In this article, our senior contributor E. J. Minnis replies to queries sent in by PPM readers. Please send any questions to: The Editor, PPM, PO Box 551, Dee Why, NSW 2099.

Edward McKay sent an email asking the following: I've only recently started doing my own speed ratings and have a couple of questions for you.

What is the best method of arriving at the horse's "base" run? Should it be one of the following:

(a) Highest rating this preparation.

(b) Highest of the last two or three starts.

(c) Last race at today's (or closest) distance irrespective of how long ago.

What is the best way to determine the daily track variant? Currently I simply consider all races of 1600 metres or less and then cut the highest and lowest par difference averaging the remaining races per 200 metres.

As a general rule, recent form is the best form to use. But this is not always the case, such as with firstuppers, or horses moving up or down to a more suitable distance.

Some handicappers produce speed ratings that are a basic "end product" and leave it at that, while others use them more as an analysis tool, incorporating them into a broader handicapping approach.

Knowing the fastest horse in the race, while important, can be misleading if the pace of the race under preview is likely to be unsuitable, i.e. the analysis has identified the race as one in which there is likely to be very little early pace, with one or two horses likely to race on or close to the pace, with the others likely to get back midfield or worse.

If the fastest horse is a get-back type it is unlikely to be a winning chance if the front-runners run a muddling early pace and are capable of coming home in a sub 34 seconds for the last 600m sectional.

Only the very best of get-back horses would be able to stand the leaders up a start and be capable of coming home in what would amount to be a sub 33 seconds last sectional - invariably in such races one of the front-runners or on-pace runners will win.

In regard to determining a base run, I would use a weighted average of a horse's last three runs this preparation, discounting any abnormal runs, where the horse may have missed the start, got blocked for a run, etc., resulting in the horse being badly beaten, or where one of the runs was over an unsuitable distance.

The weightings should be on the basis of 50 per cent of the horse's last run, 30 per cent of the second-last run and 20 per cent of the third-last run, with at least two runs being used. If only two runs are used, then the weightings should be 65 per cent for the most recent run and 35 per cent for the other run.

As a demonstration of this technique, I will use my speed/ pace ratings for Lonhro, the favourite in this year's Doncaster Handicap, comparing his ratings with those of the eventual winner, Grand Armee.

It is worth noting that my ratings are not a pure speed rating but take into account a pace component, which in my opinion is an essential part of the pre-race analysis.

Using my ratings process, 100 is considered par, with those higher being positive and lower negative.

Prior to running in the Doncaster Handicap, Lonhro's last three ratings were 90, 99 and 132 - one very much below par, one very much above par, with the other being a neutral rating.

As the rating of 90 was achieved in his last start, the George Ryder Stakes with a slow early pace, it was discarded as being unrepresentative of the horse's ability.

Using the weighted average approach, his ratings were 99 by 65 per cent equalling 64.5 plus 132 by 35 per cent equalling 46, giving a projected final speed/pace rating of 110.5.

Grand Armee's last three ratings were 121, 118 and 114 - all ratings above par.

The weighted average approach came up with a rating of 121 by 50 per cent equalling 60.5, 118 by 30 per cent equalling 35.5 and 114 by 20 per cent equalling 23, giving a projected final speed/pace rating of 119.

It is now history that Lonhro ran disappointingly in the Doncaster Handicap, finishing fourth, while Grand Armee gave Gai Waterhouse her sixth win in the race, winning at what I considered to be the very generous odds of 12/1.

This approach will not always work as well as in the above example but, as many speed handicappers will testify, when it does, some spectacular odds can be obtained.

While there is not much doubt that the current form is the best form to use in previewing a race, sometimes this is not possible, i.e. with first and second-up runners, and races from a previous preparation have to be used.

In regard to first-up runners, their performances at both or either of their last two preparations should be used, while with second-up runners the first-up run is okay so long as it is appropriate, but it should be used in  conjunction with a suitable run from its previous preparation if possible.

