Because I was not younger, I found handicap races appealing. I also have a lack of talent, so I don’t think I would ever be able to run well in a handicap race. The popularity and anger that people feel when they are treated unfairly have been the highlights of my experience with handicap racing series.
I was horrified to see a few five-mile races, where one winner beat his handicap time four times and the other won the second race two weeks later beating their target by five minutes. Many runners who had PBs were unimpressed by the handicap target. The worst was a race series with 12 (handicapped), races that saw the winner of 10 consecutively. The series was not modified in any way, despite the fact that certain runners were easily beaten while others were struggling to get close.
My club secretary (Beverley AC), asked me to explore ways that we could increase the accuracy of our handicap target calculation. It is amazing to see that you can Google any topic and find it. “running handicap”The search returned virtually nothing. Software not found. No discussion forum. Methods not available. It was easy to find information about horse racing, as well as how to maintain and record handicaps. It was a information desert for the operation.
I felt that every viable way to reach targets must meet certain criteria…
1. Runners must be able understandThe way they reached their objective
2. All should be able to use the same method
3. Verifiability should be a requirement for the targets
4. Targets should be representative of the current ability level.
If these criteria were met, we would have a good chance at pleasing the majority of people almost all the time.
Any number of these I had heard of. “methods”This is used to arrive at target times. It involves just a bunch of guys (the handicap committee) working together in an attempt (guess?) People’s times for finishing and those of others are based on their PBs at an arbitrary distance, usually 10K. Problem with the PB method is that you can no longer consider an individual’s time to be relevant. So what can you do to replace it?
The Riegel Formula was also used in handicapping to determine times for races at different distances. Peter Riegel developed this formula from his research on elite and semi-elite athlete performances. It takes the form t2 = t1 * (d2 / d1)^1.06 and, in plain English, says that if the distance run is doubled then the speed declines by 6%. The many running calculators on the Internet use this formula. The Cameron formula, a more complicated formula that uses a different reasoning process than the Riegel formula, tends to produce similar results. Only when the base race is shorter, for example, a 10K marathon, does the prediction differ from the Riegel formula, do the results start to diverge. The Cameron formula is more accurate in these situations.
One problem when using PBs for future target calculations is the inability to ensure that they have been performed under a wide range of conditions. You can have some hot days and others cold. Some are windy and others not. Others point out the points where wind and elevation changes would have a greater impact (e.g. Boston Marathon, which sees a drop of nearly 900ft between its start and finish).
A simple idea struck me: what if runners were allowed to take into consideration their previous 3 races? Then, they would be able to use the Riegel formula, or a variant, to make each race equal in distance at 10K. If there were a way of adjusting for elevation changes, we could produce a runner who is a little shorter. “flat 10K”We’d all be running on roughly the same pace by then. Assuming that the time to reach the goal in the next race is equal to the time in the previous 3 races, you would adjust this time first for distance and then for elevation.
This hypothesis was tested using data samples from previous seasons. Initial calculations were made with an Excel spreadsheet and proved to be very encouraging. Based on Dr Tim Noakes’ work, the elevation change calculations were made based on his research. It is important to make meaningful comparisons of time when runners’ base times do not come from the same races.
Many things came to light when I tried to apply the system in real life.
1. Many runners found the system to be very useful.
2. While a spreadsheet may be able to function properly, mistakes can easily be made that affect the accuracy of results. They are difficult to spot or prevent. This feature is common in many spreadsheets.
3. Once you have more than one runner, even with the spreadsheet, it was quite difficult to keep track of everything. Beverley AC was home to 160 people, at least half which were involved in the 10 race handicap series.
4. If you are looking for an objective assessment of the running ability of a runner, then it is not enough to consider races in handicap series. CurrentAbility to keep track of all races that are run by an individual
5. Some runners or circumstances required adjustments to ensure fairness to all. It was difficult to find a method that didn’t seem arbitrary and wasn’t open for criticism from anyone who might challenge your work.
There were special circumstances that required an additional look.
A runner who hasn’t raced for six months.
Runners suffering from a temporary injury.
Runners who go beyond the normal
No race history for new runners
A majority of clubs who run a series handicap races have some sort of point system that calculates the series results. A sliding scale was used in the past that gave you 10 points for being under your target time, 9 points for each minute under and so forth. We modified the system to score better than 4% below. 3% less earns you 9 points. The percentage system is more balanced between short and long races, as well as between average and high-flyers. It’s always been difficult for fast runners to win in handicap races. This is partly due to their tendency to be consistent, so it’s much easier to run 2 minutes less if your time is 55 minutes.
These are the solutions we came up with…
1. If a runner hasn’t been running a race in some time, we won’t be able to forecast a time. We assume the race time will be equal to that of the runner and give points starting from the middle (ie 6 points with our system). We will let them run the race again if it is not too far from their target time based on one race. Otherwise, we’ll award six points until they reach a reasonable average.
2. If a runner has a temporary injury or illness, it is usually easier to ignore the negative result and continue running if their previous race was successful.
3. We will treat any runner who shows a decrease in performance as an entrant, and create a new standard of performance.
Sometimes, runners perform in ways that are not expected. It doesn’t matter if it is better, it will feed back into the average of their next race. If the average exceeds 3 races, they are subject to a penalty equaling one third. You might also find someone who does not try hard to finish a race. “Who would do such a thing?” you might ask. It happens to someone in training for an important race and simply running the race to train. This is where it becomes simple to question the runner as to why it was taking so long and exclude it from the next calculation based on its insignificance of the current form.