Burlington Bike Path, Burlington, Vermont - Parks, Recreation & Waterfront

Do Footballers Change Position as They Age? What the Data Can and Cannot Show

It is one of football's most repeated observations: as a player loses pace, he drops deeper, moves inside, or reinvents himself in a new position. The claim is intuitive and the anecdotes are famous. But it is also a testable claim, and the data available to test it is weaker and stranger than the confidence of the assertion suggests.

The question, stated precisely

The loose version — "players change position as they age" — is too vague to examine. It bundles three different claims that need separating:

  • That a player's recorded position changes over his career.
  • That his function changes while his recorded position stays the same.
  • That his output declines and the position change is a response to that decline.

These are not the same question, and public football data answers them with very different levels of confidence.

What the data can actually see

Position information in most databases comes from team sheets and squad listings. It is a label describing where a player lines up at kick-off, and it is coarse by design. A player can move from an advanced role to a much deeper one within the same label without the record moving at all.

Beyond the label, four types of evidence are usable:

  • Average position and touch distribution, which describe where a player actually operated rather than where he was listed.
  • Action mix — the proportions of his involvements that are dribbles, progressive passes, aerial duels, tackles, shots and so on. This is the most sensitive indicator of a changing job.
  • Minutes and appearance patterns, including how often a player is substituted and at what stage.
  • Output rates such as goal involvements or progressive actions per ninety minutes.

Note what is missing. Sprint counts, top speed, acceleration and total distance are collected by clubs and by some competitions, but they are inconsistently published, so most public analysis of ageing is inferring physical change from on-ball behaviour rather than measuring it.

What the patterns tend to look like

Across large samples, three things show up reasonably consistently.

The first is that when recorded position does change, the direction of travel is overwhelmingly one-way: toward the team's own goal. Forwards become midfielders, wide attackers become full-backs or inside midfielders, midfielders occasionally become defenders. Movement in the other direction, later in a career, is rare enough to be treated as an exception rather than a pattern.

The second, and more common, is that the label does not change at all. What changes is the action mix inside it. A wide forward whose dribble volume falls while his passing share rises has changed job without changing position. A striker taking fewer touches in behind and more with his back to goal has done the same. This is the version of the phenomenon that most players actually experience, and it is invisible to anyone reading position labels alone.

The third is a shift from reactive to anticipatory actions. High-intensity, late-arriving interventions — recovery sprints, chasing lost causes, contesting fifty-fifty situations — decline in frequency, while positional and pre-emptive actions, such as interceptions taken from good starting positions, hold up better or increase. Whether this reflects declining capacity or accumulated judgement is not something on-ball data can settle, and honest analysis should say so.

The pattern is not uniform across positions

One detail gets lost when the question is asked about "players" in general: the scope for repositioning is not evenly distributed across the pitch.

Wide attackers have the most options. Their skill set — one-versus-one ability, crossing, stamina along a touchline — maps onto several other roles, which is why the winger-to-full-back and winger-to-inside-forward transitions are the most frequently observed of all.

Central midfielders have the next most, because the difference between an advanced eight and a deep-lying organiser is a matter of starting position and instruction rather than a different technical toolkit.

Centre-forwards have fewer. Dropping into a creative role behind the striker demands a passing range that not every forward has developed, and the alternative adaptation — becoming a physical reference point who plays fewer minutes — is a narrowing of the role rather than a change of it.

Goalkeepers are the outlier, effectively immune to repositioning and ageing on a schedule of their own. Their careers are typically extended by the fact that the job's core demands are less dependent on the attributes that decline earliest.

Any analysis that pools all outfield players together is therefore averaging over groups with structurally different options, which flattens exactly the variation that makes the question interesting.

The confounders, which are unusually severe here

This is a question where the obstacles matter more than the findings.

Survivorship is the largest. Any sample of thirty-four-year-old professionals contains only those who remained employable at that age. Players whose decline was steep and unmanaged left the sample years earlier. So a study showing that older players adapt successfully to new roles is partly measuring the fact that failing to adapt removes you from the data. The apparent resilience of ageing players is an artefact of who is left as much as it is a finding about ageing.

Minutes selection is the second. Older players are increasingly used in matches and situations that suit them — at home, against weaker opponents, from the bench in specific game states. Their per-ninety numbers therefore describe a curated set of circumstances, not a representative sample of football.

Team context is the third. A role change is frequently a squad-need decision rather than a player-decline decision. A club short of full-backs may move a winger there at twenty-six, and the resulting data would look identical to an ageing-driven repositioning.

League migration is the fourth, and it is systematically underweighted. A player moving to a less demanding competition in his thirties will often show stable or improving numbers. Nothing about him improved; the peer group changed. Any ageing analysis that does not control for competition level is partly measuring transfers.

Finally, there is label lag. Squad lists and profile pages update slowly and inconsistently, so a recorded position change may postdate the actual functional change by a season or more. Platforms such as RubiScore maintain per-season club and appearance histories precisely so a career can be reconstructed season by season rather than from a single current profile, but the underlying labels still carry this lag.

What the evidence cannot settle

Three claims commonly made in this discussion are not supported by public data, and it is worth being explicit about them.

  • That the position change caused a career extension. The counterfactual — how long the player would have lasted without the move — does not exist.
  • That physical decline is the mechanism. Without published physical data, this is an inference from on-ball behaviour, and behaviour is also shaped by tactics, instruction and role.
  • That there is a single peak age. The commonly cited peak band is an average across very different profiles, and the shape of the curve differs markedly by position and by playing style. Goalkeepers and physically dominant defenders age on a visibly different schedule to players whose value depends on acceleration.

Verdict

The honest answer is a qualified yes, with the emphasis in an unexpected place. Formal position changes late in a career are real but relatively uncommon, and they are the least interesting version of the phenomenon. The widespread, measurable change is the redistribution of a player's action mix within a stable position label — less running behind, more receiving to feet; fewer recovery sprints, more anticipation; less volume, more selection.

What the data cannot do is confirm that ageing is the cause in any individual case. Survivorship, minutes selection, tactical instruction and league changes all produce the same statistical signature. The pattern is genuine at the population level and unreliable at the level of a single career.

How to read an ageing player's numbers

  • Compare action mix across seasons rather than headline output, because output is the last thing to move and the first thing to be misread.
  • Check the competition level in every season before treating a trend as a trend.
  • Read minutes context: starts versus substitute appearances, and the game states he was used in.
  • Look at average position and touch maps rather than the position label, which is the slowest-moving field in any player record.
  • Treat a single season as noise. Two consecutive seasons pointing the same direction is the minimum worth interpreting.

Season-by-season appearance, position and club-history records for individual players are published on rubiscore.com, which is the level of granularity this question needs — career averages will not answer it.