This post is about what measurement you “should” use as you train to monitor your on-water performance. I see a lot of people at the clubs where I coach sign up for lactate testing once a year or once every few years, and it makes me wish they would pay me instead of that person. (sigh) And every once in a while I come across someone who has just discovered the heart rate monitor and is all excited about using it during training. But then I listen to the collegiate coaches and coxswains tell their rowers “bring it up to full pressure” and it gets me to wondering what the best way to monitor yourself during a piece really is.
There are a few options:
Rating
Perceived effort
Heart rate
Lactate production
Boat speed relative to land
Boat speed relative to the water
All have their advantages and disadvantages. I’ll come out right now and say that my favorite is to use boat speed relative to the water, because it’s a fairly easy quantity to measure and interpret – though not completely straightforward [see my next post for more information]. But I’ll make an effort to discuss the pros and cons of each method so you can decide for yourself what you want to use.
Let’s start with rating, or number of strokes per minute. Everyone who’s ever rowed (or walked, for that matter) knows that the more strokes (or steps) you take per unit time generally correlates with speed. So it’s often convenient to phrase workout intensities in terms of strokes per minute -- I do this on the fly, when I have to explain what a certain workout is like. An easy workout might be done at 18-20 strokes per minute, while a hard one would be above a 28. But the more experienced you get at sculling, the more skilled you will be at separating effort from rate.
This is a very good skill to learn, in fact. It’s usually easier to learn to pull hard at low rates than it is to be light and quick at high rates, but both are important skills. Pulling hard at an 18 or below allows you to practice long, locked-on, efficient strokes without building up too much lactate, so you can do these workouts for long intervals. And if you’re doing something like a bungee row, or towing a can, or tying plastic shopping bags onto your bungee, you’ll get the perfect rowing-specific lifting workout. I like doing low-rate, high-pressure workouts (in moderation) to develop the efficiency of my stroke. If I do too much over the course of a month or two, it starts to aggravate an old overuse injury and I get sore, but bits of it here and there sprinkled throughout the training year really do help me stay long and locked on when I scull.
The harder skill to learn is how to pull lightly but at a high rating. And to those of us steeped in the dictum never, ever to rush the slide, it’s even more of a conundrum. But the reason it’s advantageous is exactly in racing. If you think about rowing high and light, you’re racing much more aerobically than if you’re just muscling it every stroke. It took me a really long time to learn this, and I wish I’d gotten it a LOT sooner. (but such is life) You can practice it yourself if you have a SpeedCoach or if you’re on an erg. Start at a moderate pace, something you can keep up for 30 to 60 minutes, and note your splits or speed. If you start at something like 20 strokes per minute, then stay at that rating for a few minutes to establish a rhythm. Then, take it up two beats per minute every three to five minutes (depending on how long you plan to stay on the erg that session). The important thing as you do this is to maintain the same speed throughout. That is, if I’m at a 2:25 per 500m at 20 strokes per minute, I want to be at a 2:25 at a 24, too. There’s a limit to how much you can push this; if you’re trying to do the same pace at a 20 and at a 32, you’re going to have to rush horribly at 32 to get that same pace. In fact, you might find you’re working harder on the recovery than on the drive. But if you’re at a 24 and rowing lightly, then you will feel a little like you’re just going back and forth, which is a different ratio than you might be used to. That’s okay.
And that also leads us to our next proxy: perceived effort. This means “go by feel”, and it’s usually measured in terms of what’s called a “modified Borg scale.” You can read more about different RPEs (ratings of perceived effort) on that link, but the basic idea is to make “no effort” the lowest number on your scale and “all out” the highest number on your scale. Of course, this description creates several questions right from the get-go. For example, while “no effort” is usually pretty straightforward, “all out” is a little more complex. In rowing, particularly, does that mean “the hardest you can go at all?” Or does it mean “the hardest you can go for a particular distance?” If 100% effort (or “full pressure”) means the first thing, then you wouldn’t really want to race a 2K at full pressure, would you? Unless you didn’t care at all about lactic acid building up and making you stiff and unable to move. Instead, you might decide, after some practice pacing yourself and calibrating your RPE to different times, that your optimal racing pressure was 75% or thereabouts.
