Your Best Racing Strategy; How About a Positive Split?

By: Brad Deel

One of the more common pieces of advice you’ll hear about running a race is that the best strategy for your fastest possible time is to run even or slightly negative splits.  As someone who tries to be an evidence based runner, I always ask whether the evidence supports theory.  In my semi-regular challenge of running’s conventional wisdoms, I think the evidence may be pointing toward a slight positive split as the best strategy.  Pete Pfitzinger has even suggested planning a one minute positive split for your best marathon so I don’t think I’m entirely off base.  Hear me out if you will.

Scientists at the University of New Hampshire constructed a study with 11 members of the University’s Women’s Cross Country team.  They each ran two 5Ks to establish a baseline.  Then, they ran three more 5ks with very different strategies.  The first strategy was to run the first mile at their 5K pace.  The second strategy was to run the first mile 3% faster than their 5K pace while the third strategy was to run their first mile 6% faster than their 5K pace.  They could run the rest of the race however they wanted.  The surprising result was that the 6% faster pace produced an average time of 20:39, the 3% faster pace produced an average time of 20:52 and the even pace produced an average time of 21:11.    

One of my favorite web sites is which is something of an extended blog by two South African exercise physiologists.  They analyzed world record performances in the 10,000, 5,000, mile and 800 over the past 50 years or so.  Until recently, the first and final kilometers of both the 10,000 and the 5,000 were the fastest of the race.  Some have noted that world records since the mid 1970’s have been more evenly paced.  True, but I think that owes more to the use of paid pacers in record attempts.  Guess what?  You won’t have a professional pacer in your next race.  The mile was evenly paced while the results in 800 were extremely clear.  In the 26 world records analyzed, the second lap was always slower.  Their conclusion is that it is impossible to run an optimal 800 meters with a faster second lap.  Not difficult.  Impossible.

My final piece of evidence for your consideration comes from the West Virginia 5K Championship on June 23, 2012.  All of the top 25 runners ran faster in the first mile than their overall pace.  All of them.  It is not until you reach #29 that you find someone who ran the first mile at the same pace as the overall pace and the first finisher who ran the first mile slower than the overall pace finished 32nd.  There were 68 finishers under 20:00.  I chose 20:00 as an arbitrary time that most recreational runners would consider at least reasonably fast.  Of those 68 finishers, only four of them ran the first mile at a slower pace than their overall pace.  I find it very difficult to believe that 64 of the top 68 finishers, including all of the top 25, ran a sub-optimal strategy.  It seems much more likely that the optimal strategy – at least if you can’t employ a professional pacer, is to run a positive split. 

In other words, run like hell and then hang on for dear life.

See ya out there.

Our goal is to help our community
by promoting fitness and
raising millions for charity.

Our Featured Races

Sign up for our Newsletter!