Can We Predict Future Water Polo Talent Using Body Size and Fitness Tests?
- Darren Bezzina
- Jul 16, 2025
- 3 min read
Updated: Jul 22, 2025
When it comes to spotting future stars in sport, we often rely on the eye test — who’s fast, who’s strong, who stands out.
But what if we could use measurable data to guide those decisions?
That’s exactly what this Maltese study by Robert Spiteri set out to explore:
Can anthropometric (body size) and general performance variables help identify talent in junior male water polo players?
How the Study Worked
Participants: 43 U15 male water polo players
Groups:
SEL group: 16 players selected for the U16 national team
NON group: 27 players playing only at club level
Anthropometric tests:
Height, body mass
Waist, hip, and chest circumference
Upper limb span, dominant hand span
Body fat % (via 4-site skinfolds)
Performance tests:
Push-up, Sit-up, Chin-up
Triple-hop jump (THJ), Counter-movement jump (CMJ)
Other variables:
PHV offset (maturity)
Years of training
Key Findings
No major anthropometric advantage - The SEL group did not show significantly larger body measurements than the NON group overall. However, body mass and waist girth were significant predictors in the final model.
Maturity matters - The SEL group was significantly more biologically mature, even at the same chronological age.
Performance matters - The SEL group showed significantly higher scores in push-ups, chin-ups, and CMJ, indicating greater upper- and lower-body strength and power.
It’s not one factor — it’s the combination that counts - A binary logistic regression identified seven variables that, when considered together, correctly predicted selection status in 88.4% of participants.
Years of training
Body mass
Waist girth
Hand span
PHV offset
Chin-ups
Counter-movement jump (CMJ)
However, these variables should not be viewed in isolation. The predictive power of the model lies in how these variables interact as a group. When tested individually, most variables — particularly anthropometric ones — lost statistical significance. The model works because of their combined effect, not because each one is a strong predictor on its own.
🔬 Binary logistic regression is a statistical method used to estimate the likelihood of a player belonging to one of two groups (SEL or NON) based on several physical and experiential variables.
What This Means for Maltese Sport
This study offers valuable insight into how talent is identified in one of Malta’s most physically demanding sports — water polo.
Rather than proving that top athletes are simply “bigger” or “stronger,” the study reveals a more nuanced reality:
Maturity matters — The most successful athletes were further along biologically, not necessarily taller or heavier.
Physical Performance counts — Strength and power (push-ups, chin-ups, CMJ) clearly differentiated high performers.
Experience matters — Years of training was a strong predictor of selection.
Objective data can help — The predictive model showed we can use real data to improve how talent is assessed.
The Hidden Risk: Maturation Bias
The study raises a red flag:
Athletes who mature earlier are more likely to be selected — not always because they’re more talented, but because they appear stronger or more physically ready.
This creates a real risk of selection bias, where:
Early developers get selected
Late developers get overlooked
Long-term talent may be lost too early
A Call for Smarter Talent ID
Spiteri’s study contributes to a growing body of evidence suggesting that Malta needs more structured, long-term, and data-informed approaches to talent identification. These systems should:
Use tools like PHV offset to account for maturation
-Incorporate physical, technical, and psychological data
-Avoid over-reliance on current size or strength
-Encourage longitudinal tracking — not one-off tests
Full Citation
Robert Spiteri (2024). Anthropometric and General Performance Measurements for Talent Identification in Junior Water Polo Players [Masters dissertation, MCAST].
Let’s build talent systems based on science — not just size.
Comments