You are hereBlogs / pmallory's blog / Peter Mallory's Blog

Peter Mallory's Blog

By pmallory - Posted on 18 March 2015

16 March 2015

I promised myself not to get sucked into another fruitless discussion, but my esteemed contrarian colleague from a couple of days ago was by no means satisfied by my response to him in my most recent newsletter, so here goes one LAST response:


Hello Peter,

You're not getting away with that so easily! This is an abstract from literature review and a study that was performed on the New Zealand Olympic team:

Position of peak force

The oar angle at which peak force should occur is a well disputed issue with considerations from mechanics, biomechanics and physiology. Mechanically, it was traditionally thought that the peak force should be applied when the blade is perpendicular to the long axis of the boat. Here the transverse component of the rower's force would be minimal and force would be maximal in the propulsive direction (Celentano, Cortili, di-Prampero, & Cerretelli, 1974; Körner, 1979; Martin & Bernfield, 1980; Smith & Loschner, 2002; Spinks, 1996). Physiologically, it was also reported that a rower's metabolism operates at a higher efficiency if they row "middle pressure" strokes as opposed to "hard catch" and "strong finish" strokes (Roth, Schwanitz, Pas, & Bauer, 1993).

Conversely, front-loaded force profiles are thought to be preferable because the resulting power curve is more evenly distributed and will allow for reduced velocity fluctuations and increased mechanical efficiency (Kleshnev, 2006b; Nolte & Morrow, 2002). It is also known that the elastic energy stored in the bend of the oar shaft and later recoiled, is used more effectively if an early peak force is applied (Kleshnev, 2007c). Additionally, rowing literature considering hydrodynamic efficiency and lift forces at the blades argues that a theoretical optimal efficiency exists with maximal values at the catch and finish of the stroke (Affeld & Schichl, 1985; Nolte, 1984). This can be explained in the following way: for the same propulsion, the same amount of impulse must be given to the water. This can either be achieved with a small mass and high velocity of the water (blade stalled in the water and boat levered past — as is seen through the mid-drive) or large mass and lower velocity of the water (blade moves through water creating propulsion through lift — as occurs towards the catch and finish) (Young, 1997). Because of the nature of the relationship between kinetic energy and velocity, the stalled blade situation results in more energy lost to the water for the same propulsion. Using lift is therefore more efficient, and if maximal power could be produced by the rower towards the catch and finish, then maximal propulsive forces would be achieved because less rower energy would be lost to the water (Young, 1997).

Although it is not biomechanically possible to create a completely rectangular force profile and instantaneously reach peak force at the catch, by emphasising the start of the drive, not only will lift forces be used more effectively, high local loads on the arms will be avoided, and the body will be positioned to more effectively develop force proportional to strength of the body segments (Schwanitz, 1991). Although arguments for both middle and front loaded force-angle profiles exist, most recent literature now confirms that a peak force in front of the perpendicular point is preferable, and the reasons for this are:

1) more evenly distributed power curve reducing boat velocity fluctuations and increasing efficiency
2) increased use of lift for propulsion (a more efficient means of propelling the boat
3) enhanced use of the recoil of elastic energy stored with the bend of the oar shaft
4) reduced overload on arms
5) body better positioned to allow force production proportional to body segment strength



We should continue this conversation over a beer sometime soon. I have no desire to convince you that you have been misled. You appear to be doing just fine as is, but let me respond to the above summary.

1) You and I well understand the physics definitions of force and power. Power = force x distance / time. Simple algebra. The distance, the length of the pullthrough, is constant. As the speed of the pullthrough increases between catch and release, the time goes down and so the power goes up. So if you want a "more evenly distributed power curve," force must decrease during the pullthrough as the speed of the boat increases. Why on God’s Green Earth would anybody want to do that? The whole point of the pullthrough is not increased efficiency. It’s increased boat velocity. No prizes for "more evenly distributed power curves," just for more speed at the finish. What makes sense in an actual boat as opposed to in a biomechanist's brain is maximum available effort from catch to release. This leads to maximum possible acceleration at any given point in the pullthrough, which leads to maximum accumulated speed by the release. This requires a rising power curve, but that should bother absolutely no one.

