This weekend, on a Formula One track in Italy, Nike will try to get one of three elite runners to do the seemingly impossible: Run a marathon in less than two hours.
If the half-marathon "dress rehearsal" run from March is any indication, the runners—Olympic champion Eliud Kipchoge of Kenya, half-marathon world record holder Zersenay Tedese, and two-time Boston marathon winner Lelisa Desisa—will have some help. They'll chase a Tesla mounted with a giant clock, be joined by elite pace runners (including Kenyan-born American Bernard Lagat), and get fuel delivered via scooter as they perform lap after lap around the oval race track. All of this, as well as super-cushioned Nike shoes designed specifically for this effort, will be in the hope of shaving about 3 minutes off the almost three-year old marathon record held by another Kenyan, Dennis Kimetto.
"It took about 20 years to get to this level, to reduce it by three minutes," says Robert Ojiambo, a physiologist at Moi University in Eldoret, Kenya. In 1998, the marathon record was 2 hours, 6 minutes. The record has been broken 8 times since then to shave off the additional three minutes; the mark currently stands at 2:02:57. As part of the separate SUB2 project, Ojiambo has done research to help knock off these last few minutes. But SUB2, which was started in 2007 and relaunched in 2014 by Yannis Pitsiladis, a professor of sport and exercise science at the University of Brighton in the UK, had a much less ambitious timeframe than Nike's run. "Maybe you need 20 years again," Ojiambo says. "Maybe in 2034."
By optimizing conditions and improving running economy with its custom shoes, Nike hopes to drastically speed up this natural process. Some of these optimized conditions, like the draft from the Tesla pace car, might mean the record could be unofficial. But Pitsiladis says that, maybe more importantly, it means the Nike effort won't set a new, repeatable standard with benefits for every runner. Nike's shoe for this effort, the Vaporfly Elite, won't even be available to the public until June. Even if you wear these shoes when they are released, you won't be able to run while pacing (and getting drag) off a Tesla, and you won't get custom fuel delivered to you.
"Our project is very much about innovations, and trying to clean up the world of distance running from drugs," Pitsiladis says. Many East African champion runners put up world-class times without the benefits of sports science by training with traditional methods or not optimizing their post-workout nutrition. In recent years, many of these runners, including 2016 Olympic marathon gold medalist Jemima Sumgong of Kenya, have been accused of doping; Sumgong tested postitive for blood-booster erythropoietin (EPO) in April.
Pitsiladis believes the innovations from his project can help enhance these naturally gifted runners' performances without the need for drugs. Sub2 has worked with Maurten, a Swedish sports drink company, to develop a sports drink that delivers more carbohydrate than traditional drinks without upsetting runners' stomachs. Pitsiladis has also worked with Vodafone on a smartwatch app to help with pacing. When Ethiopia's Kenenisa Bekele ran the second-fastest marathon ever in Berlin in September, he was using both of these products. "The whole idea of SUB2 is to develop these kinds of spinoffs, like the race to the moon, where we have so many new developments that can be used in all areas of life."
But even if you drink the Maurten sports drink (which you can) and use the Vodafone app when it's released to the public, Pitsiladis says there is no one-size-fits-all solution to running faster. "The way we prepare [former world record holder] Wilson Kipsang and the way we prepare Kenenisa Bekele is not exactly the same. We need to individualize the training," he says. Even the Nike project isn't the same for all three of its runners: The shoes to be worn in this weekend's race will be individualized for each runner.
So the best way to improve your own running times is to experiment on yourself. Think of it as your own SUB2—or maybe, more accurately, SUB3 or SUB4—project. This advice will help you make the self-experiment more measurable, more efficient, and just plain better.
Choose variables that could have a big impact.
When Shawn Arent is performing studies on athletes at Rutgers University, where he's director of the Rutgers Center for Health & Human Performance, he could be testing something really tiny, building on lots of other studies to push overall knowledge forward.
"It might take me months to carry out that study. But do you really have months to make small changes before an upcoming event," he asks. Arent suggests starting with things that are foundational to your training: Your running form, your nutrition, or how much you're training.
