Welcome to Equiformratings

Equiformratings uses a handicapping process based on my understanding of the
handicapping concepts of the late Dick Whitford.

At the core of the methodology is the belief that past performances of a horse
can be expressed in terms of weight. The past performances are called past
merit ratings (PMR's). The PMR's are used to estimate the likely performance
of each horse in each race for which selections are made. These estimates are
called expected merit ratings (EMR's).

The weight to be carried is then subtracted from the EMR's and the answer
obtained is used as the criterion for judging the most likely winner, next most
likely winner and so on. The difference between EMR and weight is called
the rating criterion (the RC).

The PMR's are of course achieved under a variety of race conditions,
different courses, goings, different racing codes and sometimes even
different racing jurisdictions. Estimating EMR's from PMR's is tricky but
necessary. The method used is based on a huge number of tests carried
out on UK and Irish racing results from 1998 to 2004, some 66,000 races
in all. Factors taken into account are :

1. draw bias
2. weight for age (WFA)
3. weight for experience (WFE)
4. time since last run
5. recency of earlier runs compared to most recent run
6. race distance of past races compared to today's race distance
7. codes of past runs (meaning AW, turf, hurdle, chase)
8. gender
9. headgear
10. jockey
11. trainer (only used for debutantes)

The WFA tables have been determined without any reference to official
tables, using a “what works best” approach.

WFE tables were introduced to give, for example, recognition to the fact
that a physiologically mature horse that switches from the flat to the jumps
will improve in performance for a few years notwithstanding the fact that
the horse may already have reached the peak of its physical development.
They are also useful in judging the chances for example of a horse that
switches from turf to AW or the other way round.

Draw bias is estimated over the 5 calendar years preceding the calendar
year of selection. Both PMR's as well as EMR's are adjusted for draw
bias.

Jockey skills are measured by comparing the actual performance of riders'
mounts with the expected performance ignoring the jockey.

Trainer skills are measured in a way similar to the way jockey skills are
measured, after allowing for jockey.

Finally, the RC's are used to estimate win probabilities, which are converted
to “fair prices” and expressed in odds to 1 terms.These prices make up a
100% book.

Users who bet with bookmakers should add something for uncertainty and
(hopefully) profit.

Users who bet on exchanges need to adjust prices to allow for commission
at whatever rate is appropriate, add 1, and then add a margin for uncertainty/
profit.

An important point needs to be made. The prices suggested are estimates.
They are not cast in stone, and users are encouraged to exercise their own
judgment in deciding whether or not to bet, or what price to accept.

It is in no way suggested that users bet blindly on every bet that appears
to offer value. The aim rather is to highlight bets that seem to offer value
so that users can check form for themselves before making a final decision.



Home Betting Markets Betting Software History Member's Area Methodology Printout Proofing Ratings & Results Results Latest Results 2008 Results 2007 Results 2006 Results 2005 Seasonality Subscription Test results	Equiformratings uses a handicapping process based on my understanding of the handicapping concepts of the late Dick Whitford. At the core of the methodology is the belief that past performances of a horse can be expressed in terms of weight. The past performances are called past merit ratings (PMR's). The PMR's are used to estimate the likely performance of each horse in each race for which selections are made. These estimates are called expected merit ratings (EMR's). The weight to be carried is then subtracted from the EMR's and the answer obtained is used as the criterion for judging the most likely winner, next most likely winner and so on. The difference between EMR and weight is called the rating criterion (the RC). The PMR's are of course achieved under a variety of race conditions, different courses, goings, different racing codes and sometimes even different racing jurisdictions. Estimating EMR's from PMR's is tricky but necessary. The method used is based on a huge number of tests carried out on UK and Irish racing results from 1998 to 2004, some 66,000 races in all. Factors taken into account are : 1. draw bias 2. weight for age (WFA) 3. weight for experience (WFE) 4. time since last run 5. recency of earlier runs compared to most recent run 6. race distance of past races compared to today's race distance 7. codes of past runs (meaning AW, turf, hurdle, chase) 8. gender 9. headgear 10. jockey 11. trainer (only used for debutantes) The WFA tables have been determined without any reference to official tables, using a “what works best” approach. WFE tables were introduced to give, for example, recognition to the fact that a physiologically mature horse that switches from the flat to the jumps will improve in performance for a few years notwithstanding the fact that the horse may already have reached the peak of its physical development. They are also useful in judging the chances for example of a horse that switches from turf to AW or the other way round. Draw bias is estimated over the 5 calendar years preceding the calendar year of selection. Both PMR's as well as EMR's are adjusted for draw bias. Jockey skills are measured by comparing the actual performance of riders' mounts with the expected performance ignoring the jockey. Trainer skills are measured in a way similar to the way jockey skills are measured, after allowing for jockey. Finally, the RC's are used to estimate win probabilities, which are converted to “fair prices” and expressed in odds to 1 terms.These prices make up a 100% book. Users who bet with bookmakers should add something for uncertainty and (hopefully) profit. Users who bet on exchanges need to adjust prices to allow for commission at whatever rate is appropriate, add 1, and then add a margin for uncertainty/ profit. An important point needs to be made. The prices suggested are estimates. They are not cast in stone, and users are encouraged to exercise their own judgment in deciding whether or not to bet, or what price to accept. It is in no way suggested that users bet blindly on every bet that appears to offer value. The aim rather is to highlight bets that seem to offer value so that users can check form for themselves before making a final decision.