In this article, you’ll learn all about rowing power testing: why power tests offer important information, basic testing principles, rowing power testing protocols, and how to collect, analyze and use this information in your training. We will discuss important physiology knowledge for power testing, best practices for erg testing, the different protocols of counting strokes or seconds, apps that you can use to collect power testing data, and how often to test peak power to inform your training.
Key Points: Rowing power testing tells us if the rower is improving in maximum power output (duh), the effect of strength training on rowing performance, and offers a way to estimate race pace. My favorite power tests are the 7-stroke max or 10-second test to stay within the phosphagen energy system, responsible for producing very high outputs over short durations. Whichever power test you choose, testing protocols need to be standardized and kept consistent in order to accurately reflect results as changes from training. Use full slide, good technique, recorded drag factor, similar motivation and testing environment, and same rest between individual attempts.
Table of Contents
- Why Test Rowing Power?
- Rowing Power Testing Basics
- Testing Protocols
- Data Collection
- Using the Data
- How Often to Test
Why Test Rowing Power?
Rowing researchers have demonstrated for many years that peak power on an erg is correlated to 2km performance on an erg. Intuitively, this makes sense: Faster rowers are faster rowers. More powerful rowers can produce more watts, which allows them to row faster if they are also aerobically fit. If you are aerobically fit but cannot produce high power, you will not row as fast as if you could produce more power. Rowers tend to be more aerobically fit than powerful due to the high amount of time in rowing spent training aerobic fitness, so peak power can be a more distinguishing variable that offers some insight as to the rower’s potential for 2km performance.
Peak power tests also give us information for how high-intensity training is working and if the rower is improving, without having to frequently test the goal racing distance. If we know that peak power is related to 2km erg performance and a rower improves peak power, we can assume that the rower has potential to improve 2km performance as well. Peak power sets the ceiling, aerobic endurance builds the floor.
Some coaches have systems of extrapolating 2km performance from a peak power test. For example, Rowing Canada strength coach Ed McNeely says that a rower’s 2km erg average pace should be approximately 55% of their 10-stroke peak power performance. Danish physiologist Dr. Kurt Jensen’s testing model is another form of correlative testing. Dr. Jensen’s model states that a 10-second peak power test should be 173% of the rower’s 2km erg average pace. We will come back to these later on.
We will also see the most influence from strength training on power performance on the erg. I prefer to use a peak power test on the erg rather than test one-repetition maximums to see improvement from strength training. One-rep max testing requires a lot of considerations in order to get right. The rower needs to be very technically proficient at the lift at maximum output, comfortable and experienced in the testing environment to be at their best performance, and able to adequately rest before the test to be at their full strength, as well as after the test to recover from the effort. A peak power test on the erg requires less concern for these other variables. Performance on the erg is much more specific to rowing performance than a deadlift or squat exercise. A rower could improve their squat or deadlift and not necessarily improve on the erg or on the water. It’s more likely that improved performance on the erg transfers to improved performance on the water, so I’d rather see an improvement in peak power on the erg than more pounds on a squat or deadlift max.
Rowing Power Testing Basics
About Power Physiology
There are three major energy systems in human physiology. Each one has multiple names and sub-categories to add detail. This can often be confusing for rowers and coaches new to exercise science, so I’ll provide some brief definitions and basic information here.
Key Points: All three energy systems can act simultaneously, but the degree to which each is emphasized depends on the intensity and duration of the activity. The phosphagen energy system is the one we care about for power testing, producing the greatest power over the shortest duration. For the best rowing power testing, we need to maximize intensity, restrict duration to approximately ten seconds at most, and rest for 1-4 minutes between trials. You can skip ahead to “Erg Testing Basics” if this is enough physiology for you, or read ahead for a brief explanation of energy systems.
