The 2k Erg Benchmark Test: A Drag Factor and Pacing Protocol for Comparable Scores

A rower posts a 6:52 in September, a 7:04 in November on what everyone agreed was a solid week, then a 6:48 in January. No injury, nothing unusual in the program. The obvious read is that November was an off day and January a breakthrough. Pull the monitor logs and a different story turns up: drag factor sat at 128 in September, dropped to 118 in November because it was quicker to set up that morning, then landed back near 126 in January. Add a rower who went out four seconds under goal pace in September and paid for it in the third 500, against a controlled effort in January, and most of the implied fitness swing evaporates. None of that variance lived in the athlete. It lived in the test.

The 2k erg is rowing's most heavily used benchmark and one of the easiest to run inconsistently without anyone noticing, because it hides its two biggest sources of noise in a settings menu and a pacing decision rather than on the results screen. Lock the drag factor and prescribe the pacing shape before the piece starts, and the 2k stops being a mood-dependent number and starts tracking real physiological change. The protocol below covers both, plus the arithmetic for scoring how well the plan was actually executed, so a coach can tell a genuine improvement from a lucky split.

Why the Same Athlete Can Post Three Different 2k Times With No Fitness Change

Two variables drive most of the swing in 2k scores that has nothing to do with actual capacity: the flywheel's drag factor, and how the piece gets divided across four 500m efforts.

Drag factor governs how quickly the flywheel decelerates between strokes, which changes how much force a given stroke rate has to produce. A rower doesn't just feel a higher setting as generically harder; optimal stroke rate, force per stroke, and the fatigue curve across six-plus minutes all shift with it. Two maximal tests run 15-20 points apart on drag factor are, mechanically, two different tests, even though the monitor reports both in identical units.

Pacing shape moves the needle the other way. An athlete who goes out 4-5% faster than a sustainable average and fades through the third 500 is testing anaerobic capacity and pain tolerance more than aerobic power. One who paces conservatively and finishes even or negative is testing controlled aerobic output instead. Both can land on an identical final time, but neither tells a coach the same thing about what changed physiologically unless the pacing shape itself gets controlled and logged alongside the score.

Equipment and Locking In the Drag Factor

The test needs nothing beyond a standard ergometer with a monitor and a way to log four 500m splits. What determines whether two tests are comparable happens before the athlete sits down.

These ranges are starting points, not prescriptions, for a setting that lets the athlete execute the stroke well. What matters is that the number gets chosen once, recorded to the exact integer the monitor displays, and dialed back every time, not just the same damper lever, since two ergometers on an identical lever can read three or four points apart depending on flywheel wear. Treat a drift of more than 4-5 points between sessions as enough to void a direct comparison; recheck before trusting the split.

Step-by-Step Testing Protocol

1. Warm-up (15-18 min): Easy rowing for 8-10 minutes, then four progressive 250m efforts at roughly 65%, 75%, 85%, and 95% effort with 90 seconds of rest between each.
2. Lock the drag factor: Set it to the athlete's logged number and confirm the exact figure on the monitor, not just the damper lever.
3. Program the piece: 2000m with automatic 500m interval splits so all four splits log without hand-timing.
4. Set pacing targets before the piece: From the athlete's most recent controlled 2k or a critical power test, calculate a goal average split, then assign per-500 targets: 500m 1 at goal split minus 1.0-1.5 seconds; 500m 2 and 3 held even at goal split; 500m 4 at goal split minus 1.5-2.5 seconds for a moderate closing kick.
5. Execute and call splits: Maximal sustainable effort within those targets, calling each split as it posts so the athlete can adjust the next 500 rather than discovering a blown plan at the finish.
6. Record immediately: All four splits, average split, total time, and the drag factor displayed at the start.
7. Cool down for 8-10 minutes easy.

Total time, warm-up and cooldown included, runs about 40-45 minutes. Schedule this maximal test after at least one easy or rest day; a fatigued week suppresses both the time and the ability to hold the pacing plan.

Scoring Pacing Execution: The Split Variability Index

Total time alone answers whether a piece was fast. It does not answer whether it was executed the way it was planned, which is what actually tells a coach whether a time change reflects fitness or luck.

