What science says of seat angles

The popular triathlon-related magazines have carried articles on what seat angle you ought to use. But has this subject been addressed in the not-so-popular magazines, that is, the scientific journals? Our sport’s glossy pubs have done a good job of presenting to the layman the "science" on hydration and hyponatremia and related subjects. It’s not overstating it to say that hundreds of studies have been carried out in each of those categories, with corresponding articles published in peer-reviewed publications.

But what about bike position? If Triathlete or Inside Triathlon has reported on what’s been published in the scientific journals, I haven’t seen it. What they’ve written is like what I’ve written, which is that according to the author and the mouse in his pocket, this seat angle or that one is better.

I’m not a real scientist; I just play one on the internet. But there have been real studies performed by real scientists on the subject of seat angles, and I thought I’d write about some of them. All the tests I’ll write about below had subjects riding on a modified bicycle ergometer, that is, a stationary bike with the ability to offer a range of seat angles. Not in every case could I find what the protocol was for handlebar alterations that accompanied a change in seat angles. Take for example, "Cardiorespiratory responses to seat-tube angle variation during steady-state cycling," by Heil, et al, (Medicine and Science in Sports and Exercise, 1995). This study had cyclists riding a bicycle ergometer at four angles ranging from 69 degrees to 90 degrees. We don’t know with any precision how the handlebars were altered to negate the ill effects of a bad-fitting bike. Just the same, the study showed that only at the shallowest of angles (69 degrees) did the respiratory system of a cyclist come under more stress than at any of the steeper angles.

Two years later two studies came out. The first of these was published in the same journal as the study above, and was entitled, "Influence of different racing positions on metabolic cost in elite cyclists" (Gnehm, et al). It did not specifically test seat angles, but the study is important to note. Its authors mention going in that, "The spectacular improvements of the 1-h world record in cycling in the last four years have highlighted the importance of aerodynamics in modern bicycle racing." In the early-to-mid ‘90s the record had been significantly lowered by Moser, Boardman and Obree, each of which used more and more exotic bikes and bike positions. This study tested 14 elite male bike racers in three positions: upright with hands on the tops, crouched with hands on the road bars, and laid out on the aero bars.

Gnehm looked at metabolic changes while riders were in these three positions, and noted that riders in the aero position paid a penalty of about 9 watts versus the other two positions. It also noted that this is minor compared to Gnehm’s estimate of 100 watts an aero positions saves by virtue of a lower wind resistance. This was an oft-cited study by cyclists in newsgroups and forums in the late ‘90s, but I doubt its relevancy. First, I doubt that an aero position saves 100 watts. That’s a huge number, and I would guess that 20 - 40 watts might be more like it. Second, it appears that Gnehm did not change the seat angle for his subjects when he flattened their backs and lowered their drag. What Gnehm’s study seems to indicate is that even if you put a rider in a fairly uncomfortable aero position, the power output doesn’t markedly diminish.

The second of the ‘97 studies was published in the Journal of Sports Sciences (Price and Donne) and was called, "Effect of variation in seat tube angle and different seat heights on submaximal cycling performance in man." The article stated that, "At a seat tube angle of 80 degrees, mean VO2 was significantly lower and power efficiency significantly higher compared with an angle of 74 degrees." Likewise, 74 degrees offered more efficiency than 68 degrees.

This study is interesting because it tests not only three different seat angles (68, 74, and 80 degrees) but variable seat heights at each angle. As the quote above suggests, 80 degrees is the most efficient angle at which to ride. But there is of course more to the story. What were the handlebar configurations? What were the stem heights, the hip angles? We’ll get to that.

Interestingly, Price, et al, noted that differences in seat angle preferences were specific to classes of athletes: "...in contrast, triathletes’ bicycles have steeper angles of 78 - 82 degrees." The author also references the Heil article (noted above) and suggested that one reason why Heil found that 68 degrees was the only angle in which his subjects performed poorly was in Heil’s choice of subjects, "80% of whom," Price noted, "were triathletes who normally ride at a steep seat tube angle."

This particular study simply utilized a seat shifter, a nifty device once in relatively common usage by triathletes in the early '90s. It replaced the bicycle’s seat post, and with a lever on the handlebars you could change your seat angle on the fly. I’ve tracked down the study’s author and emailed him, asking him to verify that no changes were made to the handlebar set-ups as the seat angles were altered during his study. I’ve yet to receive a reply. I don’t, however, see any notation that any change in the bikes was effected other than a change in the seat angle provided by the seat shifter.

The results of this test are, to me, startling. These were road racers riding a road race bike, and they in general rode with considerable economy at 80 degrees versus their usual angle of 74 degrees. Oxygen consumption at the given rate of exertion was about 37 ml/kg/min at 80 degrees versus about 38.5 at 74 degrees (if you’re consuming more oxygen to do a given amount of work, you’re working harder, i.e., you’re less efficient).

Why did the authors achieve this result? "The mean shoe-pedal angle changes produced by altering the tube angle," opined the authors, "would result in a decreased effective force during the first half of the pedal stroke but an increased effective force during the second half. We speculate that increasing the tube angle improves effective force transfer during the second half of the pedal stroke."

The authors also noted that this series of tests only related to riding on the flat, and to submaximal power outputs, often atypical of race conditions. Just the same, it’s an interesting result and may bear on the next and, for us, most topical study.

