Published on May 4, 2026 at 06:00 AM
A classic example of the so-called “exploration-exploitation dilemma” is ordering food at a restaurant. I’ve had pad thai a few times and know it’s good; You’ve heard Pad krapow is great but never tried it. Should you stick to the familiar or venture into the unknown? It’s a familiar dilemma, one that comes up all the time in different forms: what TV show to watch, who to date, which job offer to consider, and so on.
Athletes also have to constantly navigate dilemmas of exploration and exploitation, he says New paper In the magazine Psychology of sport and exerciseby Katja Roetz and two colleagues at the University of Hamburg. Sport-related trade-offs range from the most general (choosing the sport in which you will compete) to the most specific (deciding which brand of gels will be used during the race). But how do we know whether we are clinging too closely to tradition, or jumping too enthusiastically into every new fad?
The fact is that there is an extensive scientific literature on the decision-making process related to exploration and exploitation, which draws on topics such as behavioral economics, evolutionary biology, and computer science. Ruetz presents three main types of exploration and exploitation dilemmas you might face as an athlete:
1. Find the best option
Ruetz’s example is that kids try a lot of different sports, but ultimately have to choose one in order to maximize their potential. I think the same is true, to some extent, for adults: I love running, but I also enjoy basketball, have dabbled in climbing, and am curious about cycling. As much as I would like to pursue all of these things in parallel, I have limited time and energy, so I have to prioritize – although my goal at this point is to maximize overall enjoyment and health more than performance.
The scientific category that addresses such problems is information search, a classic example being what was called (when first studied in the 1950s) the secretary problem. You’re interviewing candidates for a job, and you have to evaluate each one and rank them against previous interviewees. The question is: At what point do you decide that a candidate is good enough to hire, rather than continuing to interview more candidates?
Under certain somewhat artificial circumstances (for example, if you assume that you can only hire someone immediately after interviewing them, and you cannot then go back to hire previous candidates), you can come up with a mathematical rule. Of the total number of available candidates, you should interview 37 percent of them Without hiring anyonein order to get a sense of how good the overall pool of candidates is. Next, you should appoint the first candidate who is better Of anyone I’ve met so far.
The numbers specified here are not useful because the restrictions are very unrealistic. But the general approach makes sense: You should first explore your options, and allow yourself to control the choices only after the initial period of exploration. As it happens, this meshes well with his findings Scholars who study talent development: Elite adult athletes typically played more different sports as children and specialized later than elite children who did not reach the next level. explores then exploited.
2. Knowing when to switch to a new strategy
The example here is switching from an established routine to a new trainer or training program, although it can also apply to topics such as nutritional supplements (adherence Beet juiceor switch to baking soda or Broccoli extract?) and nutrition (Double your carbohydrate intake?). In my book Jane the explorerThe main example I used in my chapter on the exploration-exploitation dilemma is speed skater Niels van der Poel, who… An amazingly unconventional training approach The lead-up to the 2022 Olympic Games has rocked the endurance world.
A similar field of scientific study is animal feed. If you are a bee sucking nectar from a patch of flowers, you eventually have to decide when to abandon your current patch of flowers and fly off into the great unknown in search of a better patch. There’s no guarantee that this other hypothetical patch will be better than your current patch, and you’ll have to expend valuable energy getting there, so the decision to go ahead is a leap of faith.
The mathematical approach to foraging (which actually seems to be consistent with how animals in the wild make foraging decisions) is called the marginal value theory. The bee has a rough estimate of the amount of pollen available in a “typical” flower patch, so it sticks to the current flower patch until pollen levels drop below this average level, and then moves on.
The crucial point here is that the existing flower patch suffers from diminishing returns, because the bees gradually use up the pollen in the patch. So it may be a smart decision to stay where you are for now, but sooner or later you will have to move on. You could say that coaching philosophies follow similar dynamics. You move to a new coach or adopt a new philosophy (Norwegian traininganyone?). After a period of adjustment, you can make rapid progress. But after a few years, you’re no longer improving as quickly, perhaps because your body and mind have adapted to the stimuli of the new training program. Is it time to make another switch?
Training philosophies are harder to measure than pollen levels, but the same pattern applies: long periods of exploitation interspersed with short periods of exploration.
3. Find the right balance
When you’re in the arena, whether literally or figuratively, the answer to the dilemma of exploration and exploitation is usually “both.” Ruetz uses the example of strategy during a tennis match: If you normally play from the baseline, should you hit the net?
This decision is based on a wide range of factors that are constantly changing in real time. You may gain a temporary advantage by surprising your opponent, but the success of the strategy will depend on how well you hit the ball and whether your opponent can return your shots. Even if you succeed at first, your opponent may soon adapt to your net game, tipping the scales in favor of a return to the baseline. There’s no permanent answer here: it’s an ever-changing mix of exploration and exploitation.
The scientific model for this type of decision-making is a gambling game called “One-Armed Bandits,” where you have a pocket full of coins, and have to choose between a set of slot machines (“One-Armed Bandits,” as they are sometimes called). You don’t know the odds of winning for each machine, so you have to try a lot of different machines (i.e. explore) to see which ones are best to play (i.e. exploit). But you also have to keep exploring, because the odds of each machine are constantly changing.
There is no perfect solution for more realistic versions of multi-armed bandits. But there are several shortcuts that help increase your chances of getting a good result. My favorite is one called the higher confidence algorithm, which scientists sometimes explain as “be optimistic in the face of uncertainty.” In essence, it involves choosing the option with the greatest amount of realistic upside.
For example, if you’re confident you can finish in the top five by sitting in the pack, but you think you have a 10 percent chance of winning the race by building a big lead, the Supreme Confidence algorithm tells you to go for it. You might win, and even if you don’t, you’ll have less regret. (Regret is also a mathematical quantity here: the difference between how things could have ended and how they actually ended.)
This last example – racing conservatively versus going for it – is one I think about often when I think about the exploration-exploitation dilemma. My biggest unfulfilled dream as a runner was to run a sub-four-minute mile. My best time in the 1500 meters is 3:42.43 (equivalent to (According to World Athletics) to the mile in 4:00.03. I can’t count the number of 1,500 mile races I’ve done in my career, but strangely enough I’ve always stuck to the same slow start, fast finish strategy. I could never get to the halfway point fast enough to walk a four-minute mile. Looking back, this omission baffles me. How could I have spent all those years pursuing a goal and not once explored a fairly obvious method for achieving it?
I don’t know if reading Roetz’s paper (with the help of a time machine) changed my career. But I spent a lot of time reading and thinking about the exploration-exploitation dilemma while writing Jane the explorerand I came away thinking it was a useful and underappreciated framework for thinking through training and racing decisions. This won’t necessarily provide a simple or direct answer, but it will get you thinking about the right questions.
For more race science, sign up Email Newsletter And check out my new book The Explorer Gene: Why we seek great challenges, new flavors, and empty spots on the map.
