Talent identification: tests and assessment batteries.

Part of the complete guide: How to improve your archery score

"That kid has talent." You hear it constantly on the shooting range, based on a glance, an impression. Sometimes the intuition is right; often it is not. The interesting question is not whether talent exists, it does, but whether we can measure it with something more solid than a feeling. And the answer, today, is largely yes.

Research on talent identification in archery has developed assessment batteries covering physical capacity, stability, and vision, and even predictive models built on measurable parameters to estimate an athlete's potential. These are valuable tools, especially for those who work with groups: coaches, clubs, and federations, precisely the settings I work in.

This article covers what talent means in the context of archery, which tests exist and what they measure, how predictive models work, and, the point I return to at the end, what talent assessment can tell us and what it will never be able to tell us.

What "talent" means in archery

In sport, talent is not a single quality. It is a set of predispositions that make development faster and raise its ceiling. In archery the relevant components are physical (strength, isometric endurance, body structure), motor (postural stability, fine coordination), visual (ocular function and dominance), and mental (emotional regulation, capacity for motor learning).

Because talent is multidimensional, it cannot be captured by a single number. An athlete can be exceptionally strong physically and fragile emotionally, or the reverse. That is precisely why rigorous assessment batteries return a profile, as shown in the figure: a picture of strengths and vulnerabilities, not a grade from one to ten.

Assessment batteries

An assessment battery combines tests that cover different dimensions. The aim is to build, in a few hours, an objective picture that would otherwise take months of observation, and that stays comparable over time.

Physical and functional tests

Physical tests are the most straightforward to administer and the most widely used. They do not measure "how well you shoot today," but how solid the foundation is on which technique can be built: a strong posterior chain, good scapular control, and adequate isometric endurance are the conditions under which technique takes hold and holds up over time.

That said, physical tests should not be conflated with talent itself. A strong result here tells you that physical bottlenecks to development are absent, not that development will follow. Commitment, method, and consistency are still required. Physical tests rule out limiting factors; they do not guarantee outcomes.

Machine learning and prediction

In recent years, several studies have tried to go beyond reading individual tests, using machine learning to combine multiple parameters, fitness, anthropometry, stability indices, and estimate expected performance or an archer's potential. The approach is promising: a model can identify combinations of factors that the human eye cannot weigh simultaneously.

It deserves careful scrutiny, however. One study in this area reported very high predictive accuracy, but on a small sample with no external validation. A real risk with models built this way is that the model learns to recognise parameters already correlated with baseline performance, without actually measuring future potential (data leakage). Until validations on large, independent samples exist, these tools should be read as a promising direction, not as reliable instruments for making decisions about athletes. A model can say "this profile, on average, tends to develop further," not "this athlete will become a champion."

Reliability and validity

A test is only meaningful if it is reliable (it produces the same result under similar conditions when repeated) and valid (it actually measures what it claims to measure, and bears a genuine relationship to real performance). These are two requirements that research verifies explicitly, and that practitioners should demand before using test results to make decisions about athletes.

In practice: be sceptical of improvised tests and "feel-based" assessments presented as objective. A serious protocol is standardized, repeatable, and documented. That is precisely the difference between a data point you can compare across time and an impression that shifts with the observer's mood.

What it is actually for, and what it is not

This is the point that matters most. Talent assessment is not for eliminating people, and using it that way is a mistake in both human and technical terms, because development is far too variable to write off a thirteen-year-old on the basis of a single test. It serves a different purpose: understanding where an athlete is strong, where they are vulnerable, and therefore where to direct the work and how to allocate the ever-limited resources of a club or a squad.

Used that way, it is a valuable and genuinely constructive tool: it turns vague judgements like "has talent" or "not the right type" into readable profiles on which to build a development pathway. In the end, what you do with talent matters more than the talent itself, and a good test tells you exactly where to start.

Arcos: from individual testing to a system, and what it could become at a national level

Everything described so far, the tests and assessment batteries, has an obvious practical limitation: it requires time, expertise, and above all, a coherent way to collect and interpret data. A test administered carefully at one club and done differently at another produces results that cannot speak to each other. The numbers stay locked in the drawer of whoever collected them, and almost all their value is lost.

That is the starting point for Arcos, the tool I am developing: a way to bring structure to this work, standardize assessment protocols, collect data consistently, and make it readable over time, both at the individual athlete level and across a group. Not a gadget, but the infrastructure that is currently missing for turning scattered assessments into useful knowledge.

The more interesting question, though, is what could happen if a tool like this were adopted not by individual practitioners, but at a national federation level. When assessment data from many archers is collected using the same protocols, something becomes possible that no individual can achieve alone.

• More robust talent identification. With data collected consistently across a large population, tests stop being isolated snapshots and become reference points. It becomes possible to understand what is genuinely predictive and what is not, on Italian data, not figures imported from other contexts.
• New research and new literature. A standardized data archive is the raw material of research. It opens the door to archery studies conducted in Italy, on Italian archers, which today barely exist.
• Coach education. When data is readable and shared, coach education stops being a matter of passing on opinions and becomes a matter of reading evidence, grounded in real, measured cases.
• Targeted physical preparation. Knowing where a group is strong and where it is vulnerable, in numbers, makes it possible to build physical and technical preparation around genuine priorities rather than impressions.

This is a vision, and visions need to be tested in the field before they become promises. But the logic is straightforward and robust: assessment on its own is worth little. Standardized assessment, collected at scale and read over time, becomes a driver that improves the whole system, from the grassroots to the top. Arcos is built to be that driver.