The thoughts of the handicappers.

EC1 said:
Grandouet is better than that..he had the ground and fitness against him..ROR likewise... rating off either is illogical..imo.
Click to expand...

Did he appear to not run his race? Or Zarkandar not run his - which if you believe that Grandouet didn't then it's a given that he didn't?
 
I think you are all being unduly harsh.

Earlier in the thread, the author reviewed the Morgiana, and came to the conclusion that Go Native was "simply a better horse" than Hurricane Fly "at this stage of their careers".

He presumably came to this conclusion by ignoring the tactical nature of the race where HF set it up for Go Native, the fact that HF was a length up at the last (which he pinged, as he did every other obstacle), the fact that HF is entirely capable of finding plenty, the fact that GN had a fitness advantage, and the fact that GN has had a good look up HF's hoop every time they've met.

Personally, I think Go Native might have won if he'd stood-up, but it was by no means in the bag, and looked about 10/11 the pair to me, before GN hit the deck.

Yet Go Native is not only the "better horse", but "simply" so.

Whoever the Author is (and I assume we're talking about Mordin?), his logic seems more than somewhat flawed, and possibly even verging on the dribblings of a raging fud-heid.
 
Grasshopper - I believe that this is the official handicappers chatter.

EC1 - hes rated Zarkandar 4lbs below his best, based on an assumption that Grandouet has replicated his previous best performance. To assume that Grandoeut has not replicated that, as you are suggesting, would mean that Zarkandar has run some way below his best.

Nothing about the race that I watched suggested that Zarkandar didn't run his race (or Grandouet). Which is not to say that either of them couldn't do better in future.

Maybe they did all run a complete stinker and the race should really be rated around Minella Theatres mark of 111 which would have Zarkandar running to about 150? Quite plausible given the ground.
 
Having now checked the results, I don't think it's safe to conclude that any of them ran their race. It was no more than a racecourse gallop. The final outcome may well point to the three having run somewhere close to where the perceived level of the abilility of each would place them but, according to my figures, the race itself was 25lbs slower than Oscar Whisky's. That should sound alarm bells right left and centre.

As I said, I believe in races within races. It was probably a racecourse gallop with a valuable pot for the winner but if it proves ultimately to be a true reflection of the respective abilities [of the big three] it could just as easily be put down to coincidence.

In the absence of the computer power that Timeform would have us believe they have at their disposal and the obvious massive brains that interpret form for us lesser beings, there isn't much else raters can do than apply some form of yardstick handicapping - the same yardstick handicapping that identified Far West, Highland Lodge, His Excellency, Coneygree, Unioniste, Zarkandar and Oscar Whisky as having strong winning chances on the day.

My frighteningly cringeworthy logic tells me the above but in terms of allocating figures to the race, I've decided to rate it around Grandouet running to the same figure I had him on last season (170+), meaning that Zarkandar has run to within a pound of the figure I had for him at Wincanton and Rock On Ruby 7lbs shy of his Champion win.

It may be frighteningly cringeworthily simplistic and it's the kind of race I'd simply just add a '?' to (to remind me not to put too much trust in the figures) but it does seem to provide a 'best-fit' idea of each performance on the day. Which, after all, is often all the OH can do.
 
simmo said:
Grasshopper - I believe that this is the official handicappers chatter.

EC1 - hes rated Zarkandar 4lbs below his best, based on an assumption that Grandouet has replicated his previous best performance. To assume that Grandoeut has not replicated that, as you are suggesting, would mean that Zarkandar has run some way below his best.

Nothing about the race that I watched suggested that Zarkandar didn't run his race (or Grandouet). Which is not to say that either of them couldn't do better in future.

Maybe they did all run a complete stinker and the race should really be rated around Minella Theatres mark of 111 which would have Zarkandar running to about 150? Quite plausible given the ground.
Click to expand...

i think its possible the minella form line could be relevant..i don't think the race as a guide is worth much at all tbh..fitness and ground ?'s make the form of little help to anyone

on the day Z won but didn't look to relish it any more than the other two
 
simmo said:
Given that all handicapping involves using a yardstick of some sort, you appear to be suggesting that handicapping or just applying ratings is inherently wrong.
Click to expand...

