THE DERBY 2014

Sandown, I justify it with the overall predictive accuracy of my ratings compared to what's publicly out there. There are methods involved which I'm not prepared to go into here even if I could. I wouldn't expect anybody else to reveal the nuts and bolts of what they believe gives them an edge. Epsom is a fast drying track and as Ruby Walsh said before racing the rain was only on top of the ground and hadn't really penetrated. As for the Coronation Cup specifically, as I've already said there is margin for error, possibly the race was much slower run than I think. In relation to the Derby this makes no real difference as I know the 'ceiling' for the race due to the last race plus other factors.

so it's 'don't ask specifics, you'll just have to take my word for it' .... rather unsatisfactory on a horse racing forum don't you think

Grendel, I'm not saying you're wrong, if it makes sense to you that's all that counts. My point is that I believe Australia is already better than the bare winning margin over KH, no improvement required, he already has the ability. I'll be opposing Kingston Hill where possible in the best races.

Grendel, I'd have to show you loads of spreadsheet workings to prove my case, not really possible on here even if I had the inclination, which I don't. I thought the point of the forum was for people to use their own methods, ideas, etc, to put forward their opinions on probable outcomes of races and perceived value. How much other punters reveal about their own methods is entirely up to them, I'll judge them on the overall accuracy of their opinions.

We had 3 races over 12f on Saturday, so we can afford to leave the last race out if need be. Handicaps are always imo the most reliable for setting allowances and i have no problem using the same allowance for both the Derby and Miss marjorie's race. using the same allowance for the Coronation Cup gives a slower rating than you might expect for a G1 but this wouldn't be unusual for a small field race,especially as CDA is probably best at 10f and Flinstone was having first run.Doen't change anything as far as those horses are concerned for the future in any event.

Figgis, I enjoy reading what you have to say on here and actually think you contribute well to the mix.  You have your modum operandi and understand it well for your requirements as do I and I have no interest in knowing the nuts and bolts of your operation. Sandown asked you to explain how you reached the conclusion that Saturdays Epsom races were run progressively faster as the afternoon went on .... he didn't ask you to explain the big bang theory just a justification of your conclusion

Doen't change anything as far as those horses are concerned for the future in any event.

This is entirely my opinion on the Coronation Cup, I'm fully aware that I may have overcompensated for the ground in this case but as far as I'm concerned it doesn't change anything either way.

Grendel, it's not just that I don't want to explain everything (although that is a big part), it would be very difficult anyway. It would be just like you having to explain with accuracy how you arrived at a particular going allowance on a difficult card. Did you just go for a best fit of past figures for all races? If so, how did you arrive at those past figures? Did you average things out or take a particularly strong view of one race and rate the others around that? Etc. I imagine it would also be difficult for Sandown to explain fully his sectional methods in this limited environment even if he wanted to.

fair enough Figgis but it appears to me that you fit your going allowances to expected performances which sort of defeats the object of doing speed figures imo

If I'm right about Australia then, for me, it augurs well for his future prospects. The more a horse is fully tested in the Derby, or any big race ,the more it's likely to shorten its career at the top. Workforce, Camelot and Authorized all paid for having hard races in the Derby, either in the short term or long term. Australia was fully tested in the Guineas but could've had a harder race than he did in the Derby, imo.

Grendel, well expectations play a part, as they do with any ratings really, but I certainly wouldn't rate a race purely on expectations, as I've said I don't use race standardisation. Another part is actually watching the races (something which often gets neglected in the world of numbers) and how they were run. It struck me years ago just how many races didn't appear to be falsely run (or run too fast), the beaten distances didn't fit with falsely run races, the form of those races worked out extremely well in future yet the public speed figure ratings had told me they were slow races. Something obviously didn't add up. This led me to more research, which others might think is nothing but guesswork, well if it is I'll take it over the likes of Topspeed ratings any day.

Australia has improved every start with me, from 112 in the Guineas to 121 on Sat.

Thought they went a fair pace producing a decent Derby winner, almost certain to be even better at 10f (as his trainer / jockey admitted beforehand) and can clock a higher figure at that distance. Australia would also have decent prospects of reversing form with Night Of Thunder / Kingman if ever did drop back to a mile.