For instance, he says, studies have found that reducing stride length and increasing the number of strides you do per minute can not only make you faster, but it puts less force into your knees each time your foot strikes, which researchers believe could reduce injuries. Trying a few weeks of increasing your stride rate—and striding shorter—could make you faster. Or maybe eating a pre-run meal with more fast-acting carbohydrates might give you more pep during a long run. Once you've gone through bigger, foundational variables like these, then you could move on to things like supplements, caffeine, and other incremental changes, Arent says.
Figure out how to measure that impact.
To see if your variables are effective, you've got to test them, ideally in conditions as close to the race as possible. But you're probably not going to go out and run a marathon every day to test it.
"What you need is a meaningful sub-interval," says James Heathers, a postdoctoral fellow at Northeastern University who specializes in research methodology. If you were working on shortening your stride and taking more steps per minute, he says, choose a shorter distance to try it. If you do speed work, as many runners do, alternating fast intervals of 400 or 800 meters with slower breaks, you could do your faster, shorter striding during those intervals for a few sessions to see if you improve.You can count how many times your right or left foot hits the ground over a minute, or invest in a GPS watch that calculates your stride rate for you.
Eight hundred meters of results don't equal a marathon, of course. "You need to think about what the components are of you performing what you've done that are the reasonable things that relate to how well you'd do it in a marathon," he says. This is where the "meaningful" part of sub-interval comes in: If the shorter strides are at a much different pace than your marathon pace, it may not translate over.
Once you've found a meaningful sub-interval, make sure you're giving yourself lots of ways to measure it, Arent says. Keep track not just of how far you go on a training run, but how long it takes. Wear a heart rate monitor to see how your body's reacting. And also, he says, keep qualitative notes: Write down how your body felt during the workout with the new variable, how you felt mentally and emotionally during the workout, how you felt after, and how you felt as you recovered. Online training logs like Strava and Dailymile can help you here.
In his studies on body chemistry and biomarkers, Arent asks these kinds of questions. While the biomarkers often change before the athlete's feelings do, the body chemistry and emotional state usually wind up converging. Without a lab at your disposal, a journal of data and your feelings can help you track progress.
Test only one (or two) variables at a time.
Maybe you got gung-ho and read an article on pre-workout nutrition, another on stride length, one that said certain types of music can improve your workout, and one about pre-run caffeine and now you're ready to try EVERYTHING at once.
Slow your roll. If you really want to know what's working, try working in one change at a time, Heathers says. "Is it going to take longer? Yes," he says. "But you also need to go through a period of figuring out how these things are interrelated for you. [And if you try to test a bunch of variables at once,] you won't have any coherent thoughts about what any individual one means."
You also may not be "performing" that variable as well as possible if you're trying to concentrate on 5 others at once, he says. If you're testing a slower pace on long runs, as some elite marathoners do to maximize time on their feet, and you're also trying to increase stride turnover, you might not do either one well—meaning you won't know if either one helps.
Arent agrees that minimizing variables is key to success, but since you're probably short on training time, he thinks you could probably handle two non-competing variables—maybe changing what you eat and your pace on long runs.
Test your variables for at least a few weeks.
Most research studies last 8 weeks or more. You don't necessarily need to try something for that long, but as long as a new change isn't causing you pain, give it a few weeks, Arent says—even if you're not seeing results right away.
"When you make certain mechanical changes, just because it's a change, you might get a little worse before you get better," he says. When a golfer tries to change his swing, it might take some getting used to, but eventually improves his score. "But if you don't notice a difference within three to five weeks, you're probably not going to stick with that new idea," he says. So move on to a new one.
Giving yourself a few weeks will also keep you from jumping from new thing to new thing—or worse, trying out something totally different on race day. "It's amazing how many people do that," Arent says. "They have this strategy all through training, and they do something different on race day. And they don't know what happened! Integrate this stuff into your training. Your training is the lab."
Just like a lab, remember to test how the variable is affecting your sub-interval. After a few weeks, perform a new test, and compare it to how you did on that sub-interval the first time out. If you've reached your new goal, Heathers says, try to repeat it at least once. Researchers try to repeat their results to confirm them, and it's useful in personal experiments, too: The first time you succeed at this goal, it could be extra motivation or adrenaline pushing the result. Heathers says you don't need a million confirmation trials, but "it's normal when it's twice."
Read This Next: If You Run, You Should Be Doing This Workout