« New Directions in Steering | Main | Wiggling Around at the Finish »

July 27, 2012


Feed You can follow this conversation by subscribing to the comment feed for this post.

Bill Ennis

Hi Karen;
I've really been digesting this post the past few weeks and started to focus on the proper ratio in my stroke cycle. The proper ratio was one of the first things I learned when starting rowing three summers ago, but had forgotten about. Once I started to pay attention again to the proper drive:recovery ratio, I noticed that the boat run was better, my speed went up and the whole thing felt smoother. About the same time I stumbled upon this video(maybe you've seen it):
It's a bit artsy, but it really demonstrated to me what I think you are talking about in this post. I don't know if it was a drill but it appeared to be how she rowed throughout the video. Is this an accurate depiction of what you're talking about?
Thanks again for the blog.

John Greenly

I'm not a great rower, but I am a physicist. I've just discovered your very excellent and helpful posts- thanks!

From a physics viewpoint, with a given amount of work being done by the rower, the boat would make the fastest progress (average speed over the whole stroke cycle) if it could go at a constant speed all the time. That's a simple consequence of the fact that the drag force on the boat (the force that slows you down) is roughly proportional to the square of the speed through the water. That means that while the boat is going faster than the average speed the drag force increases more than it decreases when the boat is going slower than the average, so the drag averaged over the stroke is bigger and you lose net speed. (This is a good homework problem to work out for any physics student out there).

Of course, it is impossible for the boat to actually go at a constant speed. First of all, it's accelerating during the drive, which is unavoidable. What can be controlled to some extent is what happens on the recovery. Just at the end of the drive, the whole system of rower and boat has its maximum momentum. The rower is stationary in the boat, so they both are going the same speed. Now the rower has to get back to the stern, to the catch. The rower weighs much more than the boat, so as she moves to the stern, actually it is the boat that moves forward: the boat speed through the water increases, as you can see by watching any race video, how the bows surge forward during the first part of the recovery. So, the rower certainly should not lurch sternward at the beginning of the recovery when that extra boat speed would do the most damage in increased drag. Therefore, it's good for the hands and arms (very light weight) to move sternward first, followed by the upper body (more weight, but still only part) and finally the whole body moving on the slide. And, it's best to do this all very smoothly to avoid sudden lurches in boat speed. This all agrees well with what we all think is correct technique.
Let's look more closely. Remember that drag is slowing the whole boat/rower system down all the time during the recovery. That means that the rower does less harm with her sternward movement later in the recovery when the whole system is going slower. So, again, the sequence of hands-back-legs is the right way to make the forward impulse on the boat happen more near the end of the recovery than at the beginning, and keep the boat speed through the water most nearly constant. You can see this with very smooth scullers- watch the bow of the boat on the recovery, and you see a nice steady forward impulse over almost the whole recovery, rather than a sudden big lurch at the beginning (or any other time). That's the best indicator of how much "run" the sculler is getting out of the boat.

Now the rower stops moving sternward at the catch. That really means that the rower stops pulling the boat forward underneath her, and at that instant the boat speed and rower speed are equal, and the whole process begins again. The ideal would be to make the recovery just right so that at that instant of coming to a stop on the slide for the catch, the boat has just slowed down to exactly that speed that matches the rower so there is no sudden change. Is this possible? In principle yes, and it does mean that the rower comes up to the catch smoothly, not with a sudden stop- but it takes exquisite feel by the rower to achieve it. Getting this just right also requires just the right drive/recovery time ratio to match the boat speed and stroke rate. Basically, the faster the boat speed and stroke rate, the higher the drive/recovery time ratio has to be. (another good physics homework problem!) We all can feel that to some degree- it feels natural as we speed up to make the recovery relatively faster compared to the drive to get a smooth feel to the whole stroke, and that's why.

If the boat is still moving forward faster than this perfect matching speed, then there will be a decrease in boat speed just as the rower stops moving to make the catch. However, this is not so bad, it just brings the boat down to that matching speed a bit too suddenly. What really "checks" the boat is when the rower starts moving to the bow before the blades are in the water: "rowing it in" at the catch. This shoves the boat backward, and decreases its speed through the water below that matching speed that should have been the slowest speed in the stroke. That's bad, because the boat is going to have to go faster sometime later in the stroke to "catch up" with the rower: again, more net drag over the whole stroke.

So, the two most important times are, (1) the first part of the recovery when the system is moving fastest and extra boat speed through the water increases drag the most, and (2) just at the catch when the system is moving slowest and any check in boat speed through the water requires a compensating increase later. Reducing these extremes in boat speed variation is what makes for efficient and fast sculling. Of course, bouncing the boat up and down is also bad for increasing drag, but that's another story....

John Greenly

Hi, it's been a long time, but I finally got around to thinking this through carefully, and now after re-reading what I wrote, I'd like to emphasize and try to better explain some of the points I made.

First of all, the only way to understand what happens during the recovery is to keep in mind that the rower is NOT moving backward toward the stern, but rather the boat is moving forward under the rower. This is simply because the boat is so much lighter than the rower. If you use a dynamic erg you see this immediately- you stay nearly still over the floor, and the machine moves forward and back under you. The same is true in a boat, but it is hard to perceive, because you are moving over the water and there is no fixed reference. You can see what is really happening if you look at a video where the camera is moving smoothly alongside. The rower's body moves very little in the frame while the boat moves forward and back underneath. It's surprising how hard this is to see, though. I think our natural intuition is that the boat is a large thing , so it must be still and we are moving-- not so!

