VDW examples analysed via the approach he showed in his 1988 Mackeson discussion (Pegwell Bay)

[JennyK]

There has been some discussion of the 1988 Mackeson approach on the "Taking A Fresh Look" thread but to save cluttering that one I'll post further examples here. Discussion of the 1988 Mackeson analysis and a couple of others can be found there.

The essence of the 1988 Mackeson approach is two numerical pictures which VDW described in some detail and a third he did not describe but is a necessary inference given everything else he wrote.

The first numerical picture is as set out in VDW's article of 28/03/81, item 39 of "The Golden Years of VDW", and results in an initial short list of horses meeting two qualifying characteristics: in the first 5 (non handicaps) or 6 (handicaps) of the betting forecast and having one of the three lowest last three race placings totals from within the 5/6.

The second numerical picture is set out in the 1988 Mackeson discussion and appears to consist of a way of reducing the qualifiers from the first stage where there are more than three by taking the three with the highest ability rating (win prize money divided by the number of races won) but also augmenting them in two possible ways. First, by adding any horses from outside the first 5/6 of the forecast who have much the same last three race placings totals and ability ratings as the remaining qualifiers from the first stage. Second, by noting the horses within the field who, from their last races, came from those with the three highest class values (win prize money), not already noted as second stage qualifiers.

The inferred third stage starts with the qualifiers from stage two, ranked in descending order of ability rating and assessed for their form status - either a form horse or not a form horse - with the final list reduced by any who fail on the "other factors" by which it can be inferred that VDW principally meant going, distance, weight (in two distinct ways) and course type.

Using that methodical approach, and assisted with the comments on each of the Mackeson runners VDW gave, leads inexorably to Pegwell Bay as the class/form horse, as it does to Nomadic Way, the winner of the Cesarewitch in the same year, which is unsuprising as VDW specifically said it would. It also finds another winning class/form horse claimed by VDW in the following year, Braashee.

It appears that VDW did not give any selections publicly in the early 1990s but provided six from 1995 and 1996, of which one was Valiant Warrior, the winner of a class 109 handicap chase at Newcastle on 17/02/96. VDW said that the six were found "differently from anything I have shown to date", so it would be no surprise if the 1988 Mackeson approach did not find Valiant Warrior. Result of trying in the following post.

 

[JennyK]

Assuming his selections were made pre race (though he never proved that they were), VDW necessarily had to work from the forecasts (he seemed to use the Life's or the Mail's) rather than the returned betting. I don't have the forecast for the Valiant Warrior race and have had to use the returned betting.

The first six, with their last three race placings, were:

3 Easby Joker
8 Chief Raider, Gnomes Tycoon
9 Strong Deel, Valiant Warrior
12 Northants

so five go forward to the second stage.

First the subtractions via ability ratings

41 Gnomes Tycoon
38 Strong Steel
32 Valiant Warrior
28 Chief Raider, Easby Joker

So Chief Raider and Easby Joker are discarded.

Then the possible additions: no other horse from outside the first six in the forecast has a comparable last three race placings total and ability rating comparable to the three qualifiers so far.

The horses coming from their last time out races with the highest win prize money are:

104 Gnomes Tycoon, Strong Deel, Valiant Warrior
72 Northants
71 Snitton Lane

The first three are aready stage two qualifiers. Northants and Snitton Lane join them.

Thus at the third stage there are five possibles, in descending order of abiity ratings:

66 Snitton Lane
41 Gnomes Tycoon
38 Strong Deel
32 Valiant Warrior
29 Northants

Of the five, Gnomes Tycoon, Strong Deel and Valiant Warrior are in my interpretation of VDW's form horse criteria, form horses, and Snitton Lane and Northants are not, so three left:

41 Gnomes Tycoon
38 Strong Deel
32 Valiant Warrior

all coming into the Newcastle race from the same one a fortnight earlier where they had finished 2nd (Strong Deel), 3rd (Gnomes Tycoon) and 5th (Valiant Warrior). A notable feature is that while both Gnomes Tycoon (4/1 2nd favourite) and Valiant Warrior (5/1 3rd favourite) were near the front of the betting, Strong Deel (33/1) was the outsider of the eight and running from 8lb out of the handicap. It also seems as though Valiant Warrior not pushed hard when he lost touch approaching four out: the form book comment is "off the bit a long wayfrom home, finally gave up four out". He didn't give up in the sense of being pulled up but clearly came home at his own pace, though 30l ahead of the 6th (and last).