2) This idea that it is marginally more efficient to attempt to move the boat at either end of the pullthrough than it is in the middle (Affeld & Schichl, 1985; Nolte, 1984; Young, 1997) makes very little difference to the rower in the boat. According to the abstract, there are advantages and disadvantages throughout the pullthrough: vector issues but “theoretical optimal efficiency” toward the beginning and end, minimal transverse force component issues but stalled-blade inefficiencies in the middle. But regardless, each portion of the pullthrough is still an opportunity to help the boat along in its journey toward the finish line. Kernschlag and the abstract say to emphasize the front end at the expense of the rest of the pullthrough. Schubschlag doesn’t say not to emphasize the front end – the front end is good. It merely says to do your best at each phase of the pullthrough as it comes along, i.e. emphasize the cumulative organic whole.

3) There was a lot more bend in oars and a lot more recoil energy stored up in a bent shaft (Kleshnev, 2007c) back in the days of wooden oars, and followers of George Pocock especially took advantage of it. Nowadays the shafts hardly bend at all, and so the phenomenon is not particularly relevant to the experience of the athlete in the boat. Our rowers don't row with slide rules in their hands.

4) According to Steve Fairbairn, Karl Adam and Harry Parker, there is no need to worry about "overloading" the arms (Schwanitz, 1991). I trust their judgment FAR more than that of any researcher or biomechanist.

5) “Force production proportional to body segment strength” (Schwanitz, 1991) is the most universal belief in rowing today. The legs are the strongest, so they should initiate the pullthrough, followed by the back and the arms in that descending order, usually overlapping to a certain extent, and with the quantity and quality of effort inevitably decreasing through the three muscle-group phases. This is the basic tenet of Modern Orthodox Technique, first described in detail by Allen Rosenberg in the 1960s and ‘70s. Today it is the de facto authorized rowing technique of FISA and virtually every rowing manual which has been written in the last half century. I have only one question in response to this idea. Can anybody name any sport on Earth where muscle groups are intentionally used sequentially? Can anybody point to any creature in any body of water on Earth that propels itself with anything but an integrated organic whole-body effort?

You should read Jimmy Joy's The Quantum Sculler, which is mostly metaphysics, but it makes a really important statement, basically that any "point" chosen along the cycle of the rowing stroke is an illusion. For instance, the midpoint of the pullthrough does not exist EXCEPT in what leads up to it and what results flow from it. All that counts is the organic whole.

Any time you tell a rower to place more emphasis on one portion, you are actually telling them to place less on all the other portions. By definition! That's what the words mean, for heaven's sake. No way around that. Simple logic. To my way of thinking, you should tell your athletes to do the best they can at every (illusory) point in the pullthrough with the goal in their heads to maximize the cumulative result by the finish.

The last bit of pseudoscientific claptrap in the abstract above is the concept of an unattainable ideal "rectangular force curve." Force = mass x acceleration, and the bigger the mass, the more force it takes to get the mass to move. Think of a rocket lifting off. It barely moves . . . and then slowly but surely it gains speed. The fact that a rowing force curve looks something like a parabola in the first half of the pullthrough is NOT because you are steadily building effort. It is simply a reflection of the inertia that must be overcome. Then the curve gradually falls away in the second half not because one is trying less hard but because the challenge of continuing to add acceleration to the shell becomes steadily greater until the result inevitably dwindles to zero.

If you like, you can tell your athletes that their effort should "feel" like it would be a "rectangular curve" when graphed. They do all they can immediately at the catch and at every point from there to the release.

But actually, that’s NOT how it feels to them. It is easy to pull hard at the catch. In fact it is easy to pull harder than the boat can respond to. You can pound TOO hard . . . but then you must rebound and recommit, and the resultant two-part pullthrough is universally understood to be poison to boat speed. Then toward the finish your perceived effort must steadily increase until the last few inches take the most skill and speed that you can muster, so you feel a bit like you are straining harder and harder and harder from catch to release as the boat picks up speed.

Another way to think about this is that we want our athletes to maximize the effective work that they do. Work = force x distance and is represented by the area under the rowing force curve, and the shape that maximizes the work is a parabola, a symmetrical haystack, not a left-leaning haystack or any other shape a biomechanist might dream up.

Now no more discussion. Let me buy you a beer sometime, and let's talk about the flowers in spring.

Everybody else, check out my Facebook page for a cogent comment and a link from Jim Dreher of Durham Boat Company. You rock, Jim!