Dollar Hill Race, 2006. Age against Time
We all know thattime catches up with us but at what age can we expect our course times to slowdown and how slowly or quickly will times increase once the rot has started? � The recent Dollar hill race gives a goodinsight to these questions. The chartbelow shows the time for male competitor�s to complete the course plottedagainst the competitor�s age on the race day. With only 135 data points the data set is fairly limited but has obviousrelevance to our sport.
The data shows thattimes are flat between age 20 and mid to early 30�s. � After mid 30�s there an apparently inexorablerise in times and the rise appears constant up to the end of the available data. � The rate of increase is approx 50 seconds peryear or about 1% slower each year compared to the average time for a 30 yearold competitor. However such changes inthe average performance hide large variations between individuals at anyage. Whist times of around 80 minutes orless recorded by people up to age 50 also suggests large differences betweenindividuals.
Figure 1.Time to complete the course against age on raceday for male competitors. � The line shown is a Epanechnikov smoothing function using a 50% setting. � Only those competitors� declaring a clubaffiliation and completing the course in less than 2hrs 30mins are included inthe plot giving a count of 135. �
Figure 2 shows the same data but for female competitors however with only 34 data points andmuch smaller age range interpretation is much more limited. � Nonetheless, both the age at which timesstart to increase (early 30�s) and the rate of increase with increasing age (50seconds per year) are both very similar to the male dataset.
Figure2. � Time to complete the course against age on race day for female competitors. The line shown is Epanechnikov smoothing functionusing a 65% setting. Only thosecompetitors� declaring a club affiliation and completing the course in lessthan 2hrs 30mins are included in the plot giving a count of 34.
It has been saidthat hill races are won and lost on the descent. � Does that apply to the Dollar hill raceearlier this year? We can examine thisby checking the correlations between the times for each leg and the overalltime. All legs show high correlations withthe final time but the highest correlation is with leg 1 (ascent) and thepoorest with the final (descent) leg which suggests that although all legs areimportant, good ascenders will more likely win the race.
Correlationsof time for each leg with overall time.
| || leg1 || leg2 || leg3 ||leg4 |
| Correlation with overall time ||.949 ||.916 ||.925 ||.884 |