In the past I have made posts here stating how accurate the picks usually are on Timeform.com. But, just like any website that offers horse race predictions, they miss the mark sometimes. One of the things that I like to look at is the recent form data provided by Timeform. They give data that you can’t easily find on Equibase. One is beaten lengths and the other is the odds for each horse in their past 5 races. If the horse was a favorite in one of their last races, it will be denoted with an “f” after the odds. They also write a short comment on each horse. Some of the comments are hard to interpret because they use a lot of UK slang. Thats because they are based in Yorkshire, UK.
I decided to write an Excel macro that compares the race conditions of the previous 5 races to the current race, calculates average odds, beaten lengths, purse, and average finish position. Some horses have less than 5 races to compare, so I also made a macro that looks at 3 past races. The data is very easy to copy and paste from the website into an Excel form. I have to add surface, purse, and speed data from other sources. The condition compare is interesting because it generates a number from 0 to 3 based on how many matches are found (surface, purse, and length).
I’ve been running tests with data from live races. The problem I have is that Timeform.com posts race data on the day of the race. It takes less time to paste the data into the form, but I still have to gather a few missing pieces from elsewhere. With the macro I use for Equibase, I can get the data 3 or 4 days in advance of a race. So far, there doesn’t seem to be an easy way to find the winner. The factors that I find most useful are average finish and beaten lengths. But one bad race can skew the numbers. I still have to be on the lookout for horses running at a lower purse that had an increase in speed in their last race when surface and race length are now the same as races prior to their last race.