Dressage Scoring, Decoded: From the Judge's Booth to Your Final Percentage

You ride down centerline, salute twice, and some time later a number appears on the board: 68.077%. That number follows you. It decides ribbons, year-end awards, and whether a ride counts toward a qualification. Yet a surprising number of riders — including experienced ones — have never traced the arithmetic from the judge's first mark to that final percentage.

This is that trace, start to finish. Nothing here is opinion; it's how the scoresheet is built and how the math runs.

The scoresheet: movements and directives

Every dressage test is a numbered sequence of movements — not "exercises," movements. The 2023 USEF Training Level Test 1, for example, has thirteen of them, from the entry halt to the final halt at X.

Each movement on the sheet comes with a directive: the printed criteria the judge is evaluating for that movement. For the entry at Training Level, the directive reads "regularity and quality of trot; willing, calm transitions; straightness; attentiveness; immobility (min. 3 seconds)." The directive is why two visually similar rides can score differently — the judge isn't scoring what the movement looked like in general, but how it answered those specific questions. (If any of the vocabulary here is new, the glossary covers it.)

The judge marks each movement on a 0–10 scale, and half points are allowed. The scale has official definitions:

A 6 is not a bad mark; it is "satisfactory." A 7 is "fairly good." The emotional distance between a 6.5 and a 7 is enormous for the rider and half a point for the math. Keep both facts in mind.

Coefficients: the test telling you what it values

Certain movements carry a coefficient of 2 — their mark counts double. On the 2023 Training Level Test 1 sheet, six of the thirteen movements are coefficient movements, including the 20-meter trot circles and the free walk.

Coefficients are not decoration. They mark the movements the test designers consider most diagnostic at that level — the ones that best expose whether the basics are actually installed. The free walk carries a coefficient at the lower levels because it is nearly impossible to fake: a horse that is tense through the back or distrustful of the contact will show it the moment the reins are offered forward. A double-weighted 5 on the free walk costs you as much as two ordinary 5s, which is precisely the point.

Practical consequence: if you're managing a test strategically, the coefficient movements are where preparation buys the most percentage.

Collective marks: the summary at the bottom

After the last movement, the judge awards collective marks — holistic scores for the ride as a whole. On the current USEF national tests (the 2023 test cycle, effective December 1, 2022 through November 30, 2026), the collectives are:

Note where the double coefficients sit: Impulsion and Submission. The test rewards the quality of the training, not just the horse's raw gaits — a horse with modest natural movement but honest suppleness and acceptance can out-collect a spectacular mover who is braced against the hand.

FEI tests took a different path: they carry a single collective mark, and under the FEI rule changes effective January 1, 2026 it is called "Harmony" (previously "General Impression"), with criteria emphasizing cooperation, lightness, and sensitivity of the aids.

Errors: course versus test

Two kinds of mistakes carry deductions, and they are not the same thing.

An error of course is going the wrong way: turning at the wrong letter, skipping a movement, picking up the wrong sequence. The judge rings the bell, you're shown where to restart, and the deduction lands on your final score.

An error of test is a rule-of-the-test violation while on the correct pattern — rising when the test requires sitting, or failing to take the reins in one hand at the salute where required.

On current USEF national scoresheets, the deduction scale is printed at the bottom: first error, 2 points; second error, 4 points; third error, elimination. Those are raw points off your total before the percentage is calculated — on a 260-point test, a first error costs a bit under 0.8%.

FEI tests are harsher. Under FEI rules for senior tests, the first error of course deducts 2 percentage points from the final score, and a second error means elimination. (Reduced scales apply in junior, pony, children's, and young-horse divisions.) Deduction rules get adjusted over time, so before a championship ride, verify the current values in the USEF rulebook or FEI Dressage Rules rather than trusting folklore — or an article.

The panel: who sits where, and why it matters

At most local shows, one judge sits at C. At championships and CDIs, a Ground Jury of three, five, or seven judges surrounds the arena, and the judge at C is the Ground Jury president.

The letters are positions around the arena: C at the center of the short side opposite the entry, E and B at the midpoints of the long sides, H and M flanking C, and F and K added at the entry end for seven-judge panels.

Each position sees a genuinely different test. The judge at C looks straight up the centerline and sees straightness, geometry, and the quality of transitions on the line. A judge at E or B sees the horse in profile: the frame, the engagement of the hindquarters, whether the poll is the highest point. A halt that looks square from C may be camped-out from B. This is why panel marks legitimately differ — the judges are not disagreeing about the same picture; they are scoring different pictures.

Each judge produces their own complete scoresheet and their own percentage. The final score is the average of the judges' percentages. More judges means single-position blind spots get averaged out — which is one reason championship scores are considered more robust than one-judge scores, and why how DPI weighs results accounts for the context a score was earned in.

From marks to percentage

The arithmetic is plain:

1. Multiply each movement mark by its coefficient and sum them.
2. Add the collective marks (times their coefficients).
3. Subtract error deductions.
4. Divide by the maximum possible points for the test, multiply by 100.

Training Level Test 1 has a maximum of 260 points. Score 177 of them and you get 177 ÷ 260 = 68.077%. That's the whole trick — the percentage that drives national rankings and every rider's season goals is a division problem, typically carried to three decimal places, which is why scoreboards read 68.077 rather than 68.1.

Ties happen. Under the USEF rulebook, tied totals are separated by the higher score for collective marks — and if the collectives are also equal, the tie stands.

One warning about comparing percentages: a 68% at Training Level and a 68% at Prix St. Georges are the same arithmetic but not the same achievement. Difficulty rises through the levels — Training, First, Second, Third, Fourth, then the FEI levels: Prix St. Georges and Intermediate I (the Small Tour), Intermediate II, and Grand Prix, with Grand Prix, Grand Prix Special, and the Grand Prix Freestyle (Kür) forming the Big Tour. This is also why qualification standards — the MERs (Minimum Eligibility Requirements) and minimum-score thresholds published by USDF and USEF — are set per level and division, and why the specific numbers should always be checked against the current official documents for your year.

Why the system is built this way

None of this structure is arbitrary. The whole scoresheet is an instrument for measuring the Training Scale — rhythm, suppleness, contact, impulsion, straightness, collection. The movement marks sample it locally: this circle, this transition, this lengthening. The coefficients concentrate weight on the movements that expose the basics. The collectives measure it globally, across the entire ride — with the double weight, on the national tests, sitting exactly on impulsion and submission, the middle of the scale where correct training lives or dies.

Read your next scoresheet with that in mind and it stops being a report card and becomes a diagnostic: the marks tell you where on the Training Scale the holes are, the directives tell you what the judge was looking for, and the coefficients tell you what to fix first.

If you want to see how those percentages accumulate across a season — yours or any rider's — DPI's index and qualification planner track them from the source sheets forward.