What Your 2K Actually Measures, and What It Does Not

A 2K is a number on a watch, and two questions sit behind every one of them: what does this actually measure, and what have I been told it means that it does not mean?

Here is the short version. A 2K measures the power you can hold for about seven minutes, and that power is a blend of aerobic and anaerobic capacity rather than one clean physiological quality. What it does not measure is boat speed, because an ergometer result and a boat result are different measurements of different things: in the one study that put the same rowers on both, the two shared 28.9 per cent of their variance ([6] Nevill 2010). Neither half of that answer is an excuse for a disappointing time. It says the number is worth having, and it says what to do with it.

The context

Start with what the test is, because a lot of the confusion comes from treating it as a smaller version of a race. It is a self-paced maximal effort of 2,000 metres on an air-braked machine, and the whole thing lasts about seven minutes. It is also the default benchmark of the sport: in a systematic review of 34 studies and 909 rowers, the 2,000 m test was the protocol in 79 per cent of them ([2] Borges 2025).

It is also a very repeatable measurement, which is why it has survived. Eight well-trained rowers rowed a 2,000 m trial on a Concept II on three occasions three days apart: the coefficient of variation for mean power was 2.0 per cent, the retest correlation 0.96, and the variability of the finish time itself about one third as variable as the power figure ([1] Schabort 1999). That is roughly three seconds either side of your true time on a seven-minute piece. The protocol side of that is covered properly in our guide to what makes a test repeatable.

So the trouble with a 2K is not that the machine is unreliable, or that one bad morning poisons the number. It is quieter and more interesting than that: the test is a stable measurement of a capacity, and a lot of the things people want it to mean are not capacities at all.

The core idea

What it measures: a composite, not a single quality.

The best physiological work on this question comes from World Championship finalists. The strongest correlates of 2,000 m ergometer speed were power at maximal oxygen uptake, maximal power and maximal force, all at r = 0.95, with maximal oxygen uptake at r = 0.88 and oxygen uptake at the blood lactate threshold at r = 0.87. Together those variables explained 98 per cent of the variance, and when the model was checked against 18 more elite rowers the limits of agreement on time ran from -1.5 to 6.9 seconds ([3] Ingham 2002).

Read that list again. Maximal oxygen uptake and maximal power sit on the same list. One is the ceiling of your aerobic engine and the other is the top of your anaerobic one. A 2K is not a VO2max test with a bit of suffering at the end. It is a blend of both, and the blend shifts as you get fitter.

The aerobic half has the larger body of evidence behind it. VO2max and peak power output are the strongest and most consistent correlates of 2,000 m performance, at r = 0.83 to 0.99 across those same 34 studies and 909 rowers ([2] Borges 2025). The lactate-threshold literature agrees: across 24 studies and 797 athletes, the power or velocity at a blood lactate concentration of 4 mmol/l was the most commonly used test, with correlation coefficients from 0.53 to 0.96, which suggests 28 to 92 per cent of the variance in 2,000 m performance could be explained by that one number ([4] Kilbey 2025). Six studies that rated well on risk of bias found very strong correlations above 0.85.

The anaerobic half is not a footnote, and this is where the received wisdom goes wrong. In 40 NCAA Division I women, the 7-stroke peak power, 20-second and 60-second maximal effort tests were administered at four points across an academic year, with the 2K rowed three weeks after the last of them. The 60-second test was the best single predictor of 2K time (R² = 0.611), the 20-second test second (R² = 0.536), and the 7-stroke peak power test the weakest of the three (R² = 0.4171). Power in the 60-second test correlated negatively with 2K time at r = -0.78 and power in the 20-second test at r = -0.73 ([5] DeLeo 2026). A one-minute all-out effort predicted a seven-minute time better than a peak-power test did. That says the difference between rowers over 2,000 m is substantially about what they can still do when they are tired.

What it does not measure: boat speed.

Here is the study that should change how you read a 2K. At a British training camp, 49 junior elite male rowers completed both a 2,000 m Concept II model C test and a 2,000 m single-scull test on the water. Water speed was 3.66 metres per second against 4.96 on the erg, and the two performances shared only 28.9 per cent of their variance ([6] Nevill 2010).