On the face of it, this race would, on 1988 Mackeson lines, lead to the conclusion that Gnomes Tycoon was the class/form horse. Although there are reasons to think Strong Deel's lto place was something of a fluke, Gnomes Tycoon had a slightly lower last three place total than Valiant Warrior, a markedly better ability rating and had beaten him well lto.

Thus if VDW was using his 1988 Mackeson approach with the 1995 and 1996 selections, which in itself is a big IF, the "differently from anything I have shown before" must lie in why VDW preferred Valiant Warrior to Gnomes Tycoon.

A cynic might say, because he won, and note that Gnomes Tycoon fell at the first (and never ran again), and suggest that had GT stayed on his hooves and won, Gnomes Tycoon would have been one of the 1995/6, not Valiant Warrior. But there may be a consideration here which would explain VDW's selection in a genuine way, which could be the "differently from anything Ihave shown before" matter to which he referred.

 

[Chesham]

LTO TF Analysis

STRONG DEEL (IRE) 's recent form didn't inspire and he was 8 lb out of the handicap, too, but he ran the race of his life, finishing strongly, having been held up; it will be interesting to see how the handicapper reacts to this performance, for if he chooses to ignore it then Strong Deel will take some stopping next time out.

GNOME'S TYCOON looked in good shape and ran a fine race on his first outing for twelve weeks, taking a keen hold and moving up to challenge the winner entering the straight, but not so fluent as that horse three out and two out, and unable to do any more.

DARK OAK had more on his plate than when winning last time and couldn't go with the leaders entering the straight.

VALIANT WARRIOR is usually so consistent, but he turned in a lack-lustre display, pushed along from six out and clearly fighting a losing battle on the turn.


analysis after VW won NTO

In the end, not the strongest of handicaps for the money, with second, third and fourth all looking to have improvement to find coming into the race. VALIANT WARRIOR , runner-up in this race the previous year, put a rare poor effort behind him with an emphatic performance, leading three out and ridden clear from the next, though as none of the placed horses had obvious chances at the weights beforehand, it's probable he didn't improve so much as it might appear at first glance.

 

[Pitmatic]

JennyK

Just to add to your post by providing the Sporting Life and Daily Mail betting forecasts.

Valiant Warrior doesn't appear in the Daily Mail forecast, but does in the first six of the Sporting Life.

The returned betting does correspond to the Sporting Life forecast, so the initial picture you've outlined is the "correct" one.

 

[JennyK]

Pitmatic

Thank you.

The one thing I have always found concerning in VDW, prior to getting to grips with the 1988 Mackeson article, was his apparent reliance on the inevitably subjective betting forecast. The various possible additions made at the second stage of the method as set out in the 1988 article to a large extent mitigate that. It always feels particularly satisfactory when one of the two forecasts to which we know he referred (and of course until it closed the Sporting Chronicle's), matches the returned betting.

I've just got access to a Haig for the 1980/81 NH season, so before trying another of the 1995/1996 six, I think I'll check out the 07/03/81 five to see if they work via the same methodical approach or whether perhaps the problematic "Erin numbers" played a part. I've got what I assume are the Life forecasts so can check all five against the returned betting.

 

[Tufnel]

Pitmatic

Thank you.

The one thing I have always found concerning in VDW, prior to getting to grips with the 1988 Mackeson article, was his apparent reliance on the inevitably subjective betting forecast. The various possible additions made at the second stage of the method as set out in the 1988 article to a large extent mitigate that. It always feels particularly satisfactory when one of the two forecasts to which we know he referred (and of course until it closed the Sporting Chronicle's), matches the returned betting.

