Star Excursion Balance Test: Protocol and Normalizing Reach Distance to Limb Length

A clinic runs the Star Excursion Balance Test on two athletes back to back. Athlete A, a 6'2" post player, reaches 71cm anterior on the leg she's screening. Athlete B, a 5'4" point guard, reaches 58cm on the same leg. Athlete A's number wins by 13cm on paper, except it doesn't tell anyone anything, because her limb is roughly 15cm longer than Athlete B's to begin with. Divide both numbers by each athlete's own limb length and the point guard's relative reach turns out to be the stronger of the two. This is the most common mistake made with the SEBT: reading raw centimeters as though they mean the same thing on every limb they're measured on.

The fix has been sitting in the literature for two decades. Normalizing reach distance to limb length converts a number that mostly reflects how tall someone is into a percentage that reflects how well they control their body over a moving base of support. What follows is the full 8-direction protocol, the landmark used to measure limb length, the formula that turns a tape-measure reading into a comparable score, and what the two studies anchoring this method actually found, effect sizes and sample limitations included.

Why a Raw Reach Distance in Centimeters Doesn't Tell You Much

The Star Excursion Balance Test asks an athlete to balance on one leg at the center of a grid and reach as far as possible with the other leg along eight lines radiating out at 45-degree intervals, without losing balance and without putting weight on the reaching foot. It exposes deficits in dynamic postural control, the kind of single-leg stability an ankle or knee needs mid-cut, not what a static balance-board reading picks up. That part of the design holds up well. What gets skipped is what happens to the number afterward.

Reach distance scales with leg length almost by definition: a longer limb sweeps a longer arc before balance breaks down, independent of the underlying motor control. Gribble and Hertel's foundational 2003 work found raw reach distances correlating meaningfully with limb length across their subject pool, taller, longer-legged subjects posted longer raw reaches regardless of how stable they actually were on one leg. Compare two athletes, a roster, or the same athlete across a growth spurt using raw centimeters, and a real share of that difference is anthropometry, not balance.

Equipment and Measuring Limb Length Correctly

Limb length is measured with the athlete lying supine, legs fully extended and relaxed, as the straight-line distance from the anterior superior iliac spine (ASIS) to the most distal point of the ipsilateral medial malleolus, recorded to one decimal place. Take it twice per leg; a gap beyond 0.5cm means remeasure rather than average bad data. This ASIS-to-medial-malleolus landmark is the version Gribble and Hertel's normalization work validated, and the one that shows up in most published SEBT normative studies since. A different landmark, some clinics measure from the umbilicus instead, will produce a different percentage from the identical reach, so pick one method and never switch it mid-athlete.

Step-by-Step Testing Protocol

1. Warm-up (5-8 minutes): Light aerobic activity plus dynamic hip and ankle mobility.
2. Measure limb length bilaterally as described above, before any reaching begins.
3. Position the athlete barefoot at the center of the grid, hands on the iliac crests, and keep them there for every trial.
4. Practice trials: a minimum of 4 per direction per leg, 6 is the more conservative number most protocols default to, since the first several attempts in any new direction carry a real learning effect that fades by trial 4-6.
5. Test trials: 3 recorded reaches per direction per leg. The athlete reaches with the free leg, lightly touches the line with the most distal part of the foot without shifting weight onto it, then returns to a controlled two-leg stance.
6. Discard and repeat if the hands leave the hips, the stance foot lifts or rotates off center, the reaching foot is used to gain support, or the athlete can't return to a stable stance under control.
7. Recording: measure each valid touch point to the nearest 0.5cm, and keep the longest of the three valid trials per direction as the score, not the average, since the goal is peak reach capacity, not typical reach.
8. Order: complete all 8 directions on one leg before switching, same order every session (anterior, anteromedial, medial, posteromedial, posterior, posterolateral, lateral, anterolateral, relative to the stance leg), rest 1-2 minutes, then repeat on the other leg.