Kingston Hill took a massive 20lbs hike from 99 in the Guineas to 119 on Sat

Have last 5 Derby winners:

124 - Workforce (subsequent best 111 in the Arc)
121 - Australia
115 - Camelot (subsequent best 90)
109 - Pour Moi
87 - Ruler Of The World (improved to 105 after)

I was also struck by how many high rated races just weren't cutting it in future, with too many excuses put forward like they 'bounced' afterwards, which does happen but not as often is put forward.

New Timeform rating for Australia = 120p
Racing post = 125

Timeform = 129p excuse me.

Authorized an effortless Derby winner then gets beat in the Eclipse by Notnowcato crossing the track. He then ran a cracker in the Juddmonte beating Dylan Thomas but then didn`t turn up  in Paris where Dylan Thomas won totally reversing the York form. They are not machines but the better ones replicate their form more often than not. Frankel is top of the class in the replication stakes.

I think Australia will replicate more than most as long as the trainer states he avoids soft ground. Has to be a worry re the Arc.

Felt if you read any of the 'Coolmore' thread you will know I have serious reservations about their race selection process. They said Camelot was bottomed in the Irish Derby on heavy ground then ran him in the Arc on the same going. And his selection process with RVW left a great deal to be desired. I hope they have learned from their mistakes and noticed what the other tops horses have done. If the Arc is soft there is always either the Champion Stakes or the Breeders Cup turf. Not the BC classic.

Far be it from me to question trainer decisions but I was amazed Camelot was put in the Arc particularly after he had had a hard race in the St Leger.

Still he was 2/1f and did beat his stable companion St Nic who is the same horse as CDA so couldn't have been that bad.

More than half the field beat SNA that day, including 150/1 shot Haya Landa buy about 7 lengths. I don't think any punter would read that form literally.

*by

No? A 2/1f and a 14/1 shot. Are you sure? Beat him 3 lengths on ground they both hated. Both rated about the same but Camelot had the 3 year old allowance. I would think that part of the result was pretty accurate.

I would think that part of the result was pretty accurate

How can you take a small section of a result and say that part was accurate? While ignoring that it was run on heavy ground and both horses were thrashed 12 lengths plus by the moderate winner. It was either true form or it wasn't.

What on earth are you talking about F? On that ground I suggest that is as good as Camelot is. And on that ground I suggest that is as good as St Nic is. So when they run on that ground against horses of that calibre I expect them to be beaten. So yes I think it was true form. Don't you?

What on earth are you talking about F?

Still he was 2/1f and did beat his stable companion St Nic who is the same horse as CDA so couldn't have been that bad.

That is what I'm talking about, a thoroughly ludicrous analysis of the form.

I think you will find CDA and St Nic are about the same horse on FORM. That's why St Nic was a short priced favourite to beat CDA in last years King George before his injury. You may think it is ludicrous but that's what the formbook says. Like it or not.

So let's get this straight, you acknowledge that St Nic and Camelot are both unsuited to heavy ground, yet you say Camelot couldn't have been that bad to beat SNA as he is the same horse as CDA, while wilfully ignoring the fact that CDA positively thrives on heavy ground. You've got to laugh I suppose

Do you think so? I think potentially his best form has been on good and faster ground. I know he isn't as inconvenienced as some are on soft ground but I don't think the races he won on very soft ground he wouldn't have won on good ground. I know that is not what all the pundits say but that's what the formbook says.

I agree, he thrives because the others are inconvenienced, but the point is you're comparing the form of two horses that were very much inconvenienced to another horse that wouldn't have been in the slightest.

For me it's a travesty that geldings aren't allowed to compete in races like the Arc. Yeah I know it's what the breeders want but it just means we're crowning champions that may not have necessarily been so without potentially better horses being excluded.

Oh dear. That is so much rubbish. St Nic finished 5th and 14th in the Arc and that's about as good as CDA is. He, St Nic, wasn't good enough to win a classic and was only good enough to win the peripheral Group 1's that are uncompetitive and not necessarily the target of others. Just like CDA. You really don't want to believe all the tripe that's dished out F. 

And Solemia?