Now, remember that what you do to maintain a given average speed is to do work during the drive that is equal to the energy dissipated by the drag force of the water on the boat during the whole stroke. The drag acts all the time, so of course if you just sit still after the finish, you and the boat will continuously slow down. In order to take the next stroke you need to get back to the catch though, and that means pulling the boat forward under you, so the boat has to move faster through the water than if you just sat still. That means more drag (the drag force goes up about with the square of the speed). So, the question is, how should you go about drawing the boat forward under you to minimize that extra drag penalty?

The key to the answer is to remember that because of the drag, you and the boat together are going slower by the time you get to the catch than you were at the finish. So, in order to minimize the drag increase due to drawing the boat forward, you should begin the recovery slowly, when you're moving the fastest through the water, and then speed up smoothly right through to the catch, to keep the boat speed through the water most nearly constant. If you look at video of fine scullers, this is what you see. The boat moves smoothly forward during the recovery, rather than jumping forward at the beginning- or at any other time.

So, how do you do this? Remember Isaac Newton and his third law: action and reaction. The more of your mass you move, and the faster you move it, the faster the boat moves forward in reaction. Here's where some of the conventional rowing wisdom doesn't hold up very well. What about "fast hands" and early preparation of the body angle? Well, fast hands is moving a fairly small mass fast, leaning the body to the stern is moving more mass, and as the knees lift, almost the whole body mass is engaged. So doing any of these things suddenly will lurch the boat forward, and doing them fast, early in the recovery will hurt the worst. At a given stroke rate you have a certain amount of time, say one second, to complete the motion, so if you do it slowly at first, you have to go faster later. There's not just one unique way to coordinate that one second's motion, but clearly it must be a smooth combination of movements. Looking at the best scullers in recent years, you see the hands do not shoot out far before the body moves, the body angle begins to close sternward gently quite early, the knees don't break upward with a sudden jerk, everything finally comes together smoothly and accelerates to the catch. Mahe Drysdale is a good example, he seems rather slow to begin the recovery, then accelerates smoothly to a very quick reversal at the catch. Or look at the astounding Murray/Bond pair. There simply isn't time at that stroke rate to do sequential hands then back then legs, it's all one motion, accelerating to the catch.

Another thing you can see in videos is a sudden decrease in boat speed right around the catch. This is bad, right? Checking the boat? Well, actually it is unavoidable. Think of the whole stroke sequence. During the drive, you are actually leaving the boat behind, and during the recovery you are pulling it back to you. It has to go faster than you on the recovery, and slower than you on the drive, so that decrease in speed is inevitable, and it is quite sudden at race pace. I was wrong in what I wrote before- when you work it out, it is not possible for a sculler at race pace to eliminate this sudden decrease. There just isn't enough time available in the recovery. It is possible at low rating by making the recovery last longer so you slow down to match the speed at the catch. I think you can feel this smooth transition when you are sculling at a low stroke rate with a long, slow recovery.

What about the idea that if you come up fast to the catch, you check the boat because you have to push on the footplate to stop yourself? This is just not true. Remember, you are not moving to the stern, you are pulling the boat toward you. All you need to do to stop at the catch is to stop pulling your feet toward you. You can prove this to yourself: come to the catch, stop, and just coast without dropping the oars in. There's no check of the boat no matter how suddenly you stop at full compression- you just glide onward. What really checks the boat is when you reverse the motion and push the footplate away before the oars are in the water. Try it! After you stop and glide in the catch position, push with your feet, still without dropping the blades in. That certainly does push the boat backward.

I've been looking at videos, trying to see how the great scullers negotiate the catch. There seems to be a lot of variation in how the blades are brought to the right speed to enter the water cleanly as it moves by the boat. Some open the back angle, some use elbow bend, and some use a subtle combination of fingers, hands, shoulders, letting the motion arise from the natural contraction as the muscles are engaged. None use legs, motion of the seat. That would check the boat. So much for "feeling the catch in the feet". Here's a link to a very nice set of videos that you can step through frame by frame: http://invernessrowingclub.org.uk/personal/

Olaf Tufte uses quite a lot of elbow bend, Ekaterina Karsten seems to open her back angle, Rumyana Neykova is very subtle. Tufte does something astounding that I haven't found any one else do. He actually starts the blades into the water while his seat is still moving sternward on the slide- no chance of checking the boat there, and he gets absolute maximum use out of the very first part of the drive. See also his recovery overall- a very subtle and smooth blend of motions with a smooth acceleration to the catch, definitely not separate arms then back then legs.

One final thought. What Tufte is doing is even more than eliminating the possibility of checking. Even before driving with his legs at all (his seat is still moving sternward, remember) his blades are already beginning to drive the boat forward. How can this be? He is working off his own inertia to pull on the handles and drive the boat before he pushes with his feet. You can prove this to yourself. Begin stationary on the water, sitting somewhere around half slide, with arms extended and the blades squared and immersed. Relax your legs completely, try not to let them exert any force. (if you have a wing rigger, just rest your legs extended on top of it). Then give a quick hard pull with your arms. The boat will move forward, you will move backward in the boat. You have just used your own inertia, and old Isaac's third law- you pulled the boat forward, so you moved backward. If you release the blades neatly as the handles hit your chest, the boat will keep on going. Not only have you pulled the boat forward, you have pushed the entire Earth backward- all with your own inertia, and Newton's third law!!

The comments to this entry are closed.