ItsOnlyRunning wrote: Well, according to this article optimal pacing is to go out 3-6% faster than goal pace for the first mile.
interestingly, the paper doesn't actually show that at all.
I concede that the paper does actually say that, in the sense that the specific words, "In order to optimize 5-km performance, runners should start the initial 1.63 km of a 5-km race at paces 3-6% greater than their current average race pace" do appear in the paper, but what they did, the study they carried out, the experiment they conducted, does not make that point.
1) the study had only 11 subjects. this is not a statistically significant population size and therefore anything this paper might show is nothing more than an interesting anecdote. there is also no control group. when testing a training or performance intervention there should be a control group so that you compare the study subjects with the control group. in this study, the subjects were compared with themselves, which is, essentially, high school level playing at science for demonstration purposes. it isn't actual science.
2) to show that strategy A is optimum, you would have to show that it was better than any other possible strategy. all they did was test three possibilites. therefore, all they can say is that strategy A was the best of those three. they have literally no idea what the optimum strategy is, or whether or not there even is an optimum strategy.
3) their own words state quite clearly that what their experiment showed was that initial mile pace "can be 3 to 6% greater than current average race pace without negatively impacting performance." which means you don't get worse if you do this. it said nothing at all about making improvements.
4) their 3-6% pacing strategy was based on current best time, as performed on the treadmill in the lab, not the target time for the race or even on the runners known personal best. so, this has nothing to do with racing, it was about equalling their most recent performance. fiddling with initial pace does not make you worse than the run you did three days ago.
this is a very good example of a scientific paper claiming to show something the experiment protocol and subsequent analysis does not support. it is a very common occurence and the reason why I recommend a thorough read of any science paper you want to quote.
Cheers.
https://www.ncbi.nlm.nih.gov/pubmed/17149992