The phosphagen system produces the greatest amount of energy for the shortest amount of time by doing breaking down adenosine triphosphate (ATP). Due to the interaction of ATP and creatine phosphate, the phosphagen energy system is often also referred to as the ATP-CP or ATP-PC system. After approximately 10 seconds of very high intensity output, the intensity falls as the duration lengthens and we shift our emphasis in energy production to the glycolytic system. The glycolytic system breaks down sugar (glucose) to produce ATP. The phosphagen and glycolytic systems are often referred to together as the anaerobic energy systems, because they do not require oxygen to produce ATP. After approximately two minutes of high intensity output, the intensity falls again as the duration lengthens and we shift energy production emphasis to the oxidative system. The oxidative system is also known as the aerobic system because the body uses oxygen to produce ATP. This takes longer, but also lasts longer, so the aerobic system powers low-to-moderate intensity outputs longer than five minutes in duration.
It’s important to note that all energy systems act simultaneously, with one emphasized over another based on the output demands in intensity and duration. In “Developing Endurance,” exercise physiologist Randy Wilber refers to an analogy of energy system function as a symphony orchestra: “The orchestra includes several instrument groups, and each group plays softly, moderately, or loudly depending on the musical score.”
Read More: Is a 2k Aerobic or Anaerobic?
While the aerobic energy system provides the most amount of energy in a 2k race, peak power testing means short duration performance with the highest possible exertion and full recovery between efforts to stay within the phosphagen energy system. If we prolong the duration beyond the approximately 10-second capacity of the phosphagen system, or if we reduce rest time and do not allow for full recovery of the phosphagen system, the intensity of output will fall and energy production shifts into other systems. Highest possible exertion also means simplest technical environment possible. The more complexity required of the performance, the more the performance will be limited by technique or strategy rather than pure physiological output.
Erg Testing Basics
The rowing ergometer is an excellent device for testing physiological power. It’s a full-body, concentric-focused movement with fairly simple technical parameters. There is one exception to its excellence, which we’ll discuss in the data collection section later on. Overall, the erg lends itself very well to power testing protocols as long as we abide by our basic physiological principles to stay within the appropriate energy system and design our tests to be valid and reliable.
Validity and reliability are two principles of any form of testing. Validity means testing what you think you’re testing. For an extreme example, if I told you the length of something based on my measurement of it using a weight scale, you would (I hope) tell me that this is not a valid form of measurement. If we intend to test peak power and our test duration is outside the range of the phosphagen energy system, this is not a valid test of peak power. Reliability means producing consistent results from a test method. If I told you three different measurements for the length of a single object, you would tell me that my measurements are not reliable. If a rower produces three very different peak power scores in three attempts at our peak power test, we have a reliability problem.
Validity is the easier problem to address. For the most valid power test, we need to be on the erg performing at maximum effort for approximately ten seconds. Don’t use a bike, don’t test for over 10 seconds, don’t vary your effort levels, or you’re testing something other than rowing power.
Reliability is more challenging because it requires a high amount of communication, attention, and consistency. The fewer variables that are involved, the easier it is to produce reliable results, but it can be hard to remove variables in the real world. How much did you sleep the night before the test? What did you eat the morning of the test? How did you warm-up before the test? What music did you listen to during the test? How long did you rest between test attempts? How important is this test to you? Were you aware of your score or your teammates’ score during testing? Did you row full slide, with full effort on all strokes, at the same stroke rate, with the same drag factor, using the same technique for each test? All of these factors and more can influence performance. We may not need to control every single one of them, but the more that we can keep consistent, the more informative our results will be.
In summary, we want two major things from our rowing power testing protocols. Validity means that we’re truly testing peak rowing power by maximizing effort in a simple technical environment for approximately ten seconds. Reliability means that we’re able to produce consistent results from one test to the next by controlling enough variables in our testing environment. If we achieve validity and reliability, changes in test results will reflect actual physiological changes and this will tell us if our training is working.
Note that “performance” exists outside of “testing.” When we talk about “race performance,” we want all tools at our disposal to increase output. Eat your special pre-race meal, wear your lucky underpants, blast your favorite music, etc. Performance is the ultimate goal. Testing exists to gather information on training efficacy, so leave your lucky underpants in the drawer and keep all those motivational variables minimized.