All handicapping absolutely does not involve using a yardstick of some sort. The hanging of a set of ratings in any race based on the notion that one horse in that race ran to form is convenient, but scientifically flaky. A more robust way of establishing the worth of a race is described here by by ex-forumite Simon Rowlands: http://bit.ly/hRdpkE

This isn't a Timeform love in by any means, but it's clear that the methods used by any individual who utilises race standards and applies rigorous mathematical principles are going to provide truer results than some bloke who decides to rate a race through a random horse because the result "feels right" to him.

Grandouet is a terrible horse to hang a rating on, because the likelihood of him being exactly the same horse as he was a year ago is almost nil. Injury/lack of fitness are factors which could be expected to retard his performance, while greater physical maturity is a factor which might see him improve since his 2011 win. Only by accepting that these factors cancel each other out perfectly can he deem Grandouet to be the correct means of measuring the merit of the race.

Again, it's possible that Dickinson's ratings are spot on, but that isn't the argument. It's also possible that he's actually rated the race using proper methodology, but simply uses yardstick terminology to explain it to the public, as the BHA refuse to disclose the exact methods they use, or indeed if different handicappers use the same method.
 
rorydelargy said:
Grandouet is a terrible horse to hang a rating on, because the likelihood of him being exactly the same horse as he was a year ago is almost nil. Injury/lack of fitness are factors which could be expected to retard his performance, while greater physical maturity is a factor which might see him improve since his 2011 win. Only by accepting that these factors cancel each other out perfectly can he deem Grandouet to be the correct means of measuring the merit of the race.
Click to expand...
I'm not arguing against that. But how else do you evaluate the race? Zarkandar was coming in from a small-field handicap, the form of which is difficult to accept at face-value while Rock On Ruby and Grandouet were having their first run of the season, and all this happened on ground which was among the heaviest, according to my going allowances, ever raced upon anywhere in the last 30 years. At least we knew G had had a couple of racecourse gallops so there had to be some prospect of his being close to fitness and we saw last week that the same yard's Sprinter Sacre turned up at least as good as before.

I think we are talking about a Timeform love-in. Not only that, but we're talking about Timeform being so superior to everything else around that everything else around is to be scorned as a waste of time, space and energy.

If Timeform was any good, I might subscribe to the view but I have first hand knowledge of how dreadful Timeform ratings are and if they think they are the be all and end all then they are sadly deluded.

I'd also be pretty certain that both the Official Handicappers and Raceform compilers are more than aware of how Timeform use race standardisation and do likewise themselves. I can accept that it works in the situation referred to in the link but when we're talking established form then yarrdstick handicapping becomes just as valid so long as the evidence is there. I wouldn't be able to quote a figure with any accuracy but I'd be pretty sure Timeform rated Oath an average Derby winner when all other evidence made it obvious he was well below that, as subsequent events proved. Timeform shouldn't start throwing stones from within the glass walls of Halifax.
 
Last edited:
Desert Orchid said:
I think we are talking about a Timeform love-in. Not only that, but we're talking about Timeform being so superior to everything else around that everything else around is to be scorned as a waste of time, space and energy.
Click to expand...

No, we're not, and I'm certainly not trying to undermine the ratings of others, including yourself.

We're talking about applying rigorous methodology to ratings, an example of which is outlined in Simon Rowlands' article. He, and Timeform, are not alone in believing that some form of race standardisation, or at least pre-race expectation is used to quantify post-race ratings,The expectation of how certain key horses may perform, and the apparent overperformance of horses with proven limitations (for example) are going to be important indicators when reaching a figure, but that's rather more involved than always looking for a single individual on which to hang the entire race.

That's by the by, to a large degree, as my comments about DD are due to what I see as the nonsensical choice of Grandouet as a yardstick, rather than simply the use of yardstick handicapping per se.
 
