Getting Extra From the ‘Extras’ in Proform

May 26th, 2020

Given the lack of racing I can give extra time to research of different betting and racing strategies I found myself exporting some more data to run some tests on the influence of a factor on an outcome.

Personally, the lockdown from a betting perspective has been a chance to deepen my understanding of statistics and practice using tests for significance of factors and correlations. Using regressions to compare for example how much influence factor(s) may have on say the percentage of rivals beaten (PRB) and also to learn more Excel skills and formulas and problem solving, takes some effort but these are useful skills and more ways to examine your hypothesis can only be helpful to your confidence. The appreciation of work Proform have done definitely goes up when you are going through the raw data yourself!

The export tool is rather fast for the data volumes and allows you to select what you might want to focus in on, you can pick date ranges and what columns you want to include or omit allowing you to tailor your data sets for faster processing and to broaden or focus data for analysis.

(Click on the image enlarge and open into a new window)

The exporter allows you to tailor things more to your investigations and apply tools available on other software, use visuals such as matrices, graphs and scatterplots. This deeper work has led to growth in my awareness of just how much data is in the Proform Database, and if you are like me you become aware that you have been going to the ocean with a spoon, such was the fraction of functionality I was using.

Given I am still improving my knowledge on the area I would not be comfortable drawing conclusions on my results until I verify them and combine them in a way that are more coherent which is a large undertaking but would love to share at a future date.

Taking the time to go through the glossary and guides on Proform in more detail has given me some ideas of what might be of more interest to investigate further for systems, and to understand better how ratings are formed and what they represent, when they might be more important or less significant to analysis.  Frankly with hundreds of factors measured and the countless combinations in which they may be used the seemingly tedious task of combing a glossary will hardly set the pulse racing, but I suppose it wouldn’t have been included if it wasn’t important and even as being a very well read punter I learned 10 new things from this exercise alone. While it might not be exciting it will lead to better analysis value from Proform use and increased bottom line and that is the fun part!

The dynamic ratings which were added, well a while ago now, I failed to appreciate all they encompass, these numbers are an excellent useful guide and timesaver for form analysing by narrowing the field with no additional work, since PRB and Ratings capture the level of form already it avoids rework and will certainly have changed my first approach to a race henceforth!

Combining Dynamic Ratings which go on the filters to examine prior races with similar distance/ground/class/runners/etc and the horses wins/places and PRB and which horse ranks highest on these 2 factors is, when the racing returns, something I will be keeping a keen eye on.

A few ideas as to interesting analysis which I intend to do myself over time:

  • Look at class pars for speed ratings and see by yearly average, last time out and career both static and dynamic and what is the best way of finding profitable angles and winners.
  • See the potential of exotic bets such as Forecast and Tricast as the ratings can be very accurate in handicap races with established horses.
  • Make useful and impactful visuals to help you isolate and get through data faster

But do beware of the very large data sets when importing to your machine for analysis, for example adding formulas in Excel may make it very slow, Proform has millions of rows and hundreds of columns of data, so look up some tips on how to prevent freezing and perhaps run more complex longer calculations overnight and they will be hopefully ready for analysing the next day.

Also my tip from my last month of playing around with this stuff is to ‘stand on the shoulders of giants’, the Proform team and other names in the world of betting have done a great deal of work in the past, take time think about it and beware why methods were and were not used. Many of these producers of useful materials are themselves Proform users!

 While the betting space is not very well documented like other areas that get more academic attention if you are willing to look you will find some interesting stuff. It may seem like a lot of work for a casual punter but I have made the assumption that Proform users are in the main serious and methodical about their betting.

There is a lot more to learn for myself and the silver lining of having no racing is that while I spend now far less time on betting related activity, the time I have has been extremely interesting and beneficial. And I look forward to sharing some fruits of this with this community later in 2020.