It's plain to see looking through St Nicholas Abbeys form that he was much better going left handed than right handed so citing his Arc form as being representative of his ability is a bit misleading, and what is this ridiculous notion that St Nicholas Abbey and Cirrus Des Aigles are the same horse on form .... Cirrus is/was lengths better than SNA in optimal conditions

And the relevance of this discussion to this thread is what exactly?

and figgis, if you want some indication that the derby wasn't slow early on just watch the run of Red Galileo, a 100/1 shot always in last place until he ran down the centre of the track all the way up the home straight to snatch 5th place to record a RPR of 110, 11lbs higher than his previous best ... wouldn't have been able to do that off a slow early paced derby

Grendel, I have Red Galileo just franking the figure he put up in the classic trial, nothing more. As I said, I'm not saying it was a very slowly run race, I have it as an average type Derby winning rating, that couldn't have been achieved if it was very slowly run. I just believe that it could've been run a bit quicker and so could Australia in those circumstances.

if it was run any quicker the final time would have been slower as it was already over-paced early leading to a slightly lower than average final time figure ... Australia ran very close to even paced, much closer than the runner up did

Grendel I have Farhh and St Nic as the same horse based upon WHEN THEY MET and I have St Nic and CDA as the same horse based upon WHEN THEY MET and I have Farhh and CDA as the same horse based upon  met. Like it or not that is what the formbook  says WHEN THEY MET.

Grendel, so you think KH would've finished even closer given more even circumstances? I strongly disagree but it'll be interesting to see what happens.

Figgis, I strongly believe swap the draws, swap the result .... you really rated Red Galileo's 6th at Lingfield beaten 7¼l the same as his 5th in the Derby btn 9½l ..... I have the two performances as about 9 lengths different

And I don't buy the left handed right handed scenario either Grendel. That was just circumstantial. One of his better pieces of form was at Ascot. I just think they rode him wrong and that is something they said about later on in his career. And in the Arc under optimal conditions he finished 5th narrowly beating Sarafina who had just narrowly beaten CDA. Confirming their positions just about equals.

Anyway I'm off I promised myself I wouldn't get dragged into another argument. The point I intended was that CDA had won the Coronation Cup named after St Nic etc. So I'll leave it at that.

Grendel, yes, but remember I'm making no allowance for the extra distance. My findings are that in general the better 8-10f horses lose some speed over the 12f, it's less of a factor with slower 8-10f types, the extra distance can be a leveller, which is partly why I don't like to give a uniform allowance. I've also found that there isn't as much difference between wide margin impressive Derby winners and average ones as I used to think years ago, but I'm not expecting everyone to agree, nor would I want them to, if everyone agreed with me I'd have to give the game up

Australia won on merit, but pleased to see a good performance from Kingston Hill, easily 2nd best in the race. Wonder how far KH might stay and whether the Leger might be an option as he seemed to get a little outpaced for a while as things started to hot up, before getting in to gear, and was noticeably comignback at Australia at the business end.

12 lengths plus  to justify being a 4/7 chance.

Sandown, as I've said before I disagree that a horse necessarily has to have lengths and lengths in hand to be an odds on shot, for me a big factor is the confidence in form being franked and that doesn't necessarily have to go hand in hand with pounds or lengths advantage. However, I accept that it's the way you choose to price up a horse's chance. Nevertheless, I do find it unusual that you think a horse that would in theory win 7 out of every 11 runnings of a race would need a 12 lengths plus advantage. It's highly unlikely that the draw or luck in running would disadvantage a horse by as much as 12 lengths. It's possible that another horse could improve more than 12 lengths to beat it but that would also be a very rare occurrence. So what would the 4 losses out of 11 be mainly attributed to? Would that be your expectations of the horse running way below par?

From my experience of racing I'd say it's rare to see a horse with 12 lengths in hand but I see many 4/7 shots and lower, do you think most of these horses are underpriced and do you therefore lay a lot of them?

If you have prices of 8/1 about each runner in a 9 horse race then each one should win 1 in 9 times according to the market, you'd also expect if the market is correct to get a pretty close finish each time the race is run. So I do agree with Sandown that the result of the race can then be used to see how accurate your pricing up was. But 12 lengths does seem quite a lot imo. I think the market got the price right with Australia but Kingston Hill and Romsdal outran their prices based on that result though that's not taking into account ease of victory.

SG, having had the benefit of watching the race and pricing it up on a theoretical re-run, what price would you lay Australia?

I'd offer you 4/7 figgis, would you take it ?

No value for me, SG, don't think I'd be laying any bigger than 1/2 though.

Its all mathematically based Figgis, drawn from data over years showing how relative ratings advantage can translate into chances of winning which in turn can be turned into prices. The reverse is then possible, taking prices and turning that back into chances and pounds advantage turned into lenghts.

I wouldn't take the figures literally as a horse winning by 1/2 lengths but on the bridle doing it easily might be the equivalent of 12 lengths.