Body mass makes the point cleanly. It correlated with erg performance at r = 0.68 and with water performance at r = 0.039, and once erg speed was accounted for, mass made a negative contribution to water speed ([6] Nevill 2010). The erg does not charge you for your mass the way the water does, so a repeatable erg test is a repeatable measurement of something that is not quite boat speed.

And when the real thing does get measured, the parts of a race that people most want a 2K to predict turn out to be the parts the evidence cannot pin down. Across 45 instrumented on-water 2,000 m races, mean race power and mean stroke rate predicted race velocity, with effects ranging from small to extremely large and mostly decisively substantial. But technical efficiency and race conditions had effects that ranged from trivial to extremely large and were generally unclear ([7] Holt 2022). That is a gap in the evidence, not a measured zero, and it is not a licence to read a 2K as a race result.

The how-to

1. Turn the time into a pace. This is the most useful thing most rowers can do with the number, and the manufacturer's own protocol is the shortest sensible version of it. Row the 2K, note your average 500 m split, then repeat the piece about a week later starting at that pace and building in the second half. After three or four of those pieces you are homing in on your target 2K race pace. The same page advises you not to do more than one of these test 2Ks a week, and to be rested before each one ([9] Concept2 2024).

2. Compare it with your own last one, not with somebody else's. The erg-versus-water gap above is the reason. Two rowers of different masses are being measured on a machine that flatters the heavier one, so their raw times are not the same measurement. The club-level evidence points the same way. Across 200 of the best rowers from the seven largest Hungarian rowing clubs, with body composition measured by bioelectrical impedance and skinfolds rather than estimated, more skeletal muscle mass went with a faster 2,000 m (r = -0.39 against rowing time) and more maximal force with a much faster one (r = -0.79 in women and -0.90 in men), while more body mass and more body fat went with slower times (r = 0.24 to 0.33) ([8] Podstawski 2023). That paper is a study of confounding, not a ranking of rowers: its subject is that these characteristics travel together, so a raw time is never a clean read of any one of them.

3. Know when another test earns its place. You do not need a fresh benchmark every month. In 34 elite under-23 rowers, a linear critical speed model built from a 500 m and a 2,000 m time trial predicted 1,500 m time with a mean absolute error of under a second, a coefficient of variation of 0.5 per cent, and a correlation between actual and predicted times of R = .99, which let the coaches track progression without adding a third test day ([10] Astridge 2026). A pair of shorter pieces can tell you whether you are moving without another full effort.

4. If you want to watch the anaerobic end, test it directly. The 60-second maximal effort is the one to copy: the authors of that study describe it as a time-efficient way to monitor rowing performance and confirm adaptations to the training programme ([5] DeLeo 2026). It is a minute of work, and it is the part of a 2K that the finish time hides.

The trade-offs

Three honest limits.

The predictor literature is large and not poolable. Twenty-four studies and 797 athletes sounds decisive, and the review itself says otherwise: no meta-analysis was performed, because the studies vary in quality and mix parameters, protocols, testing methods and performance metrics, which makes comparison between them difficult ([4] Kilbey 2025). A correlation of 0.53 in one study and 0.96 in another is not just noise in the underlying physiology. It is partly a difference in how the threshold was defined. So "lactate threshold predicts your 2K" is true in a weaker sense than it sounds.

The samples are small. Eight well-trained rowers for the reliability work ([1] Schabort 1999), 49 junior elite men for the erg-versus-water work ([6] Nevill 2010), 18 elite rowers for the physiological model ([3] Ingham 2002), 40 women for the maximal-effort work ([5] DeLeo 2026), 200 club rowers for the body-composition work ([8] Podstawski 2023). These are careful studies. None of them is the last word.

And much of the pooled evidence leans male. The 34-study review behind those VO2max and peak-power correlations was 657 men and 252 women, 72 per cent male ([2] Borges 2025). For a female club rower reading those numbers, they describe a population she is not in.

What to do tomorrow

1. Write down the number, the machine, the damper and the conditions. Next time, do the same. A 2K is only a trend if the conditions are held still, and the protocol detail is in what makes a test repeatable.
2. Work out your 500 m split, then row the piece again a week later at that pace. Three or four of those and you have a race pace, which is worth more than the time you started with.
3. Stop reading other people's numbers. Your 2K is a comparison with your own 2K. Somebody else's time is a different measurement of a different body.
4. If the last minute is where it falls apart, train and test that minute directly rather than trying to diagnose it from the finish of a seven-minute piece.