Average split (seconds per 500m) equals total time in seconds divided by 4.

Split Variability Index (SVI) = standard deviation of the four 500m splits ÷ average split × 100, expressed as a percentage.

Worked example: two pieces finish four seconds apart, close enough that a glance at the clock calls them equivalent. Piece A splits 101.9, 103.4, 103.8, and 101.3 seconds, a 6:50.4 total with a 102.6-second average. The standard deviation is about 1.03 seconds, an SVI near 1.0%, a controlled profile: fast first 500, steady middle two, fast finish. Piece B splits 96.0, 104.0, 106.0, and 108.0 seconds, a 6:54.0 total with a 103.5-second average. The standard deviation is about 4.55 seconds, an SVI near 4.4%, an explosive opening followed by a collapse that worsened each successive 500. They are four seconds apart but not close to the same test: Piece A taxed controlled aerobic power, Piece B taxed anaerobic capacity and pain tolerance from a sprint the athlete could not sustain. Total time alone would file these as interchangeable.

What the Research Actually Shows

Garland (2005), analyzing split data from finalists at World Championship and Olympic 2000m rowing events for the British Journal of Sports Medicine, found that elite competitors across boat classes did not pace evenly. The dominant pattern was a fast opening 500m, a relatively even and slower middle 1000m, and a faster closing 500m, the same fast-steady-steady-fast shape prescribed above, with roughly 3-5% typically separating a crew's fastest and slowest split. The finding is descriptive, not experimental: it documents what elite rowers do outdoors rather than proving that shape causes better performance, and it says nothing about ergometer-specific factors like drag factor. It does give a real reference point for patterned pacing variation at the highest level, the basis for treating an SVI in roughly the 1.5-4% range as compatible with strong execution, provided the shape is fast-steady-fast rather than a random blowup.

Schabort, Hawley, Hopkins, and Blum (1999), in the Journal of Sports Sciences, had well-trained oarsmen complete repeated maximal 2000m time trials under standardized conditions, same warm-up and equipment held constant, on separate days. Total time returned a coefficient of variation near 1%, remarkably reproducible for a maximal six-to-seven-minute effort. That figure is the strongest argument for locking drag factor and pacing: a test capable of 1% reliability under tight control can detect a genuine 1-2% fitness change, but only if the protocol behind that reliability actually gets replicated. The study's own limitation cuts against loose field use: the sample was small and homogeneous, well-trained oarsmen under lab conditions, so 1% sits closer to a best-case ceiling than a guarantee, and reliability in a program that lets drag factor and pacing drift is very likely worse.

Reading the Split Variability Index

These bands pair Garland's (2005) elite pacing-variation range with the reality that club and collegiate athletes execute a plan less precisely than international finalists. Use them to judge execution quality, not as a score to chase; a higher SVI on a well-shaped fast-steady-fast profile is not the same failure as an identical SVI from a random blowup.

General 2k time bands vary by experience and body mass more than most benchmarks: elite scores commonly sit under roughly 6:00 for heavyweight men and 6:50 for women, well-trained club athletes land near 6:15-6:45 for men and 7:10-7:45 for women, and developing rowers run well above 7:00 and 7:45 respectively. None of these substitute for an athlete's own history; a 6:58 replacing a 7:15 under an identical drag factor and pacing plan beats any cross-athlete comparison.

Mistakes That Wreck Comparability

What a Time Change Actually Means

A time drop with a stable, patterned SVI is the cleanest signal a program gets: capacity likely moved, execution stayed constant, so the change is believable. A total time that holds steady while SVI tightens from 4% to 1.5% is a real result too, just a different one; the engine may not have changed much, but the ability to execute a race plan under fatigue clearly has.

Retest on a 6-8 week cycle rather than after every block. A single maximal 2k costs real recovery time, and weekly retesting mostly captures day-to-day readiness and pacing variance rather than a physiological shift worth reacting to. Hold drag factor and the pacing targets constant across that cycle; the only thing that should change test to test is what the athlete's body does with them.