So, if I rate a horse a 4/7 chance (very rare occurance) the chances are that it has a large advantage and/or little in depth competition.

There are plenty of times when a market becomes distorted by various factors and the Derby this year was one of those times imo. Using the same method after the outcome helps me to see whther the actual price retuned was a fair reflection of a horses chance. As a matter of interest, I found that oz ran like its SP i.e a 6/4 shot whereas KH ran like a 5/2 chance so in that sense it had greater value and represented each way value. I didn't rate it a 5/2 shot incidently but I did see it as being value and had a small ew bet on it.I did rate Oz a 6/4 chance and I back it but not because it was value but I wanted to cover my a/p liabilties and show some profit for all the effort involved. I spread my risks I guess.

Sandown, that's fair enough, but it's this bit I'm interested in

So, if I rate a horse a 4/7 chance (very rare occurance)

4/7 (and much shorter) chances in the actual market are not that rare, wouldn't you consider them to be a lay if they don't have at least 12 lengths in hand?

Yes I would but I would need to be of the view that their edge was considerably less. When I lay ahorse, like say Dawn Approach in last years Derby, I need to a strong opinion on a key factor e.b DA not staying 12f. I'm not a natural layer of horses and derive no non-financial satisfaction from doing so and therefore i tend to look for a lot of value in that case.

Again, fair enough.

I would have layed you 4/7 on Oz until the cows came home, not because I didn't thin it wouldn't win (I thought it would0 but because i was of the strong opinion that that would have been the wrong price. I did lay 5/4 and backed it back on the day at 7/4 so you can see that any odds on would have been va very atrractive lay. I know a few big hitters who did just that, incidently. (I don't count myself in that category, if i ever did)

I know that football odds can be priced by goal advantage through statistics and racing can obviously be viewed the same way, but I believe there are more nuances involved in racing that can give you an advantage over the averages of statistics. Maybe there are in football too for some people.

The narrower the spectrum of chance you believe a horse to be, the more you'd need to know about it.  I don't believe it's possible to be confident that a Derby runner is specifically a 4/7 chance bearing in mind the lack of information you have about the horse, I said before the race I thought it was between evens and 6/4 ... that's as specific a chance as I was prepared to give it.  In this years prix D'isphan, I rated Cirrus Des Aigles about a 4/9 chance (so I backed it at 5/6) with his main rival Olympic Glory needing a fast pace to run at which he wouldn't have got in that race plus other factors.
Put simply, I believe there was not enough available information about Australia to categorically say he was a 4/7 shot, 14 similar scenarios would not result in anything like 8 victories in my opinion, it just so happen that this year it was but that says nothing about long term success.

By the very nature of the race there is very rarely a ton of information about Derby runners, it has never stopped me beating the odds in the long term, and that has been done by usually backing no more than one horse. So either I've just been lucky and am overconfident or I've been doing something right. That is going to sound big headed, maybe it is, but it's all about what races you specialise in and how good you get at your speciality. People seem to have the attitude that just because they can't see why a horse should be a horse price then nobody else can. I have a garbage record in races like the Cambridgeshire, rarely have a bet in it in truth, but I wouldn't go thinking that nobody else can gain an edge in it where I couldn't.

Stevie G

But 12 lengths does seem quite a lot imo.


Shergar won by 10,2,2 lengths and I would rate that a 4/7 shot

*horse should be a short price

Don't get carried away Figgis, unless you're very old then the sample size of your record in the derby may not be big enough to give a true refection of how good you are on the race.

Figgis, how many derbys have you backed a runner in, and how many winners from how many selections?

People seem to have the attitude that just because they can't see why a horse should be a horse price then nobody else can


You often take comments very personally as though I and others who disagree might be knocking your methods but this isn't really the case. It's more a case of what I know about the market and pricing may not fit with your view. You may have backed X% of the last umpteen Derby winners but that's a small sample and in any event each race is different. As I said before, prices and markets is the not the same thing as confidence level. i tried once to measure my degreeso of cinfidence in outcomes and found that I ended up over-stating the case for horses which I thought could win. Not saying that you do, but in the end I would say its all about the price you take and the stake you put on.Nothing else matters.