Another occasion when earlier formlines may need to be used is with middle-distance and distance performers who, having had a few runs this preparation, are now moving up to a suitable or winning distance range.

In such cases, their best performance at the distance of the race under preview from their previous preparation is probably the most appropriate race to be used in the analysis, given that speed ratings usually work best on races at 1600m or less.

I would exclude any two-year-old races, those under 1200m and any with muddling pace during the early part of the race, concentrating on races between 1200m and 1600m.

However, the distance range may have to be varied, as at least four races at any given meeting should be the minimum used to determine the daily track variant.

Mark Watson from Lilydale in Victoria asks: In a particular race, how are horse number listings arrived at?

In the same race, how are weights arrived at by the handicappers? Is it supposed to give all horses the same chance of winning? If so, would a certain horse's weight change depending on the class of race?

TAB numbers are determined by the weights, with the topweight being allotted TAB No. 1.

Weights are determined by the handicappers according to how the handicapper rates the horse, as well as the restrictions placed on the horses by the conditions of the race.

For example, in Quality Handicaps there is a restriction as to the maximum weight the topweight has to carry,
usually around the 60kg mark.

While in ordinary handicap races, be they open or restricted, there will be a minimum weight a topweight has to carry, usually around the 55kg mark, which sometimes can lead to the weights being raised (the latter occurs when the original topweight/s are not declared as final nominations in the race).

The more races a horse has won, or the more races a horse has performed well in, means that a horse will be weighted higher than a lesser-performed horse.

The allotted weights will also depend on what class of race a horse has won and what type of race is being handicapped: the basic principle being the lower the class of race, the higher the weight.

The handicapper's job is to handicap the horses in such a way that all have an equal chance of winning. At least, that's the theory, but in practice that doesn't occur, which is due to other factors that the handicapper does not know and cannot take into account, such as the jockeys, the state of track and luck in running.

The exceptions to the above principles are races where the handicapper has no influence, such as set weights and weight-for-age races. In set weights races, the allocated weights are normally done according to the sex of the horse, with the fillies and mares receiving lower weights than colts, geldings and entires.

In weight-for-age races, weights are allocated according to age and not ability, again with the fillies and mares receiving the lower weights.

Another email, this time from David Pavis from Queensland, which is more a statement than a question: After nearly thirty-five years of punting, I have not seen one person capable of winning long term betting one horse per race. To win without backing multiple runners per race is an absolute impossibility as I think thirty-five years of experience is proof of that.

Most punters will not bet multiple runners per race, which is the main reason why so many of them are losing.

Men I was first told that it was the only way to win, I thought the bloke who told me was crazy but he was winning just about every week and I was having more losing weeks than winning ones.

Luckily he showed me how he did it about twenty-five years ago and I changed my betting habits and have never looked back since.

What I want to know from you is, do you think this is the way to go and, if so, what advice would you give a young punter starting out in the "punting business "?

I wouldn't state categorically that your approach, a Dutch Book method, is the "only" way to go, although it has a lot going for it.

The biggest drawback with betting multiple horses per race is the time and effort required to maximise returns and make it as successful as alternative betting techniques.

By this I mean that as in nearly all betting approaches, a punter should only ever have a wager when there is value on offer and with Dutch Book betting, where two or more horses are bet, then obtaining the best return on all of the selections can be quite a task.

If you are a TAB punter, you have no control over the final dividend a selection may pay, while applying this approach at the racetrack a firm contract can be obtained by betting with the bookmakers and this is the preferred approach on most occasions.

For young punters starting out Dutch Book punting, I would advise them to concentrate on betting two horses only per race, ensuring that they never bet odds-on.

In effect this means that if they outlaid $100 per race the minimum (potential) return should be at least $200, considering only horses at $10 or less in the betting market, of which one at least should be either the favourite or second-favourite, in races of between eight and eleven runners.

By E.J. Minnis

PRACTICAL PUNTING - JUNE 2003