Avoiding Performance Illusions
When it comes to racing, whether it be horse or any other, the universal question that is asked everyday by millions is: ‘How fast can they go?’
Every day punters turn to tools like race times and race vision in the attempt to answer that question by looking at previous performances. On face value one can ask, how you can go wrong armed with race times, sectionals and/or with race vision. A great argument and the statement is so true, they are great tools, until they lead you astray.
There exists an issue that is as relevant to race times as it is to vision that leads many punters astray with their decisions. A deception that is played out every day, during the process of form study that many, even at times long term and successful punters fall prey to.
All elite level punters rely on a combination of tools including race vision. Their study affords them the ability to pick up things that most people would miss and this can provide an edge. The issue is that in the use of race replays many basic decisions about the speed of the race relies on the race vision itself. It is hard not to; we are wired to see and calculate the speed of things in order to enjoy or avoid them.
While we do these calculations quite well to track a horse with our eye for the pleasure of watching it or to avoid being trampled by it, our accuracy is nowhere near that which is needed to determine the speed of the race. I am not disputing that with thousands of hours of watching race replays the skill of determining how fast a horse is running is very well honed. Unfortunately, it still falls short of what is needed.
There are two main issues facing a punter. One is the ability to pick up on the actual speed of a horse. The second is the contrast of that performance. They are both intertwined when working out the speed of a race, however we will attempt to examine them separately to better understand the limitations of each.
Speed, when it relates to horses, is nearly impossible to pick up by the naked eye. Sure, the difference between a canter and a gallop is obvious but how possible is it to pick up a difference of 0.5 second over 600m? What we are talking about is a difference of less than 0.001 of a second over each meter. Impossible! Even 0.002 per meter which equates to a full second difference is just as impossible.
The human mind is not able to deconstruct such tiny values. To put it in perspective the average reaction time to light stimuli is measured at 0.19 seconds, so when measuring time even the guy with the stop watch might lose 0.2sec before he reacts.
Contrast, when it relates to horses it is very hard to obtain. Sure, put a horse against a billboard you will see it move but how possible is it to pick up a difference of 0.5 seconds against a par time? For you to truly determine if a performance is good or bad you need to compare it to other such performances and adjust those performances for track speed on any given day. Track speed is the relative speed at which the track races at on any given day versus what would be expected on an even Good3 surface not affected by wind, lane bias or moisture.
For argument sakes, let’s say that you could pick up the 0.5 second, what does that time represent when it relates to a distance? We all know that the same time value represents a great difference based on the distance of the race. In general terms, the way that hoses run a 1000m race is faster, even to the naked eye than the speed with which they run a 2000m race.
Faced with this, it can be said, that by observing race vision you can’t but help to see a ‘performance illusion’ which is that nearly every 1000m race is fast and every 2000m race is slow. The issue is how do you tell a fast 1000m race from a slow 1000m race when they are hurtling down the straight and the difference between a fast and a slow race is 0.002 seconds per meter or 1 second over 600m? It does not stop there, take track condition into account, track surface type, track shape, camber and general conditions of the day to mention just a few things that come into play.
If you truly wanted to make decisions based on what you saw, you would need to be able to mentally create a ‘vision par’ consisting of all races based on distance and track and track condition to be able to properly compare a race to others and answer the question of race speed.
Those wise enough to accept and understand the illusions and limitations created by the human deficiency when watching race replays will tell you that they use time for the purposes of working out the difference in speed of the race.
I agree, time taken to traverse a given distance is the only sensible method we can use to measure the performance, if the times are accurate and are adjusted appropriately for track speed. Official horse racing times in Australia are not accurate enough for the purposes they are employed for. In fact, our data suggests that around 98% of all race times and sectionals are inaccurate by varying degrees.
However, this is not an insurmountable obstacle and investing large amounts of money annually will result in a punter obtaining the correctly measured race times, pre-any track speed adjustments.
Armed with accurate times and track speeds, punters still face a set of problems not unlike the ones that cause issues with using vision. There is an upside; the first issue of speed is well and truly solved. Accurate race times allow us to pick up differences between races with up to four one hundredths of a second precision.
The problem that remains is one of contrast and it is still just as relevant. Even race time is susceptible to creating ‘performance illusions’. To be able to compare a single race time against other race times, one needs a “PAR” time and an accurate track speed adjustment figure. In other words, the average time taken to run that distance. We all know that we need Track and Distance (TD) specific pars.