But a lot of high school or collegiate coaches don’t like to phrase race pace like this because they are afraid that their athletes will undershoot what they are actually capable of and go too slowly. Instead, they will call race pace (for whatever distance their athletes are going to race) “100%”. This communicates to the rowers that they must pull as hard as they can for the distance, and makes it more likely that every athlete will be giving their all. Of course, this scale also has its own drawback, which is that it biases rowers toward pulling too hard at the beginning of a race. And, in fact, the “fly and die” phenomenon is quite common among novice rowers.
And, of course, you’ve already figured out that both scales correspond to something inherently subjective. That means that not only can different people’s “100%”s correspond to different actual speeds, but also that one person’s 100% might yield one speed on a good day and lower one on a bad day.
That’s partly what spurred the movement toward more objective measurements, like heart rate. As you well know, the harder you work, the faster your heart beats. So, heart rate is correlated to perceived effort and might yield a better metric than “I feel like I’m going all out.” Also, you can measure it without any equipment at all except your heart and a finger on your pulse, though for trained athletes it's better to use a heart rate monitor because their heart rates start to slow down within about 20 seconds after stopping hard work.
The first, and most simplistic, way of using heart rate to monitor exercise intensity is simply to express all workouts in terms of a percent of maximum heart rate. Maximum heart rate is simply the fastest your heart can beat – but even there, there are some problems. One is that your maximum heart rate decreases with age. Another is that max heart rate differs by person. My max won’t necessarily be your max, though they might be close. And what about doing the experiment to ascertain exactly what our max heart rates are – even if we survive, it’s NOT going to be fun. So, here again, there is some estimating. The rule used to be “220 – your age”, so that my max heart rate when I was in college would have been something like 200 beats per minute, and now would be more like 180. I haven’t done all-out full pressure with a heart rate monitor on in a long time, so I don’t know whether I can still get my heart over 180 beats per minute. But I do know that by the time I was in my early twenties, the highest heart rate I saw on my monitor was something like 187. After that, the tunnel vision would set in, so if my heart rate went up I didn’t know about it anyway.
Another issue in using heart rate as a proxy for exercise intensity is that if you use a straight percentage of maximum, you would never get below some number. That is, you would never get to, say, 5% of maximum. That’s because your lowest heart rate isn’t zero. This number, the “resting heart rate” again varies according to person and with fitness level, but for trained endurance athletes it can range from 35 to 70. Even when I was training, my heart rate never went below about 75, but then I was a complete stress bunny all the time. I actually suspect my heart of being reluctant to increase its beat rate, preferring instead to just beat harder, at least for a while. Gotta love that pounding-chest feeling…
Anyway, the Karvonen, or heart-rate reserve, method of calculating % of max heart rate is a way of taking the resting heart rate into account. It involves first determining the dynamic range of your heart rate. To do this, you find your max heart rate (either by formula or empirically) and subtract your resting heart rate. You are supposed to measure your resting heart rate first thing in the morning when you wake up, averaging over a few days to make sure you weren’t just really excited the first morning. The problem with that for me was that my heart rate always shot up when my alarm went off, because I knew I was going to have to get up and start working out, and most of the time I would much rather have stayed in bed. In any case, you could then calculate (say) a HR corresponding to 50% effort by halving the difference between resting and max heart rate, and the adding that amount to resting heart rate. So if my max is 190, and my resting HR is 70, that gives me a range of 120 to play with. Half of that is 60, and if I add that to my resting HR, I come up with 130 beats per minute as my 50% effort. (Which, I might add, was always too easy for me for a 50% effort row. But then I could have been working too hard).