Grendel, this is just going to sound like aftertiming, but ever since Oath and including him, when some of my methods changed, the only decent sized confident bets that went astray were Fame and Glory, regrettably I changed my mind about STS staying , and Battle of Marengo when I made a complete mess of it last year. I had a small bet on Alamshar when Kris Kin won, had no bet in Sir Percy's year, and small bets on Kandahar Run (ante post) and Bullet Train in the years of New Approach and Workforce. Camelot was a standout but too short to interest me. Despite not backing Kris Kin, Sir Percy, New Approach and Workforce they were all either top rated or second top rated on my figures, which is nothing unique, I imagine a lot of other figures compilers rated them similarly. There were two sizeable ante post bets after debuts, one on Galileo and the other on Frankel. All other winners during that period were either well backed after trials or the day of the race.

There isn't really much point in stating all the above but it's there as it's been asked for. My only regard is my own confidence in what I do, I'm not trying to sell methods or tips, nor would I want to.

Sorry, Frankel was after Doncaster, not debut.

Another important point is whether the sample size is considered too small or not, it is only the same methods that I use for any other race I concentrate on, where there is a far larger sample, the only difference is the regular step up in trip involved with Derby analysis, but in general I don't think that's as difficult to factor as some people suggest.

i tried once to measure my degreeso of cinfidence in outcomes and found that I ended up over-stating the case for horses which I thought could win.

I don't think I've ever found that to any great extent, even years ago strike rates were always good compared to prices. My problem when younger was having no staking method and overstaking, which is just as bad as, or worse than, overstating a horse's chances. Some punters seem to have been born with the Phil Bull emotionless accountancy attitude to punting, not me, it was sh1t or bust

As a matter of interest, I found that oz ran like its SP i.e a 6/4 shot whereas KH ran like a 5/2 chance

Sandown, are you saying that's how you would price up the Derby field in future if they all reopposed?

Personally thought Australia was never going to lose despite not being convinced by its stamina in the end, so still pretty sure it would win at least 7 in 11 runnings ie a 4/7 or 8/13 chance.

Kingston Hill could keep returning for a match against Australia at any distance between 8-12f and would have to wait until Australia underperformed before winning one.

Given the likes of Orchestra, Geoffrey Chaucer, Western Hymn, Fascinating Rock, Ebanoran  are probably all better than they showed on Sat, does that really make Kingston Hill a 5/2 chance? Or an even money shot betting without Australia?

One thing I will mention is sometimes I back certain horses that I believe will either bolt up or flop
depending on certain things and in this way judging if they were value or not based
on the result would not work. Really if your pricing up of a race is correct then horses you price up as
2/1 shots  should win 1 in 3 times and 9/1 shots 10% and so on.

Looking back at the derby, there was always a chance something in the race would step
up on previous form and that horse happened to be Kingston hill but if you had rated Australia
far enough clear then the chances of that improvement being enough to beat Australia was
unlikely to be enough.

It's very hard to say exactly what price Australia should have been and I personally felt even money was
about right but  I really don't think we can say for sure.

What we can do is look at our own ratings or form study and ask ourselves did we allow for enough improvement for Kingston hill
stepping up in trip and did the speed Australia show in the guineas mean we shouldn't expect as much
improvement. What sort of performance should we have expected from those horses that
ran in the slowly run derrinstown. I found them very hard to weigh up.

pedrobob

Sandown, are you saying that's how you would price up the Derby field in future if they all reopposed?

No, I'm saying that giving the result of horse A beating horse B etc, and using the same method for converting ratings to prices that I use BEFORE a race, then that's how I would rate the horses given the relative positions and distances beaten.

Personally thought Australia was never going to lose despite not being convinced by its stamina in the end, so still pretty sure it would win at least 7 in 11 runnings ie a 4/7 or 8/13 chance.

Your opinion. Purely on the bare result I would say 4/10. That may seem low but across thousands of races I would be more right than you based on historic data.Of course, nothing to stop you building more into what Oz had in hand and adjusting for that and personally I would upgrade but we then become more subjective and prone to see what we want or expected to see. Fact is, Oz did not live up to expectations based on AOB's claims and whilst he won, he ran below the performance many if not most expected.


Take Oz out of it for the moment and judge KH separately. Not a bad winner,albeit  not a great winner, but a Derby winner nevertheless. Have to have him around the 123-125 mark wouldn't you?

The difference in their respective prices was not a fair reflection of their merits and that is what we are judging.

