Time rating

[lomaca]

A question to those using time rating.

I gave it away some time ago as I just couldn't make it work.
I used only on shorter distances as per common wisdom.

Never having changed any of the algorithm I use, I started to try it again on distances over 1600M a few month ago and to my surprise it performs exceedingly well.
What have I done wrong-right? Or is the short distance belief just a myth?
Similar experience anyone, or is it just an ephemeral fluke?

[Chrome Prince]

I know of many time raters that do well, especially using 200m sectionals of a race, measuring fastest time and acceleration.
I honestly think it's more suited to dirt tracks in the usa.
Turf courses, the pace is different and horses really aren't extended til much later.
But if you found an edge, then you may be looking at times differently to traditional methods.
I do know that when Beyer came here, he couldn't get his successful time ratings strategy to work here.

[woof43]

But if you found an edge, then you may be looking at times differently to traditional methods.


Looking differently at " time" may mean trying to understand "how long is a race"?
and then one may contemplate "when is a race over"?

[lomaca]

I know of many time raters that do well, especially using 200m sectionals of a race, measuring fastest time and acceleration.
I do it a bit differently because sectionals are still not available for all races, but I compared and it's not too different.

I honestly think it's more suited to dirt tracks in the usa.
Turf courses, the pace is different and horses really aren't extended til much later.
Yes I think you are right about that, to have consistent speed rating one has to have consistent track conditions, and there is no way we can have that on turf. I do allow for track speed on different tracks and track conditions though.

But if you found an edge, then you may be looking at times differently to traditional methods.
I do know that when Beyer came here, he couldn't get his successful time ratings strategy to work here.I have a few books on speed including that of Beyer's and tried to follow them as close as I could.

Don't know about any edge though, I think it's nothing more than a temporary thing. I used it on short distances for years without great success.
True I did get some good priced winners but not enough to cover the losses.
Generally I found my top speed selection would take off as a rocket lead almost all the way and fade on the last hundred metres.

I only started to look at them again when, by accident, I left my rating programme sit on time rating and found good results on longer distances.

Maybe woof43 is right, and in longer races the jockey has better control on how the race is run and genuine speed becomes more prominent?

Time only will tell. (Ouch!)

[The Ocho]

Generally I found my top speed selection would take off as a rocket lead almost all the way and fade on the last hundred metres.

Maybe now with Betfair, you can lay these horses in running when they are a reasonably low price. ;)

[lomaca]

Maybe now with Betfair, you can lay these horses in running when they are a reasonably low price. ;)The Ocho
Can't bring my head around to this laying caper.

I do know that my top time rated selections are not profitable for backing, but at the prices they are running maybe they would be losers at laying?
I mean if one comes in at $25 that is a fair chunk of money to pay out?

Any easy way to test it?

[woof43]

I do it a bit differently because sectionals are still not available for all races, but I compared and it's not too different.
Yes I think you are right about that, to have consistent speed rating one has to have consistent track conditions, and there is no way we can have that on turf. I do allow for track speed on different tracks and track conditions though.
I have a few books on speed including that of Beyer's and tried to follow them as close as I could.



A couple of points you have touched on above, one always needs to know what track variant needs to be applied and to which times (sometimes the state of track/track variant gets to a point when even a track variant cannot be applied, you don't need to be a rocket scientist to understand why that happens).
When the winner crosses the finish line say for a 1400m race, we assume he has run the full distance of 1400m for that race.
Now if this same horse competed against the track record holder (Fixed Standard, not that one would use the Track record as the Fixed standard) say for arguments sake at Rosehill the standard is 81.38 secs you create a Track constant Time X distance = 113,932 we can then look back at say a recent winner on the 31st Dec. who won in 83.43 secs divide this into the track constant = 1,365.5mtrs so if i compare this winner to the fixed standard we can see the distance difference with this 1 crosshair (everything should be measured in distance completed, not time).
There are 3 crosshairs that are used then to define the ability matrix of a horse,fixed, relative and a moving standard once these are defined we then can look at sectionals and then construct how how each race may unfold.

But then again all this is getting complicated, just stick with what your already doing.

[moeee]

But then again all this is getting complicated, just stick with what your already doing.

Your ability to turn simple into complicated is frighteningly amazing Woof43.

P.S. Can't recall if I replied , but thanks for the complicated sheets.
I have filed them in my "some things are best left to experts" Folder.