There are still some caveats for using heart rate as a proxy for effort, though. One is that there are several factors that can cause heart rate to increase that don’t mean you’re working harder. One of them is body temperature, and this turns out to be a very important one because your body heats up as you work out, irrespective of level of effort. In practice, this means that as you try to hold a constant level of effort or a constant pace, your heart rate will creep up. People have tried to get around this by simply specifying a range that your heart rate should be in for a particular type of workout, but given the amount of variation in individual heart rates to begin with, this method often turns back into perceived effort
Another cause of elevated heart rate is lack of sleep – and I’ll tell you I suffered from this all the time I was training. I suppose if I’d been independently wealthy, or willing to live in my car (oh, wait; I didn’t have a car), I’d have been able to give up working and would have had more time to rest. But no rich uncles appeared to leave their money to me. (Ernie? Steve? Bill? I’m still waiting. Just sayin’). And with no car, that kind of changed the whole equation. Plus, everyone’s short on sleep sometimes.
Given these problems, some people have decided that measuring lactate concentrations is the way to go to tell whether your training is helping or not. The theory behind it all is that [one of the things] you want to find is the fastest speed that an athlete can keep up without building up so much lactate that their performance over the relevant distance suffers . Then, if you do the same thing after a macrocycle or two of training, and you found that there was less lactic acid for the same speed, you would know that your training worked. So if the buildup of lactic acid in the muscles is what one wants to measure anyway, so why not just go ahead and measure it?
One issue that comes up is that the buildup of lactate in the blood is not the same thing as the buildup of lactate in the muscles. Lactate only diffuses into the blood after it’s already built up in the muscles. Optimally, you would want to measure the buildup of lactate in muscle during actual exercise, but this is impractical. To do this, you would need to take a small biopsy of the muscle, freeze it, and then analyze it after the fact. And muscle biopsies hurt. Not to mention, that kind of analysis is really pricey – so most people who measure lactates do so by making the measurements from blood samples, and they take those samples from either the fingertip or the earlobe. (Again, lurking in there is the assumption that the level of lactate in the blood taken from those sites will give you an accurate estimate of the speed at which muscle lactate increases). Oh – and don’t forget – you need to take universal precautions when dealing with blood or blood products. That alone makes this method less than appealing to me for club rowing, even aside from the expense of the equipment.
Then, you need to have your athlete do pieces at increasing speeds, measuring lactate after each piece, to make a graph of how lactate increases with increasing metabolic demands. There is a fair amount of complexity in interpreting the results, which I won’t go into here. But the would-be lactate measurer does need to put some thought into what precisely he or she would like to find, why that might be important for a particular athlete, and how training might be expected to change the results. (You would be looking for different changes for a longer-distance athlete than a shorter-distance one, for example; and if you wanted to make sure that a person’s strength and power output increased, you’d have to know what effect that would be likely to have on the lactate results).
In addition, there are several factors which can give you anomalous results. How much glycogen you have in your muscles is one. If you carbo-loaded the night before, you will appear to perform “worse” (i.e., have more lactate for a particular speed) than if you are a little carbohydrate-depleted. This seemingly paradoxical reaction happens because you can’t engage in anaerobic metabolism if you don’t have enough glycogen. So if you’re glycogen-depleted, you will need to rely more on aerobic (non-lactate-producing) metabolism. But that might put a cap on the top speed you can achieve, which is why it’s not recommended as a race preparation plan. Likewise, drinking caffeine before testing will increase fat metabolism and make it look like your anaerobic threshold has increased, when in fact it hasn’t. (This might be a good race prep plan, though).
Hard training within 24 hours before a lactate test will make you produce more lactate for a certain speed, too, because of the change in amount of stored glycogen or because of muscle damage. Time of day makes a difference, as well: You’ll usually be able to go about 2% faster in the afternoon than the morning, with no change in aerobic capacity. Then, of course, you need to do a second set of measurements after one training cycle, in order to determine whether training has had an effect (and to determine what effect it has had). Once you’ve verified that your measurements of the correspondence of lactate concentration to rowing speed are correct (by measuring a third time), you’ll be able to prescribe training speeds for an athlete. These will be designed to improve performance in one or another area so as to ultimately improve racing speed.