Stevie G

Looking back at the derby, there was always a chance something in the race would step
up on previous form and that horse happened to be Kingston hill

Before the Derby Oz was rated RPR of 122 and KH was 119. Difference afterwards was 2lbs (125 and 123.) Therefore, KH didn't improve more than 1 lb relative to Oz whereas Romsdal improved 5lbs, Arod 0lbs, Red Galileo 14 lbs.

I think that many downrated KH's chance because of his Gns run, because his RPT win didn't generate a fast timefigure and because the ype surrounding Oz had raised expectations of a vastly superior improvement. The figures don't show that.

You can argue all you like about the merits or otherwise of RPR's or the merits of your own figures, or anyone else's, but the thing about collateral ratings is that they stand the test of time. I use my time figures to try and get an early notice that a collateral figure understates the merits of a horse but in the long run I expect the collateral figures to reach the same level as mine might indicate. In the main, that is.

Sandown,

Can you really trust the RPRs though given that you said yourself that Marvellous was 30lb below the figure you gave for Imagine and RPRs rated them only 2lb difference. You also pointed out last year how a horse in a sprint at fairyhouse had been given a rating far in excess of anything it had done before. These RPRs often come up within minutes of a race finishing and I just don't trust them.

Your opinion. Purely on the bare result I would say 4/10. That may seem low but across thousands of races I would be more right than you based on historic data.

But surely giving opinions on here is the whole point? We can all see what the stats are and I know some people base their betting around the averages of statistics but there are many who try to beat those averages. It's like saying if you back a certain top trainer's runners blindly you'll win 20% of bets at a loss, whereas somebody may specialise in a particular yard and increase that strike rate to 40% at a profit.

Stevie G

Can you really trust the RPRs though given that you said yourself that Marvellous was 30lb below the figure you gave for Imagine and RPRs rated them only 2lb difference. You also pointed out last year how a horse in a sprint at fairyhouse had been given a rating far in excess of anything it had done before. These RPRs often come up within minutes of a race finishing and I just don't trust them.


All figures are wrong to some degree. There is no right answer and what I distrust the certainty that many have that their figure is the correct figure.At least with RPR's you can go onto the site, check which horse they have gone through to get their level, and adjust it if you wish. By doing your won figures you can better understand how your views might differ. The thing about collateral calculations is that there is much less room for error than with with time/speed figures, official handicap ratings are closely match them, and RPR's are easily accessible.

The room for error producing time figures is huge and I would have a much greater standard error for them than for collateral figures. DO I trust my own figures? Not entirley. When I come up with one which shows a big upside/projection I know that I may have egg on my face but at least I have been given an early indication that a horse may be capable of great improvement. The calulation might contain errors re going allowance,course allowances, distance allowances etc and with only 6 races on a card there isn't much of a sample. How do you know that you are using the right anchor? You don't. Perhaps talll the races were run fast yet you don't want to have all the winners improve. Perhaps they were all run slow, in which case you will use a heavy going allowance. The horses may not progress for hundreds of reasons. There is doubt everywhere.


For me ratings are only part of my approach anyway. They get discussed on this particular forum where most posters are advocates of time ratings so races get discussed in those terms. That leads to a particular view of the world. There are other views out there.Just ask brigust1.

But surely giving opinions on here is the whole point? We can all see what the stats are and I know some people base their betting around the averages of statistics but there are many who try to beat those averages. It's like saying if you back a certain top trainer's runners blindly you'll win 20% of bets at a loss, whereas somebody may specialise in a particular yard and increase that strike rate to 40% at a profit.


I don't disagree with that.

How do you know that you are using the right anchor?

This is even more of a problem with RPRs. I don't follow them but I get Raceform and read the comments, such as "the race is rated around the third", such a crude method has more pitfalls than anyone would encounter with a good understanding of time ratings.

This is even more of a problem with RPRs. I don't follow them but I get Raceform and read the comments, such as "the race is rated around the third", such a crude method has more pitfalls than anyone would encounter with a good understanding of time ratings.


No, that would be too crude. Most ratings, certainly RPR's, would use an averaging method using all the runner but dropping those runners which run way down the field and would bias the figures too much. That's not difficult to do. Where collateral ratings are at there worst (most opportunities if you like) are with lightly raced horses where the opening ratings given are based on race standards, where there are many unrated horses, and where the pace is extremely slow. I do find that the Top Speed figures are more unreliable because I think setting race standards can be the biggest cause of errors in the first place and I'm not happy with RP standards. Nor do I like some of the allowances. There is much more room for individual input and adjustment imo in time figs whereas RPR's can be more successfully automated.Pros and cons for both. But given that the collateral ratings are the first step in forming a market then its in those where we find the opportunity to disagree and form our own judgement.