I've just got access to a Haig for the 1980/81 NH season, so before trying another of the 1995/1996 six, I think I'll check out the 07/03/81 five to see if they work via the same methodical approach or whether perhaps the problematic "Erin numbers" played a part. I've got what I assume are the Life forecasts so can check all five against the returned betting.

The Erin numbers might prove interesting, though I'm assuming that GT, VW & SD will have the same rating for that race, unless I've really got the idea on them being a consistency for a race wrong.

I'm still of the view that this method has more to do with the type of methods trainers employ that VDW referred to with such comments as "strange things seem to happen to their charges" and "how many times have you seen a good thing come unstuck only to be found in the winner's enclosure next time when nobody wanted to back it" (I'm likely paraphrasing there).

Unfortunately I've got little time to do any form study today, either for today or the past, but hopefully I'll get some free time later. Looking forward to your findings from the Haig annual though JennyK

 

[JennyK]

First, Haig, at least back in the early 80s, has its limitations, the main one being that although it includes the first six home, and sometimes more, it mostly does not include every runner and therefore my hope of being able to check the forecasts in my notes with the returned betting is thwarted. However, this is less important with the 07/03/81 five as maybe it would be with others, because VDW specifies the forecasts he used and all the winners are short priced enough to know that they were in the first six of the returned betting, with the possible exception of Gaye Chance. More important, is that if a horse is not in the first six or however many Haig includes for a race, one would not know it had run. Balmers Coombe in the 2.15 Hereford is a case in point. There is no mention of him being in that race.

However, despite the limitations of Haig, the exercise has proved worthwhile.

The aim of the 1988 Mackeson procedure is to end up with a list of the possibles from which technically the highest ability-rated form horse is the class/form horse, but we know from the Mackeson example that VDW was happy to jettison a higher ability-rated form horse and view a lower ability-rated form horse as the most likely winner, ie the class/form horse.

The first of the five speaks for itself. A four runner race where the highest ability-rated horse is one of the three with the lowest last three races placings total, well clear on ability and, coming to the race after three consecutive wins, a form horse. Nothing in the "other factors" to concern so a clear-cut example of a class/form horse, which VDW regarded as strong enough to back.

With the other four races there were more than six runners so the various selection criteria demonstrated in the Mackeson methodology come into play.

2.15 Haydock

Those on the three lowest three race consistency totals from the first six in the forecast were Sunset Cristo and Tragus, both 3, King Or Country 7 and The Engineer 8, exactly as VDW showed.

In stage 2 we lose the lowest ability-rated of these, The Engineer, but gain Bregawn (similar to Bishops Yarn in the Mackeson) and those from the three highest class races lto: Fair View, The Vintner and The Engineer (back again).

The list in descending order of ARs at the start of the third stage becomes

Fair View 25
Sunset Cristo 24
Tragus 24
The Vintner 23
King Or Country 19
Bregawn 18
The Engineer 13

Of these, only Sunset Cristo, King Or Country and Bregawn are form horses, so Sunset Cristo, with no obvious "other factors" issues, becomes the class/formhorse, like Little Owl strong enough to bet but, for obvious reasons, not quite as certain as Little Owl.

2.30 Newbury

The issue here is Gaye Chance, with an apparent last three race placings total of 12. VDW does not explain why here, but there can be no doubt from his examples that he would have excused Gaye Chance's last run as being pushed up in class far too high.

Taking Gaye Chance's total as 3 and not 12, those within the first five of the forecast, non-handicap, are Gaye Chance (3), Easy Fella (4) and Fort Belvedere and Gowanloch (5). Because he does not yet want to show that he could count GY as 3, VDW added Fauloon and Acarine, (7).

Taking the true four as the product of stage 1, at stage 2 VDW would have extracted Gowanloch, as not having one of the three highest ability ratings, and added Ellen Mavoureen on the Bishops Yarn ground, and those from the three highest class races lto, Gaye Chance, Fauloon and Gowanloch (back again).

