By Dominic Gardiner-Hill, BHA Head of Handicapping

Those with an interest in the subject will remember that back in 2016 my predecessor as BHA Head of Handicapping, Phil Smith, undertook a review of the Flat Weight-For-Age (WFA) Scale – the recommendations from which were implemented in 2017.

Move forward to August 2023 and the European Pattern Committee asked me to lead on a piece of work that:

a) Reviewed, where possible, the effectiveness of the amended Scale adopted in 2017 and recommend any further changes based on updated data analysis.
b) Unify the WFA Scales used across Europe – producing both imperial and metric versions of the agreed Scale.

In the following blog I will highlight samples of the analysis which proved key to the project, explain the process undertaken, the conclusions drawn and the final recommendations.

Analysis

This section presents the pertinent analysis in the following order:

– BHA explanation of the analytical methods used
– Data drawn from all age handicaps run in Great Britain, Ireland and Germany
– Data drawn from all age Black Type races run in Great Britain, Ireland, France, Germany and Italy.

Analytical Methods

WinIV

Win Impact Value: compares the number of winners achieved with the number that would be expected by random chance. This metric assumes that all horses have an equal chance in a race; for example, in a five-runner race, each horse would be expected to win 20% of the time.

Par is 1.00, where the number of winners equals the number of winners expected by random chance. As such, if a WinIV of 1.50 is achieved then 50% more races are being won in comparison to random chance. In contrast a WinIV of 0.50% suggests that 50% less races are being won in comparison.

PlcIV

Place Impact Value: similar to WinIV in that it compares the number of placed horses against those expected by random chance; for example, in an eight-runner race each horse would be expected to be placed 3/8ths of the time.

Once again par is 1.00 – a PlvIV of 1.50 suggests 50% more places are being achieved in comparison to random chance and so on.

PRB

Percentage of Runners Beaten: The average percentage of runners beaten across the sample. A horse finishing 2nd in an 11-runner race has beaten 9 of their 10 rivals (90%), then the average of all runners across the sample is taken. The ideal number should be as close as possible to 50%.

Non-completions are considered to have finished joint last, so 0.00%, whilst any walkovers or voided races are removed from the analysis.

Heat Maps

In the “heat map” tables below those areas coloured white are those achieving, or close to achieving, the ideal results.

Those shaded in blue are those achieving results below the ideal – the darker the shade the further away it is.

Those shaded in red are those achieving above the idea – again the darker the shade, the greater the over achievement against random chance.

Handicap Data

Great Britain

Ireland

Germany

European Black Type and Listed Data

It is clear to see that the 2017 changes have had minimal impact on the overperformance of 3yos in British handicaps, particularly those at 13f+, whilst they also continue to regularly overperform in 10f-12f all age handicaps.

Whilst they also overperform at various points at 10f+ in Ireland, the issue is not nearly as marked as it is Britain whilst in Germany there is clear 3yo overperformance in 12f+ handicaps run between the middle of August through to the middle of October.

Interestingly, however, the data from Pattern and Listed races run in Great Britain, Ireland, France, Germany and Italy during the same period tells a slightly different story.

The metrics at all distance categories up to 13.49f are about as close to ideal as you are going to get but once again highlight 3yo overperformance in races run in the extended distance category (13.5f+). I therefore decided the only area which needed “tweaking” and bringing more into line this time around was the extended distance races.

Given the above, I then had to make the decision as to whether those amendments should be based on the data supplied by domestic handicaps or that by Pattern and Listed races in the five major racing jurisdictions across Europe.

I decided on the latter for the following reasons:

– The review was undertaken at the request of the European Pattern Committee whose ultimate objective was to not only review the performance of the current Scale but to unify the Scales across Europe.

– Adjustments based on handicap form in Britain in particular could potentially be of a size that had an impact on the balance of handicaps in other countries – in essence, what would work in Britain may well not work elsewhere.

– Adjustments based on domestic handicap form could potentially swing the advantage in favour of the older horses in Pattern and Listed races rather than bringing about greater parity.

– To create greater parity in handicaps, domestic handicappers don’t have to totally rely on the WFA Scale – their own methodology can be used to help balance the situation.

– As stated above, the Pattern and Listed race metrics for all distance categories bar the extended distance contests are largely satisfactory – any changes made on domestic handicap form could potentially change the dynamic in those areas.

All the major racing jurisdictions of Europe are represented on the European Pattern Committee and through 2024 I presented several options in terms of amendments to the current Scale – as expected, minor compromises had to be made through the process as those countries represented had their say on both the size of the changes and the timing of them. A final version was approved at the EPC meeting in August 2024 for implementation in 2025.

During the whole process I called upon the expertise of the senior Flat handicappers of Ireland, France and Germany whilst, on top of approval from the EPC, each jurisdiction will have presented the revised Scale to their various domestic racing and stakeholder committees for rubber stamping.

The table below is the agreed, updated version of the imperial WFA Scale with colour coded representations of the changes made. A kilogramme conversion of it was agreed for those metric jurisdictions across Europe.