Sandown, I know Topspeed and Raceform speed figures are based on different methodology but as far as I know RRs and RPRs are now one and the same thing.

Figgis, I think those comments aren't from the ratings compiler but the guy doing the analysis. yes sometimes I read in the RP analysis the form can be rated around such and such a horse but I don't believe RPRs do that.

I'm not too sure about that, SG, I've even heard Smith explain some official marks based on the same reasoning.

I've heard Smith say that too but that may be  because its easier to get across to the public to say that. Tf use race standardisation for sure and Simon Rowelands has explained why they do. My experience with RPR's is that they use an averaging method.

Having said that, it's surprising how often it averages out to place the 3rd/4th horse as the one which is shown to run to the same level of foorm so for quick analysis its not far wrong to use the third horse in a 10 runner race & pro rata from there.

Sound logic and methodology has to be at the heart of any ratings system, whether time based or otherwise, but ultimately I still say the overall accuracy lies with the skill and experience of whoever is doing the compiling. Nobody can get it right all the time but some will be right more than others, which is the aim. Taking those ratings and applying them, along with other factors, to future races and a suitable betting strategy is another matter.

It's a trade off between accuracy and degree of edge on the one hand and use of time on the other. Personally, having spent years on ratings development, I think the degree of improvement possible is now very marginal and much greater returns are to be found in pricing, staking, money management and risk assessment. Reducing the volatility in results, (reducing losing runs) spreading the risk (playing more than one horse) enables one to handle stake increments, plan ahead, reduce stress. A trade off in accepting a reduced average ROS% in return for more predictability/lower variation can deliver big advantages. All possible now with betting exchanges but not so preBF. Also, I would add that the market itself is much more of a factor than ever before in predicting outcomes than it ever was.

Further to my last post, it's great on the A/P forum to discuss real horses and real races with other enthusiasts (as opposed to the General Forum) but in truth the numbers who participate is pretty small, although more will read than contribute. The debates may get heated at times, and we obviously get on each others nerves at times, and all hold strong opinions. But in truth, we're not taking from each other, we don't really influence the market. The big players are either using automated systems and wouldn't know a thoroughbred racehorse from a donkey, much less care, or they have computer handicapping systems with teams of people inputting, or they are large players attempting to manipulate markets in order to trade and hedge, and we are the guys who help set the prices for them because we are opinion players. It's not so much dog eat dog as whale eat shark eat fish. There's still gain & joy to be derived but the world is a whole lot more than it was 20 years ago even.

I'm sure there is room for improvement on what is publicly out there. When you think of great minds applied to other areas of expertise compared to the limited amount done with time ratings I don't think there's much comparison. Spreading the risk is a good strategy but if you can operate at a reasonably high strike rate there is no need, as long as you're beating the market. I agree that the market is more accurate but to me that's mainly at lower levels. I don't find it much more accurate in big races than it used to be. Ok I can think of early morning ricks in Gp1s from the past (Kissing Cousin, Selkirk, etc) that you wouldn't get now because of bookies unwilling to take a view of their own now but in general there isn't too much difference.

I always find it interesting when we hear pundits on tv say that the market is more accurate because of more intelligent minds involved in the market. We then hear the pundits say they fancied such and such a horse but are now put off because of the drift. They've seen the horses in the paddock and go down to the start so they know that isn't always the reason for a drift. So why does the drift put them off? Is it because these supposed smart minds have spotted something in the horse's form they haven't seen themselves and are playing late? I'd suggest more often it's due to the influence of those with inside knowledge not fancying the horse on the day. Which is something that could easily be disguised by bookmakers years ago but not anymore.

I agree that the market is more accurate but to me that's mainly at lower levels. I don't find it much more accurate in big races than it used to be

Don't agree but I don't wish to reveal any more on this. Sorry.

the market is more accurate because of more intelligent minds involved in the market.

Can't respond as to whether this is being said i.e. about intelilgent minds, but don't agree that theyhave to be . The exchanges are just more sensitive and we can read for ourselves, although there is a hell of a lot of mis-direction at play.

I'd suggest more often it's due to the influence of those with inside knowledge not fancying the horse on the day. Which is something that could easily be disguised by bookmakers years ago but not anymore.