What about power testing on dynamic ergs?
There’s much less rowing research on dynamic ergs overall, and none that I’ve been able to find on peak power testing. The two main challenges to dynamic erg power testing are that the technique requirements are greater and the per-stroke load is lower than on a static erg. Peak power testing must be done in approximately 10 seconds, against a heavy enough load to force the rower to exert maximum effort for the entire duration, and in a simple enough environment so that effort and output is a bigger factor than technique. It is possible to do this on a dynamic erg, but it’s more challenging than on a static erg. Dynamic ergs are also less available to rowers and researchers, so lower availability and higher challenges means most of our rowing power testing research and practice comes from static ergs.
Rowing Power Testing Protocols
There are a few main methods of rowing power testing that researchers use for studying peak power. Coaches and rowers can use protocols very close to these research methods for the clearest results and most reliable information.
The two major variables in rowing power testing protocols are drag factor and duration of test. All rowing researchers use a dead start (ie. no momentum of the rower or flywheel), full slide through the entire test (ie. no starting sequence), unrestricted stroke rate, and “normal” stroke technique to achieve consistent results. Technique is likely to look different at 45+ strokes per minute than at normal training or racing pace, but I’m happy as long as things look generally similar: no crazy over/under-compression, early arm-bend, or deep laybacks just to crank some extra watts.
Some researchers do power testing at whatever drag factor the rower usually uses in training, while others increase the drag factor significantly just for power testing. A higher drag factor will generally result in higher peak power numbers produced later in the test. There are good rationales for both approaches, so I recommend picking whichever one you prefer and sticking with it. The important thing is consistency and testing in the same environment each time.
There are four main power testing durations in rowing research: 7-stroke max, 10-stroke max, 10-second test, and 15-second test. Below are details about each protocol.
The 10-stroke max test is the rowing power test that most rowers and coaches are familiar with. While 10 is a nice round number, the duration of testing is usually between 12-15 seconds of total output. This is outside the range of the phosphagen energy system, so it isn’t a “true” test of pure peak power. If you’re just looking at single-highest stroke, rowers will likely achieve this in approximately 10 seconds of the greatest output from the phosphagen energy system anyway. Few rowers (not using a starting sequence) will hit their highest power strokes in the 12-15-second range due to the increased momentum of the flywheel and the longer duration of effort.
Seven is an unusual number in training, but the seven-stroke max test (7SM) offers the truest test of rowing peak power. This test is completed in under 10 seconds to stay within the range of the phosphagen energy system. Rowing researchers Nugent et al. (2019) evaluated the 7SM and found that it is a valid and reliable assessment for rowing power testing, at least in male high-performance rowers when performed using their protocol: dead start, full slide, unrestricted stroke rate, drag factor 140, four minutes of rest between trials, three trials in at least the first testing session, and no music, verbal encouragement, or knowledge of results during testing.
10 strokes takes approximately 15 seconds, so the 10-stroke max and 15-second test are essentially interchangeable. The same goes for the 7-stroke max and the 10-second test. If you wish to count strokes, use the 7-stroke or 10-stroke max tests. If you wish to count seconds, use the 10-second test or 15-second test. The 7-stroke or 10-second test are my preferences to stay within the phosphagen energy system and achieve the best validity and reliability, but I won’t lose sleep over a rower or coach who wants to use a 10-stroke or 15-second test as long as they are doing a good job with reliability to test in consistent conditions.
The main takeaway for testing protocols is to have a plan and stick to the plan to keep your tests as consistent as possible. Consistent testing environments give us better information as to how our training is working to produce physical change. For rowing power testing, use similar test conditions, similar warm-up protocol, similar motivational settings (ie. music and encouragement), full slide the whole way with no starting sequence, and at least a 1-to-5 work-to-rest ratio between trials (ie. 50 seconds of total rest for a 10-second test) or ideally the whole four minutes to allow for full physiological recovery and no influence from fatigue.