Without TD pars’ it would be problematic to compare raw times between different tracks. In the example in the table above, one could, without pars, say that the average horse at the Gold Coast is 1.33 seconds faster than at the Sunshine Coast. Such a conclusion would be a performance illusion. We should all know that it’s just the way the tracks “race” that makes up the difference.
However, every day many punters fall for just this when using times and given that there are more than 4,200 individual Track & Distance pars (Australia) even if you had them, using and maintaining them would be problematic.
If only the problem stopped there! While Pars allow us to compare time between different tracks, once we get passed the track specific pars we need to consider that an ‘on par’ performance at a country track is not the same as an ‘on par’ performance in the city.
Extrapolate that to different states and one quickly realises that the further we move from a single track running a typical Good 3 the less user friendly raw times become.
Things get a bit more complicated when we take track conditions into account. On face value a time on a Heavy 10 is slow compared to a Good 3. Which would again be a performance illusion, one that can be overcome by using Track and Distance and Track condition (TDC) pars which can cast a totally different light on such races.
The ‘fast’ Good Track time may be nothing other than average and the ‘slow’ Heavy Track time may be the fastest ever run. Given though that the official track condition is often incorrect, relying on pars created by them are subject to errors. To further complicate the issue, each Track & Distance par would have to be multiplied by the 10 official track conditions, taking the number of individual pars needed to over 42,000. Things are starting to get a bit ridiculous!
Even if one was to have the above-mentioned pars and could use them, there is a further issue. These precisely measured performances would still create performance illusions. Such situations arise when the prevailing conditions on the day create a situation where many extraordinary performances are recorded and never replicated. Rock hard tracks or wind assistance could give a Black Caviar like performance to many horses.
Worse still is that such a performance would then be the cause of many thousands of dollars’ worth of bad bets even if you did have TDC pars if race times are used. To overcome this issue one would need Class pars to be able to adjust back the Race times by a portion that represents the assistance of the track or wind. When we get to this stage, the par combinations created by Track, Distance, Track condition and Class run into hundreds of thousands.
WHAT IS NEEDED is a method that will create an easily assimilated value that represents the horse’s performance and which takes all the above issues into consideration and is not only able to provide contrast across tracks, track conditions and classes but also across different states and jurisdictions.
What is needed is an accurate speed rating!
There is no doubt that watching race replays is of great benefit and race times, especially ones that are corrected are extremely important. It is even possible to say that a punter whose betting is confined to a single or a few tracks, can become a proficient expert in the idiosyncrasies of each track and the way they relate to each other.
With the way, the market is going, such specialisation is now however having impact on the ability to turnover enough money to remain profitable. With more tracks or bigger form convergence, such as one experiences in city tracks where horses can come from almost anywhere, it quickly becomes an issue for punters to be able to do form with the same level of consistency.
This is where a single number built on the foundation of accurate times that will allow its user to look at the speed of the race in comparison to others, evaluate a horse’s performance against any other horse across tracks, jurisdictions and states and normalise such performances by taking into account the prevailing conditions on the day becomes invaluable.
The simplicity and efficiency of a number when comparing horses in a race where the many diverging form lines which would otherwise have you watching dozens of races and dealing with the short comings mentioned above cannot be understated.
Thumbing through pars, working out the track conditions and adjusting performances is replaced by a single column of numbers where each horse’s past performance becomes clear and comparable at a glance in a way that time would never be able to provide.
Ratings are an integral part of a process by which you could arrive at your horse form conclusions. They do not replace race vision and like times they need to be adjusted for weight carried, in running incidents, bias and any abnormal race occurrences. What they do provide is an efficient method of tangible performance comparisons that will streamline the rest of your form study process.
SPEED RATINGS therefore, (those of which have been formulated by the above suggested processes) put in the most simplistic of terms, allows the analyst to confidently compare APPLES with APPLES on an ongoing basis without fear of the complications and deficiencies described earlier in this piece.
For access to Australia’s best and most widely acclaimed R2W Axis database used by many industry professionals go to ratings2win.com.au and/or contact me personally on 07 3103 2262 or by email email@example.com
Cheers and Good Punting
©ratings2win Pty Ltd 2016. Printed under licence to Betfair Pty Ltd. This document is for personal use only and must not be reproduced. This information is supplied “as is” and is to be used for general guidance only. Use of this information is entirely at your own risk. Neither Betfair nor ratings2win make any representations or warranties of any kind, including but not limited to, the accuracy, quality or completeness of the information contained herein. All intellectual property is owned by ratings2win Pty Ltd. Please gamble responsibly. For more information, visit Betfair’s Responsibility Gambling Policy.