I agree with this statement but not all is revealed through the exchanges, and anyway BM's have learned how to use the exchanges to manipulate SP's which before would have been done solely through the on-course market.

Don't agree but I don't wish to reveal any more on this. Sorry.

Sandown, I don't wish you to reveal anything, I just generally don't find it any more accurate compared to my own analysis and the horse's I'm backing, which is what matters to me, not a statistical average

anyway BM's have learned how to use the exchanges to manipulate SP's which before would have been done solely through the on-course market.

I'm sure that's true, but they can't do anything about those with inside knowledge laying horses, which certainly goes on whether those on tv want to acknowledge it or not.

I guess that as you find the market more accurate compared with your own analysis that is why you have seen the need to adapt your methods to backing a spread of horses and trading? I've seen no need to adapt mine so far, and to be honest trading doesn't really interest me at all, good luck to those who do it, but I always think it's better to enjoy your work if you can and that wouldn't do it for me whether I'd be any good at it or not. I also think that whatever you're doing there will come a time when you're not as good as you used to be, or the competition gets too hot, or you can't adapt to change. Hopefully I'm ok for a few years yet, assuming I've still got the appetite for the game, but you never know how things will develop and what I see as my edge could be rubbed out in the future.

I guess that as you find the market more accurate compared with your own analysis that is why you have seen the need to adapt your methods to backing a spread of horses and trading?

No, that's not the reason. I'm not a big trader either.But I do invariably back more than one horse as a rule.

I think you should always be looking to improve your methods even if you are making a good living from the game,
Figgis I notice you say you didn't do well on races like the cambridgeshire, so you could try other methods on that race though given how tricky that race is you're probably just not giving it much of a chance, you can't expect to me be winning on races like that as often as the derby.

SG, I suppose I'm just not that interested in races like the Cambridgeshire from a betting point of view. I prefer to concentrate on races that appeal to me and I know I can get something from overall. If I was putting effort into races like the Cambridgeshire I don't think I'd have made advances in the areas I feel I have, you can only do so much. It's also because I'm not interested in operating at low strike rates which would obviously come with such races, unless I was backing a few, which doesn't appeal to me and would also involve giving up some value.

Yes some races you just don't get a feel for, I tried looking at the belmont stakes the other day and just didn't know where to start. Didn't know the horses, course, trainers, breeding or anything so quickly gave up. .

Same here. I only take a casual interest in American racing. When I think of how long it's taken me to pick up on certain nuances involved in our racing that can give you an edge how can I expect to compete against local punters with years of knowledge? And the truth is I just don't have any interest in playing catch up. I suppose using valid statistics it would be possible but not with any real insight on any particular race. I don't know how it's possible to spread yourself so thin, maybe some people are doing it but I'd expect it's mainly with the help of stats and algorithms. However, I reckon anyone who specialises in a particular area is going to be more knowledgeable in their chosen field (if they're any good) than somebody trying to cover all areas.

Aidan ‏@Aidan_m_  7m
Australia confirmed a runner in the Irish Derby by Aidan O'Brien provided the ground remains good or quicker according to the RP.

They are big supporters of the race, so if its the Ire Derby (granted fast ground) then so be it. Would have preferred to see him over 10f at Sandown myself.

I found that oz ran like its SP i.e a 6/4 shot whereas KH ran like a 5/2 chance so in that sense it had greater value and represented each way value. I didn't rate it a 5/2 shot incidently but I did see it as being value and had a small ew bet on it.I did rate Oz a 6/4 chance and I back it but not because it was value but I wanted to cover my a/p liabilties and show some profit for all the effort involved. I spread my risks I guess.

Sandown, does this view mean that you will be keen to back Kingston Hill at 3/1 against 1/3 Australia in the Irish derby?

A reasonable question to ask, I guess. Whether I would be keen to, I don't know, but at the prices all other things being equal, then imo the value lies with KH. However, I suspect that the ground, pace & tactics at the Curragh will suit Australia more which evens things up a bit.At the very least, I will not be backing Australia at that price.Whether  I oppose him will depend on the final field and conditions.

You have to consider the trainers here also. KH is Roger Varian's first Derby horse, and first real Group 1 horse, so it remains to be seen how he can manage to keep the horse consistent and competing throughout the season. He may do that with aplomb. He may not. But we know AOB can do it, at least up until September time.

I`d heard Australia was definitely going for the Eclipse

Glad he opted for the easier option.