How Judging and Scoring Works

Dressage is scored by trained judges marking every movement of a test on a 0–10 scale, and the final result is a percentage of the maximum possible marks. Unlike sports decided by time or knocked poles, dressage assesses the quality of performance, which is why the judging system is built around standardised criteria, multi-judge panels, and averaging. Understanding it turns a score sheet from a verdict into usable information.

The 0–10 scale

Every movement receives a mark from 0 to 10, with half-marks allowed; half-marks have been permitted in all international tests since 2011. The meaning of each mark is standardised across FEI competition:

A 6 means the movement was shown adequately with visible faults; a 7 is correct with some positive quality; an 8 is genuinely good. Marks of 9 are uncommon even in elite sport, and a 10 on a movement is rare enough to make news. The scale is intentionally wide so that judges can differentiate, and when the marks of a five- or seven-judge panel are combined, a clear picture emerges.

Coefficients: double-weighted movements

Selected movements carry a coefficient of 2, doubling their weight in the total. Test writers apply coefficients to the movements that are most telling for the level: at the lower levels often the free walk or a key transition; at the top of the sport piaffe, passage and their transitions. A 7 on a coefficient movement contributes 14 points, so a rider’s percentage is disproportionately shaped by how the coefficient movements go, a weighting visible on any Grand Prix score sheet.

Collective marks

After the movements, judges award collective marks for the overall picture rather than any single exercise.

US national (USEF/USDF) sheets award five: gaits (freedom and regularity of the paces), impulsion (energy, elasticity, engagement), submission (willingness, responsiveness, lightness of the connection), the rider’s seat and position, and the correct and effective use of the aids.

FEI tests use a single collective mark, renamed “Harmony” from 2026 (replacing “General Impression”), defined as covering harmony and cooperation, lightness, the effectiveness and sensitivity of the aids, and adherence to the training scale.

Beside every movement, the test sheet prints a directives column stating the criteria by which that movement is judged: what should be achieved and how the judge weighs it. The directives are the published bridge between the abstract 0–10 scale and the specific exercise, and the mechanism through which the test writers steer the sport’s standards. Tests are drafted and revised on federation cycles, so the directives change with them.

In both systems the collectives carry coefficients, so overall quality weighs heavily: a test with a few good movements inside a tense, resistant picture scores below a consistent, harmonious one.

Judge panels and positions

At national shows a single judge at C may officiate. At international competitions, a panel of five judges sits around the arena at C, E, B, M and H; championships and the Olympic Games use seven, adding K and F. (The positions are named for the nearest letter: the judges at M and H sit along the short side either side of C, and those at K and F near the A end.)

Each position offers a different view. The judge at B sees the half-pass from the side; the judge at C sees straightness on the centreline; the judges near the corners see the transitions the others catch at an angle. That is the point of the panel: a fault invisible from one seat is visible from another, and the average is fairer than any single view.

How the percentage is calculated

Each judge marks every movement independently. For each judge, the marks (with coefficients and collectives) are totalled and divided by the maximum possible, giving that judge’s percentage; the judges’ percentages are then averaged for the final score.

Final score (%) = points earned ÷ maximum possible × 100, averaged across the panel.

Because each judge’s percentage is computed before the averaging, a single divergent judge moves a five-judge result by only a fifth of the divergence, which is the averaging’s built-in protection. Panel members usually land within a point or two of percentage of each other. A larger spread attracts attention, and at major events supervisory panels and analytics now track exactly that.

What the percentages mean

• Below 60%: the combination is not yet established at the level. Common at a first attempt and simply a signpost for training.
• 60–65%: solid; the basics are confirmed and moving up becomes thinkable. Many qualification thresholds sit in this band.
• 65–70%: competitive at the level.
• 70–75%: a good score reflecting consistent quality; typically places well.
• 75–80%: excellent; few combinations score here consistently.
• 80%+: elite, and rare even internationally.

For perspective, the Grand Prix Freestyle world record is Charlotte Dujardin’s 94.3% on Valegro, set at Olympia in 2014, a mark that has stood for over a decade.