The list in descending order of ARs at the start of the third stage becomes

Fauloon 18
Gaye Chance 16
Ellen Mavoureen 15
Fort Belvedere 13
Gowanloch 7

Of these, all are form horses except Ellen Mavoureen so the apparent class/form horse is Fauloon, but VDW preferred Gaye Chance and in a later article referred to him as the class/form horse. So why did he pass over Fauloon? Not for the same reason as he passed over Jim Thorpe for Pegwell Bay, but for two reasons: ratings and "form not exciting".

Prior to running down the field in a class 218, Gaye Chance had won at classes 17, 24 and 22. By contrast, Fauloon's last three were a win at class 20, 2nd at class 18 when favourite, beaten by a 25/1 shot, and 4th in a class 41. How much their respective form and how much their ratings as per the table in the article determined the decision to by pass Fauloon and see Gaye Chance as the class/form horse remains to be discovered (more later).

I'll add the other two races after I've watched the 4.20 Newmarket (where I have no financial interest but want to see whether I've been too cautious in avoiding what might have been a bet due to the presence of the Haggas horse).

 

[JennyK]

2.00 Market Rasen

Straightforward, except for one point.

The three with the lowest last three placing totals from the first six in this handicap chase are as shown in the table in the article, with no need to extract from them as there are only three horses and no Bishps Yarn addition. Adding those not in the three from the three highest class races lto adds two, so the list in descending AR order at the start of the third stage is

Greenways 12
Kenlis 11
Master Brutus 10
Silberto 9
Magic Tipp 8

Greenways not a form horse so by the narrowest of margins Kenlis becomes the class/form horse in a race where, as VDW said, the relative ability of the field is like a blanket.

Kenlis a "good thing" but not a bet for VDW. A seeming inconsistency - easy to see why not a bet, with Kenlis being so close to other form horses on ability, but he is well clear of next on class/form, Master Brutus, on ratings (of which more later).

2.15 Hereford

No table in the article for this race, but I have the first five in the betting forecast of this non-handicap as Balmers Coombe, Broomy Bank, Keengaddy, Polar Express and Wild Gamble.

Those with the lowest last three placings from these are Wild Gamble (4), Broomy Bank (7) and Balmers Coombe (9), and like the Market Rasen race no extractions and the only additions, Keengaddy and Cash Bonanza on the class of last race basis. So in descending AR order at the start of the third stage we have:

Balmers Coombe 30
Broomy Bank 12
Wild Gamble 12
Keengaddy 9
Cash Bonanza 0

Of these, Broomy Bank and Cash Bonanza are not form horses, so it looks as though Balmers Coombe, well clear on ability of Wild Gamble, is the class/form horse. But in a later article VDW indicates no, Wild Gamble is.

The reason VDW passed over Balmers Coombe can only be hyothesised, but it may well be due to the fact that he was a huge surprise when winning lto, his only win before the 2.15 Hereford. In his last race Haig lists seven of the eight runners, and of that seven he and one other went off at 66/1, the next longest being 12/1. Haig commented: "proved very moderate over hurdles in 1979/80 and this was markedly improved form".

VDW seems never to have been very interested in horses whose last run seemed something of a fluke, and unless the ratings (of which we don't have his) help, I cannot currently think of any other reason why VDW by passed Balmers Coombe for Wild Gamble.

Overall

The races work out quite well via the Mackeson approach, though in two the ostensible class/form horses - Fauloon and Balmers Coombe - were by-passed. They can both be viewed as having questionable form, compared with Gaye Chance and Wild Gamble respectively, irrespective of ratings, but VDW mentions the ratings explicitly with Fauloon, the ratings differences Kenlis vis-a-vis Master Brutus are striking, and for all we know they may have played a part in Wild Gamble's preference to Balmers Coombe. Some facts and thoughts on those later.

 

[JennyK]

From the above posts, it seems possible that the ratings VDW gave in his article of 28/03/81 were material to his distinguishing Gaye Chance from the slightly higher ability-rated Fauloon and making Kenlis a "good thing" though not a bet when only a point ahead of Master Brutus on ability.

The attached sheet shows all the horses in the four races where VDW provided a rating, and the Daily Mail ratings.

