The Sports Gene, much of the modern conversation about high-level distance running has turned to talent: where it comes from, how to spot it, and how to develop it. One piece often missing from the conversation is what it actually means to be talented. We speak about "talented runners" as if there is one specific set of criteria that we evaluate talent against, but in truth, there are several different types of talent which don't have any inter-dependability. By this I mean that just because one runner is talented in a certain way does not necessarily imply he or she will also be talented in another.
Broadly, I believe there are (at least) four different ways one can be naturally talented as a runner. Some are more easily assessed than others.
Natural running ability
This is what most people are thinking of when they say someone is a "talented" runner. They mean he or she has a high natural set-point of aerobic endurance, someone who can run fast times or impressive workouts without much or anything in the way of formal training. Some people like to explain this mostly in terms of genetics, while others point to an active childhood as the determining factor. Both, of course, are important, but for a coach, neither matters—you work with what arrives at your doorstop on day one. And if that newly-minted runner can already run at a high level without any history of training, that is always a good thing. This does not guarantee success, as I'll explain below, but people who start at a very high level of fitness naturally have a distinct advantage.
Natural running ability is also the easiest type of talent to identify: all it takes is a race or time trial. At the school I coach at right now, our best runner was spotted as a freshman thanks to a fitness test all sprinters on the track team undergo called the Cooper test. It was devised by Dr. Kenneth Cooper, sometimes known as the "father of aerobics" (rightfully so, given that he coined the word), for quickly and accurately evaluating the VO2 max of a large number of test subjects in the field, e.g. military recruits. The test is simple: cover as much ground as you can in 12 minutes, ideally running but taking walk breaks if needed. The distance you cover, in meters, is plugged in to a formula which predicts your VO2 max. Cooper's test is reasonably accurate when comparing test results to lab-determined VO2 max.
In our case, this freshman "sprinter" ended up finishing over 500 meters ahead of everyone else at the twelve-minute whistle. Within a few weeks he was running distance workouts with our top athletes.
For distance runners, any time trial of 1600 to 2400 meters should suffice for evaluating natural running ability. Longer tests can mask true running ability, because it's very hard to run a good 5k or 10k off natural running ability alone—these events just rely too much on training, not to mention proper pacing, which is an acquired skill. Sprint and true mid-distance athletes should be evaluated with shorter time trials that more accurately measures the shared aerobic and anaerobic components of the specific event—a 300m time trial for the 400 meters, for example, or a 500m time trial for prospective 800m runners.