National variations in judging practice

The 0–10 scale and the percentage are universal; the practice around them is federation business, and the differences are codified. Under the Dutch regulations, a single judge may score tests up to and including ZZ-Zwaar, two judges are required at ZZ-Zwaar and above, and a judge may assess at most 32 combinations in a competition day, with mandatory breaks built into the schedule. Germany’s LPO 2024 admits a mixed judging format for rider tests at classes L to S, in which one judge marks the individual movements technically while another judges the overall impression — a system adopted from the FEI’s Children tests. National rules also govern the mundane machinery: whether a caller may read the test (permitted at most national levels, never in FEI tests), and how scores are registered, which in Germany and the Netherlands feeds the central records that make competition claims verifiable.

The bands are not merely descriptive; the federations build them into their rules. The Dutch winstpunten system awards one promotion point for a score of 60% up to 65%, two from 65% and three from 70%, and British Dressage scales its points to the percentage achieved from a 60% floor, so the 60/65/70 boundaries carry institutional weight far beyond the day’s placings. A judge’s marks are, in this sense, also entries in the record systems that make a horse’s career verifiable.

Why judges disagree — and why that is not a scandal

Dressage is often called subjective, but the disagreement between judges has mostly mundane causes:

• Viewing angle. A slightly crooked half-pass genuinely looks straighter from some seats than others.
• Weighing a fault. Two judges can agree on what they saw and differ on whether it costs half a point or a full one.
• The moment. A movement lasts seconds; a judge’s attention is human.

The FEI Dressage Judging Manual standardises the criteria, judges train and calibrate continuously, and the panel average absorbs individual variation. The residual disagreement is the honest price of judging quality rather than counting faults.

Judge levels and credentials

Judges are certified and graded by their national federation and the FEI. National systems license judges per level (Britain’s list system, the US ‘r’/‘R’/‘S’ progression, and so on). FEI judges are graded by stars up to 5*; only FEI 5* judges may officiate at the Olympic Games or World Championships. Reaching that level takes decades of riding, judging and continuing education, with promotion between grades depending on assessed officiating at FEI events and on regular refresher courses.

Errors, grievances and what can be challenged

Clear administrative errors (a movement left unmarked, a wrongly computed total) can be corrected through the show’s procedures. A mark itself is a judge’s discretion and cannot be challenged because the rider disagrees; a 6 where the rider hoped for a 7 is not a grievance.

Judges can, however, eliminate a combination for cause: dangerous riding, signs of blood or lameness, illegal equipment, or accumulated errors of course. A merely disappointing test is never grounds for elimination.

Using the score sheet

The percentage measures competitiveness; the movement marks and the judge’s written remarks identify what to train. “Needs more engagement in the hindquarters” or “lost rhythm in the transition” is a training plan in miniature. A sheet full of 6s and 7s with scattered 5s reads most productively from the 5s upward: they are the syllabus. And because the marks trace back to the training scale, the remarks usually point below the movement (at the rhythm, suppleness or straightness gap that caused it), rather than at the movement itself.

Judging by the horse’s qualities rather than a test’s movements also exists as its own format: in young horse classes, judges mark the paces, submission and prospects of the horse directly and explain their marks publicly.

Sources

• Fédération Équestre Internationale — FEI Dressage Rules, 26th edition (updates effective 1 January 2026), 2026. https://inside.fei.org/fei/disc/dressage/rules
• Fédération Équestre Internationale — Dressage Rules (judging panels and positions), 2022 revision annex. https://inside.fei.org/system/files/16.2_GA%202022_Annex%2016.2_DRESSAGE%20FULL%20RULES_0.pdf
• Koninklijke Nederlandse Hippische Sportfederatie (KNHS) — Disciplinereglement Dressuur, versie 2026 (judging provisions), 2026. https://www.knhs.nl/media/jmid00ij/disciplinereglement-dressuur-2026.pdf
• Deutsche Reiterliche Vereinigung (FN) — LPO 2024 (mixed judging format), 2024. https://www.pferd-aktuell.de/turniersport/regelwerke-und-merkblaetter/leistungs-pruefungs-ordnung/neue-lpo-2024
• United States Dressage Federation — Purpose of Tests and test materials (collective marks), 2023. https://www.usdf.org/EduDocs/Training/purposeoftests.pdf
• Wikipedia — Dressage (half-marks since 2011), 2026. https://en.wikipedia.org/wiki/Dressage