VDW gave two sets and it is the second of these in the attachment, and inevitably it was interesting to see that his second set was the same as the Mail's plus 12 for all four runners in the Little Owl race. And most of his second set of ratings are not far from the Mail's +12, with some notable exceptions, including Gaye Chance (+16) and Fauloon (-6).

This suggests that one set had as its base the Mail's, with adjustments in the light of ... maybe even the problematic Erin-type numbers.

Without reference to the ratings, Little Owl and Sunset Cristo come out as solid class/form horses. The other three less so, though on form one can see why both Gaye Chance and Wild Gamble could be preferred to Fauloon and Balmers Coombe respectively and Kenlis was, by the smallest of AR margins, the class/form horse ahead of Master Brutus.

Whether the Erin-type numbers have any relevance here I have no idea, but something must explain the seemingly differential adjustment of the Mail ratings, if indeed as I currently suppose they were the basis for VDW's second set.

At the time, the Life did not give ratings for NH, so unless the first set was based on any the Sporting Chronicle, still then publishing, offered, I am at a loss to think from where VDW drew them. The differentials between Gaye Chance and Fauloon, and between Kenlis and Master Brutus are marked and in the same direction in the first set as the second, and both sets of ratings differentiate far more than the relevant ability ratings.

All that said, the results so far of running races through the 1988 Mackeson procedure currently lead me to the conclusion that solid class/form horses, which VDW stated he backed, stand out clearly, without recourse to either the Erin-type numbers or (assuming not explained by them) how he adjusted his second set of ratings. But plenty more VDW races to check.

 

[Chesham]

VDW - Van Der SpeilSee

Same conclusion back in 2016

 

[gerry]

When i looked at the results of the bigger races on saturday the one thing that stood out the most from the winners was there last three reults showed the horse was improving.There was signs of the ground coming right for them also with most having decent form on the track.
And finally they seemed to have form at the class of race to warnt respect in it.
So i would say the last three runs is a great starting point with some sort of improvment in them although this could be good run. ,ok run ,then other good run, but no disaster runs in there which could appeal to the consistant form.
There was a bases for looking at times also but that was dependant on going.
This dosnt mean they where VDW type selections but more that declining latest form is deff a no no and class is also crucial with a perfectly reasonable worry about a up and comer who might be below class but going to massivley improve these type of races would be adviced to leave. alone.

 

[JennyK]

Thinking further about the 1988 Mackeson analysis, it is clear that VDW went beyond the first numerical picture if by that we can assume he meant the four tables as in the 07/03/81 article, which focus the reader on those runners within the first 5/6 of the forecast whose last three race placings are one of the three lowest such totals within the 5/6.

In the 1988 Mackeson example, VDW extends that focus in several ways - he adds a horse not in the first 5/6 (Bishops Yarn), he extracts two others who were "qualifiers" from the first numerical picture (Comeragh King and Giolla Padraig) and he adds two more, one of whom was in the first 5/6 of the forecast (Smart Tar) and another who wasn't (Warner For Leisure). In doing so he invites us to widen our focus.

If that above is correct, it occurs to me that the interesting conclusion is not, as I initially thought, ah, here we have three new "rules" to add to those that enable us to create the first numerical picture, but that there is something more important being imparted, which I think makes much more sense of VDW.

The problem with too focused a view of the first numerical picture, and indeed with my initial view of the additions in the 1988 Mackeson piece, is that it has the danger of turning VDW's approach into a system, when he emphasised all along it was a methodical approach. For sure, he said somethings which would only be literally true if it were a system - that if we understood him properly we'd all have the same selections as him. But I am pretty sure he did not mean that literally. More like if you learn how to determine the class/form horse AND to judge which ones are rock solid (less than 20%), you'll end up backing much the same horses as me.

From the very first example, right the way through, there are ambiguities IF one thinks one is uncovering a system. Recently I've raised the question of what VDW really considered to be Decent Fellow's last three placings total - 7, 14, or something in between? Why, specifically in the 1988 Mackeson does Bishops Yarn suddenly appear in the reckoning?And there are similar ambiguities throughout the examples and often rather didactic statements about individual horses' chances which are not explained specifically and less-than-easy to construe advice, such as that about form in the 07/03/81 article.