Just my opinion, but race standardisation isn't nearly the tool that Timeform would have us believe, and the whole idea that the nuances of pace, ground, actual effort and class of opposition in an individual race, can be superseded by the maths of what happened in previous years, is almost laughable.
Every race is a unique event, and individual horses react differently to its circumstances, so the underlying assumption - that the same horses would produce the same result, year after year - would hardly be a tool to bet your boots on.
Yardstick handicapping, applied literally, does have it's limitations, but at least it's not hamstrung by some belief that races from years ago had any impact whatsoever on the result it's dealing with on the day.
 
rorydelargy said:
All handicapping absolutely does not involve using a yardstick of some sort. The hanging of a set of ratings in any race based on the notion that one horse in that race ran to form is convenient, but scientifically flaky. A more robust way of establishing the worth of a race is described here by by ex-forumite Simon Rowlands: http://bit.ly/hRdpkE

This isn't a Timeform love in by any means, but it's clear that the methods used by any individual who utilises race standards and applies rigorous mathematical principles are going to provide truer results than some bloke who decides to rate a race through a random horse because the result "feels right" to him.
Click to expand...

I agree with this really.

Simmo's comment about all handicapping "using a yardstick of some sort" isn't necessarily untrue though; in a sense race standardisation exponents (of which I am one) use the race itself as the "yardstick."

I would suspect the problem for many people trying to compile their own ratings is that race standardisation principles by their nature require an extensive back-catalogue of ratings around which to compile a standard. While this is one of the reasons that Timeform provides (in my view at least) a valuable service, it probably renders using the technique implausible for most others .
 
Last edited:
reet hard said:
Just my opinion, but race standardisation isn't nearly the tool that Timeform would have us believe, and the whole idea that the nuances of pace, ground, actual effort and class of opposition in an individual race, can be superseded by the maths of what happened in previous years, is almost laughable.
Every race is a unique event, and individual horses react differently to its circumstances, so the underlying assumption - that the same horses would produce the same result, year after year - would hardly be a tool to bet your boots on.
Yardstick handicapping, applied literally, does have it's limitations, but at least it's not hamstrung by some belief that races from years ago had any impact whatsoever on the result it's dealing with on the day.
Click to expand...

It's worth pointing out (before David Johnson does..) that Timeform don't use standardisation in isolation; they do take into account pace (albeit unsatisfactorily due to the lack of sectional timings) through Timefigures, ground, actual effort (including the vaunted Timeform squiggle - I'm not sure how anybody could quantify this to be honest).

While I'm at it, one area where they don't provide adequate coverage is Irish racing (particularly Irish flat racing). The lack of Timefigures is a major blow (and I suspect has to do with the difficulty of producing figures at the Curragh), but the race comments frequently lack incision and are average at best, particularly as Justin O'Hanlon and Kevin Blake have raised the RP analysis standard considerably. I don't know if focussing more on Ireland makes sense from a commercial perspective - the demise of Turform must slightly increase available market share - but your crew need to raise their game, DJ. :cool:
 
Trackside
I'm aware Timeform don't use race standardisation in isolation, but it's clear - both from SR's article, and Rory's post on the International - that the use of it is being promoted as setting the company apart from/above conventional handicapping methods.
I don't believe it does, and while I'd never condone using Grandouet as a simple benchmark, his form and his profile (judged on an amalgam of the criteria I outlined, rather than tabulated separately) will always be a better guide to what happened in the race than however many similar races were run in years previous.
 
trackside528 said:
Simmo's comment about all handicapping "using a yardstick of some sort" isn't necessarily untrue though; in a sense race standardisation exponents (of which I am one) use the race itself as the "yardstick."
Click to expand...

Precisely my point. Otherwise you'd just be throwing random numbers onto a page.
 
The only yardstick that counts is P&L.

It's largely immaterial how you reach your conclusions as long as you are making a profit on the back of them.
 
rorydelargy said:
but that's rather more involved than always looking for a single individual on which to hang the entire race.

That's by the by, to a large degree, as my comments about DD are due to what I see as the nonsensical choice of Grandouet as a yardstick, rather than simply the use of yardstick handicapping per se.
Click to expand...
I think DD's hands were tied by the circumstances surrounding the race.

