Figuring It Out
When someone publishes a list of figures you may not be quite sure where they came from, who did the measuring or how they were put together. Here's a general example – house prices. The real estate people are often telling us the average price of houses in particular cities. But what they are calculating is the average price of houses sold in the current period. They do not know the prices of houses that were not sold. Therefore the mix may or may not be typical of all houses in that area yet the figures may be taken and used in all sorts of ways to tell us how the economy is doing. That may be right or it may be misleading- you don't know.
Back to racing, there are three recent examples worth noting.
First, that very cluey website justracing.com.au picked up a series of sectional times published by TVN for Melbourne horse races and demonstrated how inaccurate they were. So, too, with contradictory penetrometers readings and mixed advice about stewards' track classifications. (Details available on the JUSTRACING website).
Second, a web blogger on globalgreyhounds.com recently published a list of Australian and NZ tracks in order of the frequency of Falls/Distanced/Tailed off as a proportion of all runners. Now this may well use correct arithmetic but does it tell us anything?
The blogger does not quote the source or the method used to extract the figures so it's hard to check. However, most of the non-TAB tracks were at the bad end of the list (starting with Port Pirie) and most of the TAB tracks at the good end, especially Hobart and those in NZ. That leads to a suspicion that (a) non-TAB data may not have been as well recorded and/or (b) that “Distanced” dogs would be more common at small country tracks where first starters, non-chasers and low quality performers are common. The results would also be confused by the proportion of small fields in each sample – again much more likely at country tracks.
In any event, both Distanced and Tailed Off would include injured dogs but whether their problems could be sourced to interference or natural causes is unknown. Either way, that data cannot be added blindly to Falls as that would be adding up apples and oranges and getting a fruit salad.
Third, the web also contains a very fine site called greyhound-data.com. It has a terrific presentation of worldwide breeding information, well designed and easy to access and read.
It also contains the racing experience of each of these dogs – starts, wins, Group races, etc. However, they are not worth a cracker because they are hopelessly incomplete. You would be lucky to find half the necessary runs for a given dog. That is, it is useless information. In this day and age good data should be readily available for such an important purpose. If it isn't, nothing should be printed.
Perhaps Greyhounds Australasia should be charged with combining its existing stud book information with a similar “book” for race form. By doing that breeders could readily check which sires and dams are producing what results from their progeny on the racetrack. (Such a facility was available in America for their dogs but I can't find the reference at the moment).
Similarly, it would not be hard for state authorities or GA to keep tab of race falls, the probable reasons for them and potential changes to fix the problem. Watching a few races would do no harm either.
There would not be a major sport in the world which does not rely on and publish performance data. It is a core ingredient of every TV presentation. Indeed, people like Bruce McAvaney will tell you what their grandmother had for breakfast on her wedding day.
This practice has galloped along in recent times as footballers' time on the field or bowlers' heartbeats or the number of forehand winners are disclosed to the nation. The public like all this stuff and participants make use of it on a daily basis to check how their charges are doing.
But that data has to be accurate, relevant and usable. Near enough is not good enough. And half the job is not acceptable.