In the light of looking at a number of races through what I now see as too strict an adherence to the new "rules" in the 1988 Mackeson example, I think I see what VDW was doing over the ten years 1978-1988 (and maybe in the later, 1995/6 examples) - encouraging his readers to take seriously his contention that the good things are identified through the balance of class, form and the "other factors" - and offering some guidance on how to proceed methodically in operationalising them WITHOUT arbitrarily narrowing the field prematurely. The additions to the first numerical picture in the 1988 Mackeson example don't provide three more "rules"; they point up the need not to be over-bound by the forecast (but equally not to ignore the underlying stats about the proportion of winners coming from the first 5/6) and to keep all possibly plausible runners on the table as it were until very near the end of the process.

Jim Thorpe is a case in point in respect of balance. First in the forecast, a perfect last three race set of placings, a form horse, and third highest in the field on ability, with the two highest (Townley Stone and Smart Tar) not currently being form horses. He has to be the class/form horse, one imagines. But no, he is by-passed at the penultimate stage because he does not pass muster on the third leg of VDW's tripod, the "other factors". Pegwell Bay, fifth equal on ability (and with Gee-A and Bishops Yarn, like Townley Stone, not currently form horses), well clear of other form horses on ability and passing muster on the "other factors" in the specific circumstances applying, becomes the class/form horse. And, helpfully, VDW underlines the significance of the "other factors" by adding, in capitals, a different circumstance where Pegwell Bay would not have been the selection.

VDW was a proven liar, in his stories to Tony Peach, his third wife etc, and seems in other ways not to have been a very nice man. But he was very clever.

 

[Thomas]

Jenny K
Have you considered and I'm only throwing it out there as I don't know,but could Jim Thorpe fall down on Vdw's percentages criteria.

The fact that JT, had won his last 3 races which equates to 33%, perhaps he didn't measure up on the other factors to raise his percentage high enough to be considered further, as I say just a thought

 

[Chesham]

VDW stated that JT fell down on Distance and Going in this class (Simples)

 

[Thomas]

VDW stated that JT fell down on Distance and Going in this class (Simples)

Good enough Chesham it's a long time since I looked at the old examples, I didn't mean to lead anyone astray.

 

[Chesham]

Good enough Chesham it's a long time since I looked at the old examples, I didn't mean to lead anyone astray.

I just think what’s the point in getting bogged down with Numerical Figures. We all have different versions of the 2nd numerical picture, the Erin numbers and two method of ratings

what ever your take on VDW surely apply that to races yet to be run.

 

[Tufnel]

I just think what’s the point in getting bogged down with Numerical Figures. We all have different versions of the 2nd numerical picture, the Erin numbers and two method of ratings

what ever your take on VDW surely apply that to races yet to be run.

The Erin figures have driven me round the bend. Every time I think I've nailed the process for getting them, there'll be one or two in a race that don't fit and it's back to the drawing board.

I know several members on here over the years have said don't bother with them, they don't matter, etc. But if they were not important then why were they ever offered up in the first place way back in the 78 PK race?

Perhaps I've just misunderstood what Bobajobber was saying? HIs clues on the open forum seemed to be about the figure being a consistency for the race (not about to be run but the LTO races, etc). Which would explain why all runners in that race had the same rating.

The SPs/Position in the market had a part to play, though again that's what I'm getting as the suggestion, and placings probably too.

The problem is, you can find a simple enough way to make many of them fit the numbers, and by that I mean the Erin and the number of examples Bob posted (presuming he is correct of course and no typos,etc) but as I say, there's always several of them that don't equate to the number suggested which is confounded further by the fact you can devise different ways to get the right figures, without actually getting the process right.

The angles I'd been looking at were points awarded for position in the betting, based on the Methodmaker and factual evidence of where most winners come from in the market. Bob made play several times of a past race showing the first 2 finishers as both being from any 1 of the first 3 in the betting, which has logic to it based on the percentages of races won by the first 3 in the market.