Rowing Power Testing Data Collection
Once you’ve selected your rowing power testing protocol, the next step is how to get the data from your erg to your logbook or spreadsheet.
I mentioned in the testing basics section that there was one major exception to the excellence of the erg as a means of power testing. Here we have it: The Concept2 performance monitor does not record data under 20 seconds, nor does it collect individual stroke data. This greatly limits our ability to test peak power in practical settings and in rowing research. You can row for 20 seconds and the Concept2 performance monitor will record your average power, peak power, total distance, etc. However, 20 seconds is far too long of a duration to test peak power, so using a 20-second test will not give you useful data for rowing power testing.
The “manual” method of peak power testing is to video record the erg screen during <20-second testing and then watch the video back to record stroke data. This is what rowing researchers, rowers and coaches, and I have done for years to record, track, and analyze peak power testing data. It is very time-consuming, requires at least one coach or assistant to record the screen with a camera or phone, and introduces the possibility of errors when collecting data for multiple rowers or transcribing results by hand.
Fortunately, as of recently in 2021, we have some great solutions via third-party apps that go above and beyond in our ability to collect and analyze erging data. Here are two created by a couple of rower-coach-developers that automatically collect, export, and analyze erging data, including peak power testing for durations under 20 seconds. I have no formal relationship with any of these developers or their products.
RowHero is a free app available for iPhone developed by David Winter (Android coming spring 2021). David is a rower and coach who wanted a better way to collect and analyze erg data for teams of rowers, so he built it! RowHero is free for individual athletes and has some scalable plans for coaches depending on what features you want and how many athletes you coach. RowHero solves a couple problems for power testing. First, it exports stroke-by-stroke results to a .csv file. Set up a 20-second timed piece (not “Just Row”), pull hard for your desired number of strokes or amount of time, and then paddle out the remaining time to get to the option to export the results. RowHero also delivers detailed information on drive length, drive speed, and distance per stroke so you can see deeper power characteristics changing too.
Remote Rowing Coach is another free app available in the US for iPhone and iPad, developed by Neil Bergenroth. Neil is a former rower and current coach who built a better system for remote or online rowing coaching and self-coaching. Remote Rowing Coach offers a ton of features that rowing data nerds will just love, but the one we’ll focus on is the built-in power testing function. Neil has set this up to do either a 7-stroke or 10-stroke max with a ton of data collection, including: time to complete, distance rowed, peak power stroke, average power, drive speed, drive length, and force curve analysis. Check out the guided tour of Remote Rowing Coach in the video below.
How to Use Your Power Testing Data
You’ve done your testing and collected your results, so now it’s time to evaluate the data. The simplest way to evaluate peak rowing power is by looking at the greatest amount of watts produced on one single stroke of the test. This offers one single number to indicate whether the rower is improving or not improving in maximum power production.
From earlier in this article, Ed McNeely and Kurt Jensen base their 2km pace estimations on the single-highest stroke, so use this number in their predictive systems. Note that McNeely’s testing protocol is 10 strokes, full slide, from a dead stop, with no rate restriction, at 200 drag factor. You will need to follow this protocol to compare your number to his 55% correlation. There are no details on testing protocol for the Jensen Model on the handy calculation page or in his chapter of Concept2’s old “Indoor Rowing Training Guide” (section 4.09), so I don’t know what testing protocol they used to determine the 173% number.
I’ve never seen a correlation for race paces of other distances, but you could track and determine this yourself over a single season to build on it in future seasons. If you are a masters rower racing 1km, test your peak power during the same week that you test your 1km, and divide your average watts in the 1km by your peak watts in the power test (eg. 300W / 535W = 0.56 or 56%). Repeat the two tests approximately once per month over your racing season and see how the numbers change. Does a consistent percentage appear across multiple tests? If so, you can then remember this number for the next season, just test your power, and estimate your 1km pace based on that number (ie. peak power W * your personal % = target 1km pace).
