Why we test

Every test in a battery should answer a question that changes what you do next. Review papers on athlete profiling agree that tests should be chosen for the decision they inform, not for convenience or habit [1, 2]. In team sports, testing typically serves five purposes:

1. Profile. Benchmark each athlete against the physical demands of their sport and position, and identify individual strengths and weaknesses [2].
2. Programme design. Use the profile to individualise training emphasis, loads and targets. A strong but slow athlete needs a different block from a fast but weak one [2].
3. Monitor. Track adaptation and fatigue across the season, so training can be adjusted before performance drops [1].
4. Evaluate. Check whether the programme is producing the changes it was designed to produce [1].
5. Return to play. Compare an injured athlete with their own pre-injury baseline, giving objective criteria alongside clinical judgement.

Key concepts: validity, reliability and standardisation

A test is only useful if it measures the right quality, gives the same answer when nothing has changed, and is run the same way every time.

Validity

Validity asks whether a test measures what it claims to measure, and whether that quality matters in competition. A valid test relates strongly to a criterion measure and to the demands of the sport. A 1RM bench press may be a valid measure of upper-body strength, but it is only useful if that strength matters for the position being tested [1, 2].

Reliability: intra-rater and inter-rater

Reliability is the consistency of a measurement when the athlete’s true ability has not changed. In practice, much of the error comes from the person running the test, so coaches should check two kinds [1]:

• Intra-rater reliability: does the same coach get the same result when testing the same athlete on separate occasions?
• Inter-rater reliability: do different coaches get the same result when testing the same athlete?

Both matter whenever human judgement is involved, such as hand-timing, rating technique, judging squat depth or measuring skinfolds. Weakley and colleagues recommend training assessors and keeping the same assessor across testing occasions where possible [1].

Reliability is often reported as an intraclass correlation coefficient (ICC). Koo and Li’s widely used guideline rates an ICC below 0.50 as poor, 0.50–0.75 as moderate, 0.75–0.90 as good and above 0.90 as excellent [4]. For S&C coaches, Hopkins argues that the typical error (the within-athlete standard deviation of repeated trials, often expressed as a percentage) is more useful than the ICC. The ICC shifts with how varied the squad is, while the typical error tells you directly how much noise to expect in one athlete’s score [3].

Standardisation

Standardisation means running each test under identical conditions every time, so differences between results reflect the athlete and not the setup. This covers the protocol, warm-up, surface, footwear, time of day, equipment calibration, verbal encouragement and the athlete’s preparation, such as sleep, food and caffeine [1]. Unstandardised testing inflates the typical error and hides real change.

Signal versus noise

To decide whether a change is real, compare it with two benchmarks:

• Smallest worthwhile change (SWC): the smallest change that matters for performance. A common shortcut is 0.2 × the between-athlete standard deviation, based on Cohen’s small effect size [3]. Weakley and colleagues caution that this blanket rule can underestimate a meaningful change. Where possible, set the SWC from what matters in the sport [1].
• Typical error (TE): the noise in the test. Because a change score includes error from two tests, its noise is roughly TE × √2 [1].

If the typical error is larger than the SWC, the test cannot reliably detect the changes you care about in an individual athlete.

Testing principles: order and standardisation

Run the least fatiguing tests first and the most fatiguing last, so that fatigue from one test does not contaminate the next. Both the NSCA’s Essentials of Strength Training and Conditioning and recent profiling recommendations put low-demand measures first and exhaustive endurance tests last, and keep the same order on every testing occasion [1, 5]. A practical order for a team-sport battery is:

1. Anthropometrics. Height, body mass and body fat. These are non-fatiguing and should come first.
2. Leg power. Countermovement jump (CMJ) and/or reactive strength index (RSI). These are short, maximal efforts that are very sensitive to fatigue.
3. Sprint speed. 0–10 m and 0–30 m. Acceleration and maximum velocity need a fresh athlete.
4. Strength. Isometric mid-thigh pull (IMTP), NordBord, ForceFrame, and 1RM squat, bench press and chin-up.
5. Anaerobic and aerobic capacity. Phosphate decrement test (PDT), 30-15 Intermittent Fitness Test, Yo-Yo Intermittent Recovery Test and the Bronco. These are exhaustive and always come last, ideally on a separate day.

Alongside a fixed order, standardise the conditions around the tests [1]:

• Same time of day and, where possible, the same tester.
• An identical, scripted warm-up.
• The same surface and footwear, indoors where possible for sprint tests.
• At least 48 hours after hard training or a match.
• Familiarisation trials before the first recorded test, so a learning effect isn’t mistaken for improvement.
• A record of conditions such as wind, temperature, sleep and recent training load, so unusual results can be explained.

The test battery and technology

A team-sport battery should cover each physical quality that matters for the sport, using the most reliable tool available for each. The table lists the tests in testing order.

Scroll the table sideways on a phone.

A few notes on the technology

• Force plates record the whole force–time curve, not just jump height. They show how an athlete produces a jump (for example, RSI-modified and rate of force development), which can change with fatigue even when jump height doesn’t [6].
• Timing gates should be dual-beam and set at a fixed height, with a standardised start position. Single-beam gates can be triggered early by a swinging arm or lifted knee [7].
• Isometric tests such as the IMTP, NordBord and ForceFrame give maximal-strength data with little fatigue, so they suit in-season use better than a 1RM [8–10].
• GPS/LPS units and linear encoders add ongoing monitoring of training load and daily readiness between formal testing blocks.

The simplest reliable test beats an expensive noisy one. Before buying technology, check its published reliability and run your own reliability trial with your athletes and testers [1, 3].

Applying the data

Test at set points in the season, compare each athlete with themselves, and turn every result into a training action.

When to test

In a congested fixture schedule, rely on quick, low-fatigue tests such as the CMJ and isometric strength tests. Leave exhaustive tests for pre-season.

Compare athletes with themselves

A squad ranking hides individual change. Judge each athlete’s change against the SWC and the test’s typical error, as described above [1, 3]. A change smaller than the typical error is likely noise.

Standardise scores to compare across tests

To see an athlete’s profile across tests with different units, convert each result to a z-score [15]:

A z-score of +1 means one standard deviation above the squad average. For timed tests, where lower is better, reverse the sign. Plotting z-scores side by side shows each athlete’s strengths and weaknesses at a glance. Turner and colleagues combine them into a single Total Score of Athleticism [15].

Interpret in context

Read results alongside position, training age, injury history and recent training and match load. A drop in sprint speed after three games in eight days means something different from the same drop in pre-season.

Close the loop

Feed results back to athletes and coaches quickly, ideally within 48 hours, in a simple format. Then link each result to a decision, such as an extra strength block for a weak athlete or reduced load for a fatigued one [2].

Key takeaways

• Test with purpose. Every test should inform a training decision [1, 2].
• Check reliability first. Know each test’s typical error and the intra- and inter-rater reliability of your testers [1, 3, 4].
• Standardise everything. Use the same order, warm-up, surface, timing and tester on every occasion [1, 5].
• Let the question choose the technology. A simple reliable test beats an expensive noisy one [1, 7].
• Interpret in context. An athlete’s change against their own baseline matters more than their squad rank [1, 15].