The question is, which is better?

I would be most interested to read some informed opinion on this topic and as a started shall provide some research of my own.

By ways of comparison I have decided to use the top rated horse on weight and time from the Wizard formguide for B,S,M races on Saturdays from 3 Nov 2001 to 30 Mar 2002

Note: in some races there was more than one top rater and in the case of scratchings the starter with the highest rating was used.

The results:

WRAT
sel 499
win 117 (23.4%)
ret 482.4 (-3.3%)
ave. div $4.12

TRAT
sel 533
win 107 (20.1%)
ret 514.3 (-3.5%)
ave. div $4.81

NB. the dividends were selected on this basis-
where the horse started (SP) at its top fluctuation (TF) this price was used.
-
where the horse firmed in the betting the dividend one turn below the TF was used.
ie. TF 3.50; SP 3.00; price used 3.20
-
TAB dividends were not considered although some probably paid considerably more

I believe this method would have returned a small profit if TF had been used on all winners as the difference would probably be greater than 3%.

So WRAT has the higher strike rate but there is a minimal difference in profits. No clear indication here.

Included in these winners were 14 winners in double figures, 3 over 20.0, with a best of 41.0.

(maybe I could sell this as a system, at least it would take a long time to go broke!)

further research to be done:

how do the dual top raters go? expect higher strike rate but significantly smaller divis.

what about only those whose TF is 10.0 or better? only need 7% strike rate to break even.

As a point of interest the horse which returned the 41.0 divi was called Vanishing Powder and won race 8 @ eagle farm on Jan 26. I seem to recall this site advertsing it as a winner under one of their selection methods.



[ This Message was edited by: thekey on 2002-08-20 21:55 ]

The Dual 100 Pointers results as follows:

sel 241
win 64 (26.56%)
return 243.5
profit 2.5 (1%)
ave. div. $3.80

pretty much as expected - higher strike rate, lower dividends.

I am a little concerned as to the longterm consistency of these figures as March was an exceptional month and turned a 17.2% loss into these break even figures.

For the last 4 saturdays there have been 6 wins from 23 bets. this is right on par for the SR but a return of 20.4 sees a small LOT.

I know everyone will suggest filters etc. to improve this but all logical filters will remove the larger priced winners reducing this to a long term loss situation.

I don't know how others feel about this but I don't feel comfortable betting a horse that hasn't won for 12 months or longer, same for a horse that has a low career SR.

This is where all the big priced winners come from.

With regards to the double figure priced horses mentioned in the previous post the results were so woeful it is not even worth printing.

weight and time plus field strength.both 91 100 .fs to1 and ahalf plus. use all 3 for handicapping success. plus of course all your own handicapping procedures.

I prefer to rely on Time based ratings up to 2100m and weight based ratings for 2100m +.

thekey

With what you have as results already (20-23% for either), have you considered backing only those which pay as an overlay. ie $5 or better for a win.


[ This Message was edited by: Fryingpan on 2002-08-26 14:46 ]

Unfortunately backing those over the $5 mark severely reduces the strike rate and thereby increases the need for higher dividends.

I don't believe that because the SR is 20% $5 is necessarily value for any given top rater.

You must consider the other contenders.

ie. if there are 6 horses rated above say 95 would the top rater have a 20% chance of winning, assuming there are more than 6 horses in the race.

you can usually recognize a straight times punter. a forlorn look and a handful of losing betting tickets.

Hi Darkydog and all

I see have you been studying your Don Scott / Rem Plante quotes very dilligently.

In my view, saying times have no place in racings is blatantly ignorant. My own approach is centred on a combination of time and the influence pace has on time, and I have managed to make an effective income for a number of years. I will say though that times are not consistently reliable if you do not consider the early pace of the race and the influence it has had on final time.

My own database of rated races over the past 6 years clearly shows that better horses run faster times than inferior horses. When I comine that with the influence of pace, the relationship and effectiveness of those ratings in identifying the difference between horses is very very good.

At the end of it all, the key is to be able to identify what time rating a horse can run in the likely pace conditions of the upcoming race. Add a few other checks like weight, distance, fitness etc. and you have a powerful indication of the competitive potential of horses in an upcoming race.
I then isolate the value, bet those horses effectively, ignore the poor value and when the dust settles at the end of a year, I find that i'm in front, all be it some years less than others.

So when I hear someone say that times don't work, all it means to me is that they don't know how to make them work.

Our betting environment is still dominated by class and weight and this is the reason why times are more effective. Very few people know how to use them well.

My sentiments exactly Osulld , couldn't agree more.

Am interested in how you can assess form using times. How do you assess the time one horse runs if the state of the track is different, or if the race is run at a muddling pace in the early stages? This approach would need to also account for the fact that some courses are "faster" than others. Seems a difficult way to assess form.