8,153
edits
Changes
no edit summary
[[File:HistogramOfBestSplits_Extreme_Partial_All.jpg|right|thumb|400px|
This graph looks at the best performance from runners who had both a positive and negative split result (12,425 or 1.6% of the finishes). Of those runners, 52% did better with a negative split and 48% on a positive split, but notice the peak for the slightly negative splits.]]
The results of 26 marathons covering 876,703 results for 754,851 runners were analyzed to look at the differences between positive splits (going slower in the second half) and negative splits (running the second half faster). Overall 13% of finishes were negative splits, with a mean split of positive 8.25%. Faster runners have a narrower distribution of splits and a mean closer to even. Looking at the 192,585 runners who have more than one finish in the same city, only 10% of the best performances are with a negative split, though a higher portion of the 2:00 to 2:30 and 4:00 to 5:00 finishes were negative. The subset of runners who have both a positive and negative split shows that 52% had their best performance with a negative split and the most common split for a best performance is negative 0-1%. Not surprisingly the biggest change in performance relates to extreme positive splits being much slower than more even splits, probably due to [[Going out too fast]]. Overall, the data suggests that it's beneficial to have a split time that is close to even, with a slightly negative split possibly being optimal.
=Method=
The New York City and Chicago marathons use electronic timing that records the start, finish and half way times for runners. These races have little elevation change; NYC has around 900 feet (300m) of ascent and descent, and Chicago is effectively flat. Data is publically available for the New York marathon from 2000 to 2011 and the Chicago marathon from 1998 to 2012. (The results of the 2007 Chicago marathon were excluded as that year the race was unusually hotand [[Impact of Heat on Marathon Performance| temperature has a big impact on marathon performance]].) The results were grouped within race location using name and age so that multiple finishes by the same individual could be compared. This gives data for 26 races, covering 876,703 results for 754,851 runners.
=Runners with multiple entries=
We have 72,259 runners with multiple entries, with the distribution shown below.