The definitive article is now passing its second anniversary since its publication. The "Effects of bicycle frame ergonomics on triathlon 10-km running performance," by Ian Garside and Dominic Doran, appeared in the Journal of Sports Sciences in June 2000. As was the case with all the studies on this subject, the tests were performed in strict lab conditions, with gas analyzers and all that stuff. All subjects rode stationary ergometers.

Unlike Price et al above, Garside utilized triathletes, but: "All participants were naive to training and racing on bicycles with steep seat tube angles (>76-degrees); all participants used a 73-degree frame geometry as standard."

As opposed to most of the testing up to this point, Garside’s protocol called for the tests to be conducted, "as fast as possible under race-like conditions." The test called for triathletes to ride a 40km simulation on both a 73-degree set-up and then on an 81-degree set-up, each followed immediately by a fast-as-possible 10km run on a treadmill.

The authors noted the improved bike/run performance in the field, "based on anecdotal testimony from athletes purporting to have experienced improved performance." But, they noted that prior to this study, "No empirical evidence exists."

Frankly, the results were groundbreaking, for three reasons. First, these triathletes absolutely blew away their "duathlon" performances in the steeper configuration. The average time it took subjects to complete the 40km/10km "brick" was about 1:50 at 73 degrees of seat angle, and it was a full 5+ minutes faster at 81 degrees.

Second, as this test was performed in England, the triathletes there were (as previously noted) "naive" to steep seat angles, that is, they all normally rode shallow. Imagine their surprise at the result! (For a further explanation of the tendency of UK triathletes to ride shallow, consider our 2001 Kona Bike Survey on UK-based entries).

And finally, these tests only measured the physiological responses to the biomechanical change generated by a steeper seat angle. As this test was performed in a lab on stationary equipment, the aerodynamic benefit one derives from the ability to achieve a lower frontal profile with a steeper seat angle was not part of the equation.

Where in this 40km/10km exercise did the time savings occur? There were some time savings achieved during the bike leg. Average 40km times were 1:04:10 in the 73-degree configuration and 1:02:54 for the riders when at 81 degrees. But it was in the first half of the run that the big time savings occurred. It took riders 24:15 to complete the first 5km off the shallow set-up, and only 21:41 after riding the steep bike (and remember, these triathletes had never run off a steep set-up before).

The time savings continued during the second half of the run, but the gap narrowed. The subjects ran 22:01 and 21:14 in the second half of the 10km after riding in shallow and steep configurations respectively.

There is one curious element to these results which is not addressed in the study. It is odd to me that as a group there was such a marked tendency to negative-split the run. While this is not as pronounced in the steep test (21:41 to 21:14) it is severe in the shallow test (24:15 to 22:01). This one element gives me pause when considering the parallel between this test and what happens at triathlons I attend.

As the authors discuss the results they say, "Unexpectedly, the time to completion of the 40km cycle section was faster under the 81-degree 'steep' than the 73-degree 'shallow' condition." This underscores the tendency, I think, for the traditional view to hold sway in the UK, which is that if steep is faster, it’s only because it helps during the run. Obviously the authors were forced to rethink that position, especially as there is still the untested (by them) issue of wind resistance to consider.

The authors only guess at the causes for the enhanced ability to perform with the steeper seat angle, and posit about the,

"greater contribution of the hamstrings and gluteus muscles (Heil et al., 1995). Although muscle recruitment cannot be determined from the present results, alterations in muscle recruitment or activation patterns can have the effect of distributing muscular work over a greater muscle mass (increased contribution of the hamstring and gluteus muscles) that would theoretically reduced the work rate per individual muscle fiber (Coyle et al., 1988)."

Juxtapose that statement, in which the operative phrase is "distributing muscular work over a greater muscle mass," with what Price says in the study above: "We speculate that increasing the tube angle improves effective force transfer during the second half of the pedal stroke." It seems that both authors feel that steep seat angles might distribute work over a greater range of the pedal stroke and in so doing lessen the peak torque that must be applied if should that power application be concentrated over a shorter arc.

The scientists quoted above uttered statements that are not unlike what I wrote a couple of months ago in my Intro to training with power: "I’m convinced that the bane of the triathlete during the bike segment is the peak power one puts out, and not only in terms of too many watts now and then during the ride, but peak power inside the pedal stroke as well." The difference, of course, between what I wrote and what the scientists wrote above is that I was simply writing from the seat of my pants, while Price and Garside had actual data they themselves generated to back up their hypotheses.

The demonstration that steeper seat angles (and not for tri bikes, but for road race bikes) is better is not knew. Gonzalez, et al, (Journal of Biomechanics) demonstrated in 1989, in "Multivariable optimization of cycling biomechanics" that the optimal seat angle for his subjects averaged 76 degrees (as is the case with all of these studies, however, this relates to riding on the flats). But Gonzalez’ study also demonstrated that riding with a cadence of 115 bpm was optimal, and this corresponds only roughly to a real-world race situation.

Though these studies were all conducted inside four walls, and took place on the "flats" and were only sparsely conducted with race-specific exertions, evidence is piling up in favor of steeper seat angles. And that is especially true when one considers the post-cycling run in triathlon, at least up through the International Distance.