[darkydog2002]

I reckon I,ll look at Time Ratings when the Official Race Club Handicappers incorporate them into their ratings.(Which is never)

[aussielongboat]

I reckon I,ll look at Time Ratings when the Official Race Club Handicappers incorporate them into their ratings.(Which is never)

what makes you think that they aren't now

Theoretical example:
2 horses win at 1200M on same day in different races.
carry equal weights - similiar or same class - maiden for example
one breaks the course record.
how do you think they would line up next time?
of course the quicker one would get more weight.

cheers

aussie

[aussielongboat]

I always thought the most useful aspect of time was how it related to pace in a race.
thus if your pick is tailed off and they run some" cheap sectionals" it would have no chance as the race would become a quarter-horse finish down the straight.

so you discard the run.

next time under similar conditions where there is a couple of early speedsters in the field you may feel that your pick maybe will be steaming home over the top and represents good value

that's as far as i could get with it.

[The Ocho]

The Ocho
Can't bring my head around to this laying caper.

I do know that my top time rated selections are not profitable for backing, but at the prices they are running maybe they would be losers at laying?
I mean if one comes in at $25 that is a fair chunk of money to pay out?

Any easy way to test it?
Well I definitely wouldn't be laying anything at $25 but a bot could be set up to lay your own selections in running if they come down to a certain price. If they are virtually leading all the way but then tiring then you would think the price would come down to maybe $3 or $4 (or less). The bot would fire off a bet and hopefully get matched and if it wins you haven't lost so much.

I guess it depends on strike rates and the like.

[woof43]

Your ability to turn simple into complicated is frighteningly amazing Woof43.

P.S. Can't recall if I replied , but thanks for the complicated sheets.
I have filed them in my "some things are best left to experts" Folder.

If I recall the sheets i provided were in response the contact address you provided.

1 sheet detailed time adjustments one needed to make when comparing a runners time from any track to The Meadows.

The other just detailed the average winners time for every track in Australia over all distances each sheet provided a count as to the confidence/validity of the data.

[lomaca]

Well I definitely wouldn't be laying anything at $25 but a bot could be set up to lay your own selections in running if they come down to a certain price. If they are virtually leading all the way but then tiring then you would think the price would come down to maybe $3 or $4 (or less). The bot would fire off a bet and hopefully get matched and if it wins you haven't lost so much.

I guess it depends on strike rates and the like.One question, how do you know what the "in running" price was?
I don't collect any data from BF.

I have very good records as to the selections final TAB prices and finish positions so I could work out the position on lays, but if there is a diff. between "in running" and final price than I'd waste my time.

Thanks to all who replied.
As I said before, I gave up on time rating some time ago only looked at it recently again, it's not something one can do easily, it's more like a black art, too many variables.

[darkydog2002]

Phil Purser of "justracing" uses them but god knows why.?

[The Ocho]

I'm not sure lomaca as to in running prices. My bot can tell you what the highest and lowest the horse traded at but doesn't give how much money was bet at each level.

[lomaca]

I'm not sure lomaca as to in running prices. My bot can tell you what the highest and lowest the horse traded at but doesn't give how much money was bet at each level.Thanks T O.

I think I stick to what I know best, here is an example how bad my time rating is compared to my class rating. (see time text)

Race 6 Pinjarra, you'd think a 1000 M race would be ideally suited for time rating, and yet? (see class text)

Unfortunately the nemesis of not being able to get matched for the full amount struck again, but this time not too bad. (see BF text)

[lomaca]

Thanks T O.

I think I stick to what I know best, here is an example how bad my time rating is compared to my class rating. (see time text)

Race 6 Pinjarra, you'd think a 1000 M race would be ideally suited for time rating, and yet? (see class text)

Unfortunately the nemesis of not being able to get matched for the full amount struck again, but this time not too bad. (see BF text)Not betting but let's see
PIJ race 8
Class No, 6, 9, 7
Time No, 3 ,11 13

[jose]

Hope you got some of the TRI Iomaca.

[lomaca]

Hope you got some of the TRI Iomaca.Hate to admit it Jose, I don't even bet a quinella!

Trust me, I had so many two legged tries and three legged quadies that I can't be bothered anymore with multiples.

I would be hard pressed to pick an exa in a two horse race.
Sad but true.

Good luck

Ps should I strech my luck?

Here it goes, no bets again
Race 9 PIJ
Class 2,5,16,3
Time 6,12,5,10

[darkydog2002]

Hi Chrome ,
Just wondering how that system of yours using TIME panned out .
Cheers
darky

[jose]

2 from 2. Good stuff.