IRE) .. 200 .. W 3/1<br>Prince Dayjur(USA) .. 187 .. 5th 9/1<br>Templet(USA) .. 156<br>Contra Mundum(USA) . 152<br>Hand Chime .. 122<br>Kingsmaite .. 120<br>Aventura(IRE) .. 119<br>Mademoiselle .. 119<br>Suffolk House .. 95<br>Crusoe(IRE) .. 87<br>Ligne D`Eau .. 66
I like to back the top 3, 4 or 5, depending on the number of runners .. so this was a break even race for me.
I’ve got a bit of spare time and feel like writing a longish post. So I have decided to write a few thoughts and explain my way of thinking about rating race horses, making a book and analysing form with a view to getting an edge when backing on horses.
I’ve researched this subject for about four years and although I don’t profess to be any sort of expert, I have picked a few things up along the way and would appreciate any feed-back or faults anyone can see with what I am saying here.
To me ratings fall into two distinct categories. There are those which calculates the horses past performance in pounds and then try and estimate what the future is most likely to be based on past performance and there are those that are not.
Ratings that are calculated in pounds are commonly known as handicap ratings. I am not interested in handicapping because the horses actual weight cannot be determined before the race and therefore the actual weight carried (or moved along) over the course of the race cannot be correctly determined. In my opinion the actual weight that the horse weighs is the most important variable in all handicapping methods. I believe actual horse weight accounts for bounce factor and unexplained improvements and reversals in form. For me this is a major draw back for the likes of RPR and Topspeed methods. Anyone handicapping horses must consider that the actual horse weight will vary between +/-1% of the nominal weight allotted for that age/sex band, in my opinion.
To me this approach is just pi$$ing in the wind, no matter how hard you try and how precise you get with your handicapping you are never going to overcome this variation, which could be as much as plus of minus 20lbs.
It’s the second type of rating that interests me .. these are basically values allotted to various factors, that do not necessarily include weight, that are used to generate a rating. This is, each horse has a value and then that value is calculated against the other values to say what price the horses should be.<br>:cool:
Dave,
I use the weight/ratings approach, although I’m having a break until April because this approach works best in the months when the going is consistent and there are no obstacles in the way!
I think you are going against the conventional wisdom to ignore weight carried because the evidence to support it is very strong indeed and has been the basis of handicapping for hundreds of years.
That said, there is no doubt that a horse’s fitness must be related to its weight, which is many times greater than the weight in the saddle. A fat horse, rather like a fat person, will be uncompetitive in a test of speed.
I certainly agree with your other main point – that there are other factors – perhaps more important than weight carried – which have a major bearing on the outcome of a race. If you can convert these variables into ratings, I think you are on the right track. I’ve been trying for years to do just that, making adjustments in the light of experience and results.
We all have an opinion as to which factors are most important, so our final ratings will be different, although our basic method might be the same:
1. Identify positive factors
2. Assign ratings to each factor
3. Calculate ratings
4. Assign odds to each horse – probably the most difficult step.
5. Place bets
6. Collect winnings and head for the sun.
My own list of positive factors has been set out previously on my ratings thread, but it would be interesting if other people gave their preferences. I think I’ve seen a few threads on the main forum over the last few years which have done that.
1.30 Lingfield<br>What Do You Know .. 190<br>Up Tempo(IRE) .. 183<br>Fairdonna .. 166<br>Hotchpotch(USA) .. 143<br>Executive Paddy(IRE) .. 131<br>Lucius Verrus(USA) .. 111<br>Binnion Bay(IRE) .. 106<br>Mulberry Lad(IRE) .. 101<br>Bucharest .. 84<br>Cativo Cavallino .. 77<br>Cool Tiger .. 74<br>Inwaan(IRE) .. 0
This is the race for today .. !
Artemis, these are my basic class ratings calculated from how each horse has perfomed over the past 6 months or so. In effect they are my own method of allotting a handicap without using pounds. Using prize money and finishing position.
I would expect the winner to come from the top 4 or 5 in the rating.
.. more later.
Hi dave jay
I agree with you.
The way I see it, every race is an examination for the horses. Various factors that will influence the result need to be assigned a rating. These individual elements will not carry the same significance, so a weighting needs to be applied. The natural conclusion is then to produce an overall merit score.
All this is very difficult to do and adhere to consistently.
A systematic approach is the only way to go if you want to stay sane.
byefrom<br>carlisle<br>
Hi there fellow prisoners
I call for a Zen like calm…………..