The next 2 quotes must surely have a lesser score and likewise any outside of the first 5/6 in the sp returns (depending on hcp or non hcp).

I'm tempted to just move on, but there must have been something worthwhile to reward those who managed to sort them out?

 

[Chesham]

HI T Tufnel

See

VDW - Van Der Speil

VDW was using a combination of Rating for the 90 Fig (Two method of Rating) he was deducting to get the 90 Fig and and adjusting accordingly

The Other figure he is Adding and adjusting accordingly.

 

[JennyK]

T Tufnel

"you can devise different ways to get the right figures, without actually getting the process right."

You are spot on, but don't seem to be drawing the appropriate conclusion.

I am open to being corrected, but to the best of my belief the 1978 Erin was the one and only time VDW explicitly used what those who think the figures weren't consistency totals with some mistakes refer to as probables or Erin numbers.

If that is correct, first, we don't know if VDW used them throughout or just in the Erin example, or maybe something in between, ie for some months or years before abandoning them. Second, given that there seem to be several methods of getting to the numbers, there is no way of knowing which, if any, is correct. For those of us who have a solution, it would be nice to know whether ours is right, but that is a gratification (or a corrective) we are unlikely to get.

Much more to the point, in his article setting out the first numerical picture, VDW used consistency figures in the sense of the sum of the last three placings (adjusted when necessary as he explained in various letters and articles, eg Mr Kildare, Uther Pendragon, Gaye Chance). And he did so in other examples including the 1988 Mackeson where he presented a (I think the) second numerical picture. Whether or not he used probables figures as well as a means of underwriting the consistency figures is an open question, but for sure they are not necessary for the identification of any of his selections (including Prominent King if one takes the "Guest" view about Decent Fellow).

A study of VDW's examples (not sure about the handicap hurdles ones which I've not looked at) up to but not including Travado etc, shows several things:

how he assessed which horses were form horses and which were not

the profiles he looked for in his selections - different for class risers (the majority) and class droppers - and what shortcomings from the ideal he was willing to accept, and why

why sometimes he by-passed the apparent class/form horse for a form horse with a lower ability rating, as with Jim Thorpe/Pegwell Bay (basically, what he meant by the "other factors").

These are all very much more important in my view than how VDW calculated the probables numbers (which we will likely never know for sure). I am also interested that in reply to one of my posts a week or so ago you said that the probables numbers - one, unproven, version of them - did not seem to be necessary to find the selections in which they were illustrated. It is a consciousness as to whether a horses is rising or dropping in class, the consistency figures, the ability ratings (including for some Flat races the speed/merit alternative to the win prize money rating) how VDW assessed form and when (and when not) to be concerned about one or more of the "other factors", which explain the examples, not the probables nunbers.

 

[Tufnel]

HI T Tufnel

See

VDW - Van Der Speil

VDW was using a combination of Rating for the 90 Fig (Two method of Rating) he was deducting to get the 90 Fig and and adjusting accordingly

The Other figure he is Adding and adjusting accordingly.

It's the Erin probables figure I'm referring to. VDW only gave us the ones in the Erin race, but Bobajobber has posted several examples with his formula (whatever that is) for races such as Pegwell Bay, Monbeg Dude (Welsh national) and It's A Steal, a selection back in 2014 that I believe you yourself brought to attention in part due to the horse's improving Erin numbers.

All I can say with any sense of certainty on this particular consistency figure is that it's a figure for previous races and stays the same for all horses who ran in the race. For example Pegwell Bay and Smart Tar both had a figure of 3 for their LTO race where PB won.

As I've mentioned before, I never saw the blog Bob had in the IS that apparently explained how he arrived at the figures, so I can't see where I'm going wrong with them if it is based purely on past betting data combined with who placed in the LTO race and possibly where they placed before or which horse's in the betting made up the places in that race before.

It's more of a bugbear that I can't solve it, but surely the rating had some purpose, otherwise why did he offer it up?