Case study · Football academy

Five seasons at Aris Limassol

360 Ergometric Tests · Data 25 June 2021 – 19 July 2026 · Published with the club's consent; no athlete is identifiable

Between June 2021 and July 2026 we put 1,620 children in the Aris Limassol academy through ergometric testing. That is 3,248 assessments in total. All the tests are field tests and were carried out at the club's own facility. Below are the four things the data showed. One of them is not flattering.

1,620athletes
3,248assessments
34months with assessments
400with 3+ repeat visits
Finding

Aris does not promote children on the basis of month of birth

Age groups are defined by year of birth. So within the same age group, a child born in January can be up to eleven months older than one born in December of the same year. At ten or twelve years old those months matter a lot: the child born earlier is usually taller, heavier and stronger, simply because they have had longer to grow. So in academies around the world, the older children are promoted more often. Not because they have more talent, but because they have more months. This is called the relative age effect, and it is one of the best documented phenomena in sports science.

Birth quarter within each athlete’s own birth year — 1,447 athletes aged 9-17
Each athlete is placed in the quarter of their own birth year — not of one shared year. In perfect balance each quarter would be 25%.
25.5%Jan–Mar
24.7%Apr–Jun
25.5%Jul–Sep
24.3%Oct–Dec
Ratio of the first quarter to the fourth: 1.05. Among the other clubs in our sample with 100 athletes or more, the same ratio reaches as high as 1.71.
1.05

What the number means. For every 100 Aris children born in the last quarter of the year, there are 105 born in the first. That is almost equal. In three other clubs in our sample the same ratio is above 160 to 100.

In plain terms: at Aris, a child born in December is just as likely to be in the academy as one born in January of the same year. Their month of birth costs them nothing.

This balance is not a given. Across all the children we assess in Cyprus, the four quarters hold roughly equal shares of births. So the imbalance we see at other clubs is not down to the country. It is down to how children are selected and promoted. Aris has the largest academy population we assess in Cyprus, and it does not have that imbalance.

Finding

The improvement was not just growth

Every child improves as they grow. The question is a different one: do they improve more than other children of their age? That is why we do not compare the raw values themselves. We compare each child's position within the Cypriot norms.

Change in position against Cypriot norms
Each athlete's first and last assessment. On average about 16 months apart.
Improvement Decline
Aerobic capacity (beep test)n = 489
+8.0
Agilityn = 590
+7.5
Maximum speedn = 593
+7.4
Speed over 30 mn = 593
+6.6
Speed over 10 mn = 593
+4.0
Flexibilityn = 573
−0.8
Body compositionn = 597
−5.2
Percentile points. The axis extends ±12.5 points either side of zero. Only athletes with at least two assessments five months or more apart are included. All the tests are field tests; aerobic capacity was estimated from the beep test.

Aerobic capacity rose eight percentile points. From the 44th percentile, below the national average, to the 52nd, above it. 62% of athletes improved. The same pattern appears in agility and in speed.

This improvement is the result of the club's training. Assessment does not improve anyone; it simply makes visible what the training achieved.

Caveat

Two things went backwards

Body composition fell five points and flexibility one. In body composition only 37% of athletes improved. That means almost two in three lost ground relative to their peers.

Why we are publishing this. A report in which everything improves is not a report, it is an advertisement. Assessment is worth doing precisely when it also shows what we did not want to see.

What it means in practice. Aerobic capacity and speed improved. Body composition and flexibility did not. These are two different interventions: nutritional guidance and a targeted flexibility programme. Neither of them would have emerged from a single assessment. It took the comparison across five years.

A note of honesty. Part of the decline concerns children who went through their growth spurt within this period. At that age the body changes quickly on its own. With the present sample we cannot fully separate what is down to growth and what is down to training, so we do not present it as a training failure.

Finding

375 children are right now at the most sensitive point

During the growth spurt the bones grow faster than the muscles and the tendons. Coordination temporarily drops and the risk of overuse injury rises. The child appears to be getting worse, when in reality they are simply growing.

Stage of biological maturation — 1,614 athletes
Each athlete's most recent assessment
68.1%
23.2%
8.7%
Before the spurt — 1,099 In the spurt (PHV) — 375 After the spurt — 140
Calculated from age, height and body mass, with the Moore (2015) model. It does not require a sitting height measurement. The classification updates automatically at every new assessment and is available to every coach.

Without assessment, these 375 children do not stand out. They look like athletes who have stopped progressing. With assessment, the coach knows who they are and what they need: adjust the load, do not raise the demands.

Why it took five years

One assessment gives a number. Four give a trajectory.

How many times each athlete was assessed
Repeat visitsAthletesWhat it allows
1 assessment701Snapshot — comparison against norms
2 assessments519Change — improved or not
3 assessments205Trend — in which direction
4 and above195Trajectory — at what rate, and whether it is changing
The athlete with the longest history has seven assessments across five seasons.

Rate of improvement cannot be calculated retrospectively. It needs history. When a club changes testing provider, it starts again from zero, because the new provider's first assessment has nothing to compare with.

What one assessment does not show

Balance in selection, rate of improvement, stage of maturation and the two parameters that went backwards: none of these shows up in a single assessment. They emerged because the same children were assessed for five years, with the same methodology, in the same database.

Data from the 360 Ergometric Tests database, 25 June 2021 – 19 July 2026. All the tests are field tests and were carried out at the club's own facility. Aerobic capacity was estimated from the beep test. Stage of biological maturation was calculated with the Moore (2015) model, from age, height and body mass. Comparisons are made against Cypriot norms built from 10,347 athletes and 22,154 assessments. Paired comparisons include only athletes assessed at least twice with a minimum interval of five months. Published with the consent of the Aris Limassol academy. No individual data is disclosed.

See the same picture for your own academy

It starts with a basic round of assessments at your facilities.

Request a proposal