Knowledge is power.
Humans standing alone are weak.
<br>byefrom<br>carlisle<br>
Good posts lads .. firstly, I do agree that the path to private valuable ratings is a Kafka type journey because you can never really know if you are right or wrong. For me, understanding how luck works was a break through .. but enough of me, who mentioned Clive Holt?
Clive Holt devised a rating system that was based on recent form .. any VDW buff, should be familiar with this method of assessment. These are the rules for Clive Holt ratings .. I believe that they were revised later on but these are the ones that I use.
Points are awarded to each horse as follows;<br>C = +1<br>D = +2<br>Form – last two runs from the current season only;<br>1 = +5<br>2 = +3<br>3 = +2<br>4 = +1
Example;<br>3:30 Kempton<br>Horse No. – Pts awarded<br>1 – 6<br>2 – 0<br>3 – 3<br>4 – 2<br>5 – 3<br>6 – 5<br>7 – 0<br>8 – 10<br>9 – 2<br>10 – 5<br>*****************************************<br> 1(8) 0-25522 Sir Douglas Beaten Favourite R Harris 4 9 2 P Hanagan<br>2(9) 16-85 Methaaly (IRE) Jane Chapple-Hyam 4 8 13 B Doyle<br>3(6) 409611- Generator Course and Dist winner Dr J D Scargill 5 8 13 J Fanning<br>4(2) 321239- Motu (IRE) Distance winner I W McInnes 6 8 9 Samuel Drury (7)<br>5(1) 11480-4 Takitwo Distance winner Beaten Favourite P D Cundell 4 8 8 L Keniry<br>6(7) 527-144 Beneking Course winner Distance winner D Burchell 7 8 8 M Fenton<br>7(4) 7031-66 Zazous J J Bridger 6 8 6 M Halford (3)<br>8(3) 2-96312 Grand Palace (IRE) Distance winner D Shaw 4 8 4 Dean McKeown<br>9(5) 0660-07 Kew The Music Distance winner M R Channon 7 8 2 C Catlin<br>10(10) 4710-39 Crafty Fox Distance winner A P Jarvis 4 8 2 J D Walsh (7)<br>*************************************
.. cheap and cheerful?
Now to turn those into a price. There are a few ways to do this and then even more ways to use the price, if there’s any interest we could persue this futher, but first of all we’ll look at a simple way of pricing the race from these figures.
You can add of of the points added together and then divide each individual score by the total to get a percentage .. <br>Score .. %age<br>6 .. 16.67%<br>0 .. 0.00%<br>3 .. 8.33%<br>2 .. 5.56%<br>3 .. 8.33%<br>5 .. 13.89%<br>0 .. 0.00%<br>10 .. 27.78%<br>2 .. 5.56%<br>5 .. 13.89%<br>36 = total
and then convert them into prices decimal odds (1/x) where x = %age chance or fractional odds (1/x)-1<br>No. .. dec .. fractional<br>1 .. 6 .. 5/1<br>2 .. 0 .. 100/1<br>3 .. 12 .. 11/1<br>4 .. 18 .. 17/1<br>5 .. 12 .. 11/1<br>6 .. 7.2 .. 6.2/1<br>7 .. 0 .. 100/1<br>8 .. 3.6 .. 2.6/1<br>9 .. 18 .. 17/1<br>10 .. 7.2 .. 6.2/1
This method gives us a 100% book. Anyone looking to make a backers book should multiply the total scores by 1.2 to make sure you have a theoretical 20% edge, for this race 36 *1.2, divide the individual scores 43.2. Doing that gives you an 80% book.
Our favourite Grand Palace is 13/2 in the RP forecast .. at 100% we have it as a 5/2 shot, on our backers book 80%, its 10/3.<br>Good Luck
Top rated today<br>Plateau13:30 Kempton<br>Hill of Almhuim(IRE)14:00 Kempton<br>Grove Creek14:30 Kempton<br>Ansells Pride(IRE)15:00 Kempton<br>Motu(IRE)15:30 Kempton<br>Rogue16:00 Kempton<br>Detonate16:30 Kempton<br>Silca Key17:00 Kempton<br>
The most difficult part for me is determining what sort of weighting to give previous form – ie how much better is a win in a cl2 race from a 3rd place in a cl3 race. The rest of the parameters, change in OR, form over dist or going etc should be insignificant when class is taken into account IMO.