A more thorough evaluation beyond single-highest peak power stroke offers more detailed information and potentially greater training guidance. We have a great example of this in two recent research studies in rowing power testing.
In a 2017 study, rowing researchers split junior male rowers into two groups for four weeks of training. One group did 30 minutes of plyometric exercises while the other group did 30 minutes of low-intensity cycling. Both groups did the same erging and on-water rowing training together following the plyometric or cycling training. The researchers initially found no significant difference in 15-second power production as measured by single-highest stroke performance, only a small, non-significant improvement in the plyometrics group. The plyometrics group improved from 585W to 629W (+44 watts), while the cycling group improved from 566W to 569W (+3 watts).
In 2019, the same researchers returned to their 2017 data to evaluate 15-second test performance stroke-by-stroke with more detailed information that they had collected using Concept2’s free ErgData app (and the manual video-recording system). When they evaluated the data stroke-by-stroke, they found that the plyometric group produced power earlier in the 15-second test, produced more average power across all 15 seconds, and improved their drive speed compared to the cycling group. The 2017 finding of a small, non-significant improvement in single-highest stroke power missed all of this more detailed, more important information. Think about rowers getting off the line in a race: more power, earlier in the sequence, with faster drives means a major advantage in early race positioning. The enhanced stroke-by-stroke approach improves our ability to evaluate performance, as well as our understanding of the influence of plyometric exercise training for rowers.
Note: I reviewed this study in the April 2021 issue of Science of Rowing.
If you’re managing an entire team and don’t have adequate resources to do a stroke-by-stroke evaluation for each power test, evaluating single-highest stroke power still provides a valuable indication of improving peak power or not improving peak power. However, if you can use one of the apps that automatically collects and exports stroke-by-stroke data, and if you have the time to go through and evaluate this data, I’ve found this to be a very informative approach. Consider using at least some of the following key performance indicators:
- Peak power: greatest watts produced from a single stroke
- Peak power stroke number: producing peak power earlier in the test (ie. stroke #6-7 instead of #8-9) generally indicates increased power
- Overall power: total watts produced in the test, not just single-stroke
- Drive speed: increasing drive speed generally means increased power
- Drive length: should remain the same as a check for technique consistency
How Often Should You Do Power Testing?
Once you have your testing and evaluation systems in place, decide how often you want to test peak power. There are two main approaches here with two different rationales: one for testing every 4-6 weeks and the other for weekly testing.
Every 4-6 Weeks: My preferred approach is to use a single peak power testing protocol every 4-6 weeks in as similar testing conditions as possible. The 4-6-week window is about the shortest amount time that we can expect to see improvement in power production. We may not do this year-round if rowers spend a significant amount of their off-season time away from the erg or high-intensity training. I am most interested in the results in the six months prior to racing the 1km or 2km distance. This is the time when our rowing and strength training is more specific to the shorter duration, higher intensity output. Gradual improvements in peak power over six months informs us that rowers are consistently improving, avoiding overtraining, and building up to a peak performance. I have seen improvements of 8-10% with newer rowers or experienced rowers returning to strength and power training, or improvements of 1-3% later in the training cycle or with very experienced rowers.
Weekly Testing: Coaches and rowers with more time and desire for testing can also use peak power tests weekly. This provides much more data over a year or season of training, and also offers indications of readiness to avoid overtraining. Peak power testing is highly dependent on central nervous system (CNS) performance for the very high intensity, short duration output. CNS-dependent performance typically shows signs of degradation due to overtraining before other longer duration, more aerobic or muscular performances. A sharp decline in peak power performance or multiple declining performances in a row likely indicates overtraining or increased stress from non-training factors. Rowers who train on their own or rowers and coaches in high-performance training situations can use this information to adjust workload individually. Junior, college, or large masters programs could still collect this weekly information and analyze it on more of a whole-team basis to look for general trends.