Finally, you might think that you can use lactate-to-speed correspondences to compare individuals. That is, you might think that if So-and-so has X millimoles of blood lactate at such-and-such a speed, you might aim for that same correspondence yourself. But be careful: for middle distances like 2K races, the contribution of anaerobic threshold to differences in speed (for Olympians, a relatively homogenous group) is only about 80%. The shorter the distance, the less contribution anaerobic has to speed differences. So it is difficult to predict race results on the basis of threshold speeds alone.
It is for these reasons that I prefer training by speed best. First, it is noninvasive – a huge win, in my opinion. Again – the use and interpretation of speed as a way of monitoring training isn’t completely transparent, but I feel its convenience outweighs a lot of the drawbacks.
First, though, comes the decision over what distance to measure your speed. Research [see Ernest Maglischo’s book “Swimming Fastest” for a beautiful exposition of all of the research on performance monitoring] has shown that a 30 to 35-minute piece, all out for the distance, provides a very accurate estimate of anaerobic threshold. It is so long that you really can't go much over your anaerobic threshold. The caveat here is that you must row it as evenly-paced as possible. If you go out too fast, your average speed will be lower than your threshold pace because you will have put yourself into acidosis too early – you’ll have flown and died. In fact, if you pace yourself well, you’ll get a more accurate estimate of your threshold speed than if you used blood measurements to begin with. And if a half-hour long piece isn’t to your liking, you can still get accurate threshold estimates with a 20-minute piece.
I particularly like this because many head races are approximately 20 minutes in length. Since I coach primarily on the Charles, my scullers can get practice steering the Head of the Charles course at the same time that they do a 20’ piece for meters. Because there are many turns in the river, there is often wind, and sometimes there’s even a current, it’s not the absolute speed I will be interested in – only the improvement from training cycle to training cycle.
Which of course brings up the issue of factors that can cloud the interpretation of speed results. This will depend on what method you use to measure speed, of course, and there are two main ones: GPS-based (or ground-based) measures, and SpeedCoach-based (or water-based) measures.
If you’re measuring speed relative to some objects on the ground or in the sky by using the start and finish lines of a particular racecourse or by using a GPS method, your results will be affected by both wind and current. How they’ll be affected is pretty intuitively clear: a headwind will slow you and a tailwind will speed you. Rowing with the current speeds you and against it slows you. Turns are a little more complicated, but since they affect ground- and water-based speed measurements equally, I will leave them out of our calculus here. Water temperature also affects both types of measurements (it’s really the density of the water – if it’s colder you go slower; if it’s warmer you go faster).
By contrast, the measurement of speed using a SpeedCoach will be affected by wind in just the same way as with ground-based measurements. But current won’t affect you at all, because the impeller measures your speed relative to the water itself. If you’re in a strong current just floating along, your speed will be zero. If you’re rowing, your speed relative to the water will be the same regardless of whether you’re going with or against the current. (It’s just hard to get your head around this if there are trees in the background of your imagination, because then you’ll go faster relative to the trees if you’re with the current than if you’re going against it. Ignore the trees).
Of course, speed relative to the ground or water will be affected not only by physiological improvements, but also by improvements in technique. This isn’t a bad thing, to me; because it’s really the overall improvement in my athletes that I want to measure. I might not care whether that is coming in this macrocycle from increased fitness or better technique. It’s all going to come down to increased speed on race day anyway. There are certainly issues of training efficiency, and whether improvements in technique might be canceled out by a poor training regimen. But at the end of the day, my scullers care about whether they are going faster. So measuring it and showing them improvement is good for their psyches.
In the next post, I’ll discuss training by splits (water speed). The complexities I’ve mentioned above will come more to the fore there, but are not at all insurmountable. In fact, they will turn into advantages. Stay tuned.