David .. how do you know that class is significant ? or should I say more significant than any other factor .. ? I’m not being antagonostic, just curious.
Thanks Dave for the information re: ratings to odds. I shall incorporate this on my spreadsheet tommorrow and report back in a week or so.
My idea behind this is the view that it may help to identify beforehand races where the ratings may be suspect in a more systematic way rather than simply avoiding certain types of races. I’ll explain all next week.
You Guys! You floor me with your techniques, spread-sheets and I don’t know what-all lol. So much so that I hesitate to post in case I get shot down in flames. Let me explain…
I am 71 in 2 days time, had surgery for cancer, a stroke last summer, got arthritis bad, wear specs and a hearing aid, walk with a stick, and my teeth are falling out. You may not believe this but my wife has a programme of work set out for me – decorating and gardening, as soon as the weather improves that would kill an elephant. The only thing that I have left that is in any way near working order is may brain, so you have got to catch me whilst I am still warm.
I have been following UK horse racing since 1951, which means I have had time to pursue every avenue of approach to selecting winners that there is, including compiling ratings. In the end I came to the conclusion that you are just as well off using any commercial ratings that you respect. The trick is to be able to price them up and bet them at your own price as the only thing that matters in the end is the odds.
Let’s take the first race on Dave’s list today 13.30 Wolverhampton. Using the Pattern Form site data I rated the 4 X 6 month columns 6, 5, 4, 3, 2, 1 in descending order and totted them up to get a set of ratings. Then I valued all those down to 10/1 as possible bets (I didn’t bet in this race) but would have bet all those that went off better than my price, but at my odds, to return 100pts including stake:
13.30 Wolverhampton<br>1 – 6<br>2 – 9 = 9/1 stake 10pts. SP 9/1 no bet<br>3 – 0<br>4 – 8<br>5 – 6<br>6 – 16 = 11/4 stake 26pts. SP 9/1 WON = 260pts return<br>7 – 16 = 11/4 stake 26pts. SP 20/1 lost<br>8 – 7<br>9 – 10 = 7/1 stake 12pts. SP 10/1 lost<br>10 – 0<br>11 – 3
Result 260pts return – 64 pts staked = +196pts
The point I am trying to make is why do all the hard work when it has been done for you? What you have to do is learn to price ratings up and stake them at your prices when they are viable.
Let’s take RPR in the same race and see how we would have gone on:
1 – 91 = 15/2. SP7/1 no bet<br>2 – 91 = 15/2. SP 13/2 no bet<br>3 – 87 = <br>4 – 88 = 10/1. SP 6/1 no bet<br>5 – 87<br>6 – 90 = 8/1 stake 11pts. SP9/1 won return 110pts<br>7 – 85<br>8 – 91 = 15/2. SP 6/1 no bet<br>9 – 89 = 9/1. SP8/1 no bet<br>10 – 89 = 9/1 stake 10pts. SP 33/1 lost<br>11 – 79
Result 110pts – 21pts stake = +79pts
I am not trying to be clever (although my wife does tell me I am a B-knowall lol) just trying to point you in the right way in the light of my long experience.<br>
(Edited by Formath at 8:08 pm on Feb. 25, 2007)
I know what you mean Alan .. it does seem like a lot of work, but I have most of my stuff automated, so its the setting up of it that takes the time and then its just a matter of pressing a few buttons.
I do like that technique you have posted up there and I’m sure I’ll get around to doing a little excel number on it. If you want a copy when I’m done, let me know.
I’m sure there are a great many things you will be able to share with us before the good weather comes and you disappear up a ladder .. :biggrin:
Quote: from Formath on 8:04 pm on Feb. 25, 2007[br] Let’s take RPR in the same race and see how we would have gone on:
1 – 91 = 15/2. SP7/1 no bet<br>2 – 91 = 15/2. SP 13/2 no bet<br>3 – 87 = <br>4 – 88 = 10/1. SP 6/1 no bet<br>5 – 87<br>6 – 90 = 8/1 stake 11pts. SP9/1 won return 110pts<br>7 – 85<br>8 – 91 = 15/2. SP 6/1 no bet<br>9 – 89 = 9/1. SP8/1 no bet<br>10 – 89 = 9/1 stake 10pts. SP 33/1 lost<br>11 – 79
Result 110pts – 21pts stake = +79pts
Hi Alan – can you run through exactly how you priced up this race using those ratings please. And why do you bother to price up a horse rated 88 but not those rated 87 or 85.
- You must be logged in to reply to this topic.