The senders of further emails have asked if I could throw any light on "the key" (as claimed to have been found by G Hall in his letter published in the Sporting Chronicle Handicap Book of 11 January 1979 (item 14 of "The Golden Years of Van der Wheil") or "the second numerical picture", something VDW suggested we needed to progress to from what he referred to as the "first numerical picture" in his Pegwell Bay evaluation (included in "Betting the VDW Way").
Frankly no, and not because they contain important truths about the method I am unwilling to discuss. Rather, I see no way we can be sure what G Hall and VDW respectively meant by those phrases.
Taking "the key" first, G Hall claims to have found something crucial from VDW's first few examples - Prominent King (item 8 of "The Golden Years of Van Der Wheil") and another batch included in a letter reprinted as item 13 of the same booklet. That crucial element apparently enabled Hall to find "a constant stream of winners" including four named winners on Cambridgeshire day 1978. VDW responded to Hall's letter, confirming the four named winners were "good things" (but not referring to other easily identified winners Hall referred to without actually naming) and congratulating him "on "spotting the key" as he describes it".
It is easy enough to lay out the details of the VDW examples on which Hall was able to draw, and those of the seven horses to which Hall referred (the four named ones and the winners of the 1978 Spring double and 1978 Cesarewitch) and see some commonalities. But which of those Hall meant when he referred to "the key" seems to me to be literally impossible to know. For what is certain is that, at the time Hall wrote his letter, VDW had not mentioned important parts of what I think of as his basic method (for example the ability rating), and thus Hall found his winners on at best a part appreciation of that method. And we can go further, and confidently state on the basis of what VDW wrote subsequently that although the four named horses (but not necessarily the three unnamed ones) were "good things" in the sense of being what we were later able to know as probables with form, it is open to question (on the discussion in the March 1981 article) whether any of the four would have been bets for VDW.
Take for example Baronet, the winner of the 1978 Cambridgeshire named by Hall. There is no question that he was one of only two probables with form in the race, and from my understanding of VDW's way of assessing form he had the better form of the two. But he had a much lower ability rating than the other probable with form and was ranked only 8th in the field. The evidence is that VDW confined his bets to class/form horses in the top four of the ability ranking, occasionally going down as low as fifth. On the old Gummy forum Lee once posted to the effect that, although if he had had a bet in the 1978 Cambridgeshire he would certainly have included Baronet, Baronet would not have been a clear single bet, and I agree. IF VDW bet in the race he would have backed the two probables with form as a book, not just Baronet.
The Hall examples are well worthy of study, but in my view not in the hope of finding the answer to understanding the whole basic VDW method, and certainly with no possibility of being certain which of their common features Hall thought was "the key".
Turning to the "second numerical picture", I think there is no way we can be sure what VDW meant. It is reasonable to assume, from the reference to the "first numerical picture" in the Pegwell Bay article, that by that VDW meant the basic platform as shown in the four tables in the March 1981 article, ie the names of the runners, their previous placings, consistency totals, ability ratings, and their cross-check ratings, though even that is arguable - he could have meant by the first the more limited numerical picture given in his Prominent King evaluation.
But long before the Pegwell Bay evaluation VDW had shown us other numerical pictures, rather different to the one I am assuming he meant was the first. For example, there are the numerical pictures given in The Old Feller example (item 24 of "The Golden Years of Van Der Wheil"), the discussions of the 1981 Spring double races (item 42) and the 1981 2000 Guineas (item 47), and the Roushayd chapter of "Systematic Betting". It is also the case that the very first table in the Pegwell Bay evaluation is a substantially extended version of the tables in the March 1981 article, and is thus yet another numerical picture.
So any idea that there is "the second numerical picture" that provides the essence of the basic method is in my view as misconceived as the view that Hall's "key" does.
Personally I believe the answer to discovering the basic method and achieving the very high strike rate VDW claimed was possible lies in pains-takingly working out precisely the operational definitions of what VDW referred to in the March 1981 article as "the elements" of the method. Then, although the order of operations is determined by logic, the exact form of the numerical analyses (or pictures) one uses to get through that order seems to me to be of no great importance.
So what are the elements, the numerical aspects of which need to be included in a VDW analysis:
1. Consistency - and to identify the consistent horses the first level raw data one needs is positions in betting forecast (down to 6th and equals in the case of handicaps, 5th and equals for non handicaps) and the totals of the last three completed placings of each runner. From that, the "automatic" consistent horses - those with the three lowest last three run totals from the first five (non handicaps) or six (handicaps) in the forecast - can instantly be identified. Then, with some additional raw data, one can identify any "discretionary" consistent horses - the contemporary equivalents of, for example, VDW selections Prominent King, Love from Verona, Son of Love and Righthand Man;
2. Probables. Sometimes there will be more than three consistent horses (indeed sometimes more than three "automatic" ones because two or more horses within the first five/six of the forecast have the same low consistency total). Here VDW applied a numerical device to try to reduce the number to three, which he termed "probables". (In the Prominent King example the four consistent horses - Beacon Light, Decent Fellow and Mr Kildare ("automatics") and PK ("discretionary") were reduced to three by the device, which generates for the four the numbers shown in the Prominent King table);
3. Ability. Straightforward - win prize money divided by number of wins - but also with younger, unexposed Flat horses, the subsidiary, time-based method shown in item 47 of "The Golden Years of Van Der Wheil";
4. Form. Which of the probables are form horses and which are not - for which yet more numerical data is needed, essentially as outlined in the second full paragraph under the Little Owl table in the March 1981 article;
5. The "balance between class, form and the other factors" (item 36 of "The Golden Years of Van Der Wheil") - for which yet more numerical data is needed in respect of the probables with form in order to identify the class/form horse;
6. "Winner in the race" characteristics - the numerical profile of the class/form horse is checked against the numerical characteristics of VDW's notion of the "winner in the race" - which requires yet more numerical data.
When one knows - as far as it is possible to know - how, operationally speaking, VDW handled each of these elements then how one brings together the various data seems to me to be a matter of convenience rather than part of the method itself. Personally I use an Excel workbook for my analyses:
in sheet 1 I include all the data necessary to identify the probables in ability rating rank order (ie stages 1-3 above), so that is my first "numerical picture";
in sheet 2 I include all the data necessary to identify, first, the probables with form and then the class/form horse (ie stages 4-5 above), so that is my second "numerical picture";
in sheet 3 I include all the data necessary to determine whether the class/form horse has all the "winner in the race" characteristics (ie stage 6 above), so that is my third "numerical picture",
and finally in sheet 4 I have the OR-based data I use to cross-check the outcomes of stages 1-6, my alternative to the two methods of rating VDW referred to in the March 1981 article, as I haven't been able to work out how he compiled his ratings, so that is my fourth "numerical picture".
But depending how one wants to organise things, one could easily either split up operations I cover with one sheet into two or more, or combine operations I cover in two or more into one. That will determine how many "numerical pictures" one then has, but it seems to me entirely a matter of personal choice rather than material to working methodically through the method.
In sum, my advice to those who have raised the questions with me is don't waste your time trying to answer essentially unanswerable questions such as what did Hall mean by "the key" or VDW by "the second numerical picture" and instead focus on answerable questions, namely how, precisely, did VDW identify (a) the consistent horses, (b) the probables, (c) the probables with form, (d) the class/form horse and (e) the class/form horse with "winner in the race" characteristics.