The full bilateral 8-direction test runs about 20-25 minutes per athlete, noticeably longer than the 3-direction Y-Balance Test variant, which trades directions for speed.

Turning Reach Distance Into a Number You Can Compare

The normalization itself is one line of arithmetic: %MAXD = (reach distance ÷ limb length) × 100, calculated separately for each of the 8 directions and each leg, both measurements in the same unit before dividing.

Worked example: Athlete A has a limb length of 92cm and posts a 66cm anterior reach, 66 ÷ 92 × 100 = 71.7%. Athlete B has a limb length of 79cm and reaches 60cm in the same direction, 60 ÷ 79 × 100 = 75.9%. Athlete A's raw reach is 6cm longer in absolute terms; Athlete B's normalized reach is over 4 points better. Reading only the tape-measure numbers would rank these two athletes backwards.

Composite score, the single number most clinics track across a season, is the average of the normalized reach percentages across all 8 directions, or whichever reduced set a given protocol uses. Asymmetry between limbs is calculated the same way, direction by direction: Asymmetry = |reach on limb A − reach on limb B|.

What the Research Actually Shows

Gribble and Hertel (2003), in Measurement in Physical Education and Exercise Science, tested exactly this problem: how much of the variance in raw SEBT reach distance comes from limb length rather than balance ability, and whether normalizing removes it. Raw reach distances correlated with limb length strongly enough in several directions that a meaningful share of the score's variance was anthropometric rather than neuromuscular; normalizing to the ASIS-to-medial-malleolus measure reduced that relationship substantially, leaving a score far more attributable to actual postural control. Their caveat matters in practice: different limb-length landmarks produce different normalized percentages from the identical reach, so a score from one clinic's method isn't automatically comparable to another's unless both used the same landmark.

Plisky, Rauh, Kaminski, and Underwood (2006), in the Journal of Orthopaedic & Sports Physical Therapy, followed 235 high school basketball players across a season using the normalized SEBT and tracked lower-extremity injuries. Anterior reach asymmetry of 4cm or more between limbs was associated with roughly 2.5 times greater odds of injury. A composite score below 94% of limb length was associated with roughly 6.5 times greater odds, but only in the female athletes in their sample; the same threshold did not reach significance in the males, a sex-specific finding later reviews have flagged as a reason not to apply the 94% cutoff uncritically outside basketball or across sexes without separate validation.

Normative Ranges and How to Read a Composite Score

These are typical ranges from published normative work on healthy trained athletes, not fixed cutoffs; an athlete's own baseline, retested under an identical protocol, is worth more than a comparison against any table. Two numbers matter more than where any single direction lands: composite score relative to the 94% threshold from Plisky's cohort, and anterior asymmetry relative to the 4cm threshold, both flagged above as carrying a measurable injury association in at least one athlete population. Neither is a pass-fail line on its own; treat a low composite or an asymmetry beyond 4cm as a reason to look closer, not a diagnosis.

Mistakes That Skew the Score

What to Do With a Low or Asymmetrical Score

A composite score under roughly 94% of limb length, or an anterior asymmetry past 4cm, is a training target rather than a reason to bench someone. A short anterior reach usually pairs with a matching limitation in ankle dorsiflexion range or single-leg quad control, and responds to weight-bearing dorsiflexion mobility work and controlled single-leg squat progressions before more reactive or plyometric loading gets added. A deficit concentrated in the posteromedial or posterolateral directions points more toward hip abductor and gluteal control, and responds better to lateral band work and single-leg Romanian deadlift progressions than to ankle-focused drills.

Retest on the same protocol, same limb-length landmark, ideally the same tester, every 4-6 weeks; SEBT composite scores move slowly enough that retesting sooner mostly captures day-to-day noise. When asymmetry rather than a low overall composite is the flag, unilateral work on the weaker limb closes that gap faster than adding volume to both sides equally.