Only three runners of consequence.
One of them coming in from handicap company.
The other two making their seasonal debut.
Heavy ground.
A moderate pace.
Conservative tactics.

In these circumstances, race standardisation can play no part as they wouldn't have been replicated at any time in this race probably in the last 20 years. This is quite different from a 20-runner maiden 2yo race on fast ground at Newmarket.

Would it have been acceptable for him to say the form of the race is impossible to assess with any degree of confidence therefore no ratings are allocated? How did Timeform evaluate this race? If they've used standardisation then they've allowed themselves to be slaves to their methods rather than masters of them.

The adjusted RPRs of the last ten winners of the race ranged from 162 (Rigmarole in 2003) to 180 (Rooster Booster 2002), yet these races were run in successive years. Would standardisation have had them closer? The average was 169 so has the handicapper merely rated Zarkandar an average winner and rated the rest of the race accordingly?

I'm not sure either approach would be acceptable. I think you have to look at the ability of the ones who contest the race on the day ather than those who contested it in previous years.

I don't agree with hanging the ratings around one horse either but you find much more often than not that if a race has been well contested and properly run that several of them will run pretty close to expectations. That wouldn't necessarily have been the case here although, in fairness, the front two have run close to expectations with only Rock On Ruby disappointing, if you want to be harsh. I think it is more important to base one's evaluation of Saturday's race on the visual evidence and this is where I rely on people with a better eye than mine. I wanted G to win so was a tad disappointed, even if he emerges best at the weights. Z impressed with his guts. Rock On Ruby appeared to blow up and ran a perfectly satisfactory race in the circumstances.

But as a rater, I want to allocate figures, so all I can do is see how the ratings work based on each of them running to form in turn and decide which outcome appeals as closest to the most likely indication of performance levels, which makes it all very subjective. I'd still want to treat the form with caution regarding future races and it wouldn't bother me too much just to leave the race unrated but ratings wold also allow me to see how each is likely to perform should they meet again in heavy ground, small fields, the likelihood of conservative tactics being employed, etc.

For what it's worth, Oscar Whisky's race remains unrated.
 
Last edited:
Interesting discussion, memory letting me down but not so long back weren't we criticising the handicapper for rating the race rather than the horses that ran in it?
 
That was essentially my criticism for his Tingle Creek ratings, Col - yes.

If he has used a race-standardisation technique to assess the Tingle Creek, then he does appear to have used it somewhat in isolation.
 
Grasshopper said:
That was essentially my criticism for his Tingle Creek ratings, Col - yes.

If he has used a race-standardisation technique to assess the Tingle Creek, then he does appear to have used it somewhat in isolation.
Click to expand...

Remember it's a different handicapper looking at the Tingle Creek and the International - it appears that some of the BHA handicappers use different methods, or at least appear to when discussing their findings.
 
reet hard said:
Trackside
I'm aware Timeform don't use race standardisation in isolation, but it's clear - both from SR's article, and Rory's post on the International - that the use of it is being promoted as setting the company apart from/above conventional handicapping methods.
I don't believe it does, and while I'd never condone using Grandouet as a simple benchmark, his form and his profile (judged on an amalgam of the criteria I outlined, rather than tabulated separately) will always be a better guide to what happened in the race than however many similar races were run in years previous.
Click to expand...

Hi Reet,

I'm more than happy to have Timeform's methods criticised, and am aware that not everything is rosy because it comes from Halifax. My role as a freelancer involves using their methodology as a basis for my analysis, but it's not inherent in my role that I act as a Timeform apologist, and I don't want it to appear that way. I'd be happy to have a decent discussion about handicapping techniques which don't have to descend into ad hominem insults, and I'm sure I've got a huge amount to learn.
 
Thanks Rory; I do hope you don't think my intention was to malign you personally in any way. :cool:
I do, however, have a bee in my bonnet about race standardisation being anything more than a minor tool, and would always rail against suggestion that it can give an answer* where the form book can't.
Cheers,
Reet

* I should qualify that to races with exposed form.
 
Last edited:
Back
Top