Center of Gravity method of Road bike fitting

As many of you know, I’m working at JackRabbit’s new store in Manhattan as the lead bike designer / fitter. Our business model is to be one of the top shops in the country for custom road and triathlon bikes. To make that a reality, we are investing heavily in tools and processes that allow us to do a better job than anyone else out there. One of our many innovations is a new technique for determining proper saddle placement for road bikes. I think it’s more accurate than any of the other methods I’ve researched, and I don’t think anyone else does it this way, so I wanted to share what I’ve learned and get some feedback from the Slowtwitch community.

The Nutshell version of the method: Most fitting systems dictate the “Knee Over Pedal Spindle” rule, where the fitter drops a plumb line from the knee. I’ve read just about every document available regarding saddle fore/aft, and I’ve seen this method described in almost every one, but have never seen an adequate explanation for it. In our method, we set up the bike (in road bike fits) so that the rider’s center of gravity is directly above the bottom bracket, with a slightly more relaxed position for novice riders and a slightly more aggressive position for experienced riders. Results of a few dozen fittings and custom bikes thus far have shown that the majority of novice cyclists (many entirely new at road cycling) prefer a position which places their center of gravity (C.G.) just behind (0-1.25 cm) the bottom bracket, while more experienced cyclists have preferred the position which places their C.G. between 0 and 1.5 cm in front of the bottom bracket. The text below is from a document which will soon be posted on our website. I’ve also included a few pictures of our design set-up, which includes 4 laboratory scales for accurately locating our client’s center of gravity in a variety of positions and geometries. As you can see we’re using the Argon18 fit bike which allows us to rotate the seat tube angle accurately and easily from 70 to 80 degrees for both road and tri fits. I know the wires and tools are all over the place and the platform is dirty. I promise I’ll clean it up before you come in.

Road Bike Fore/Aft (Not Tri Bike!):

The fore/aft saddle position on a road bike is probably the single most important aspect of a comfortable and powerful ride. Unfortunately for cyclists, the popular method for determining saddle fore/aft is based on observation and is nothing more than a rule of thumb. The method is called the Knee Over Pedal Spindle (K.O.P.S.) method. The K.O.P.S. method says that the correct saddle position is the position which places the bottom of the knee cap directly over the pedal axle when the crank arm is in the 3 O’clock position (when viewing the right leg). K.O.P.S. can put a rider at a very shallow seat angle if they have long femurs, or a very steep one if they have short ones. There is no real science behind this method; it is based entirely on observation of what many riders deem a comfortable position. The leg is made up of several bones and joints which create a lever system that turns the crank arms. If one slides the saddle position 2 centimeters forward and adjusts the saddle so that it is at the same height as it was previously, then the lever system of the leg is exactly the same as it was before hand, just rotated a few degrees forward; the knee has no idea if it’s directly above, in front of or behind the pedal spindle. From a bio- mechanical standpoint it is identical, as shown in the diagrams below. http://tinypic.com/mcdyc5.jpg

The cyclist on the bottom has been rotated forward two degrees about the bottom bracket, note that the limb lengths and frame geometry have been drawn completely arbitrarily, but are consistent between the two diagrams.

The only thing that does change as one rotates the saddle fore and aft is the location of the rider’s center of gravity. If one’s center of gravity is too far forward, then an excessive amount of weight needs to be supported with the cyclist’s core and arms, and if one’s center of gravity is too far back, then an excessive amount of weight is placed on the cyclists sit bones. Bicycling is essentially a one legged squat performed over and over again. As one squats down, they bend at their hips pushing their torso forward and their rear backwards, this keeps their center of gravity directly above the ball of their foot, maintaining balance. The goal of the saddle position is to put the rider in a position that is as natural and as balanced as possible. If one were not pedaling at all, then their center of gravity would need to be directly over the bottom bracket in order to maintain balance. Pedaling however, creates forces which need to be balanced with the rider’s body weight in order to maintain a state of equilibrium. Simply put, a rider’s center of gravity should be slightly in front of the bottom bracket in order to maintain a balanced position while riding.

Think of a cyclist riding a bike with no handle bars, but maintaining their upper body position as if they were. If the cyclist does not pedal, then their weight will cause them to fall forward. The cyclists pedaling forces however, create balancing forces to keep them upright. Stronger cyclists generate more force at the pedals than weaker ones, which is why they should be positioned a little further forward (this lengthens the moment arm of the cyclists center of gravity).

Unfortunately for this method of bike fitting, it is impossible to establish one position that will work perfectly for every single minute in the saddle. The cyclist doesn’t always generate the same pedaling force and the pedal stroke has two major dead spots in it where the cyclist isn’t pushing down at all. When the cyclist’s pedal stroke is at the 12 and 6 O’clock positions, the pedal stroke is doing nothing to keep the cyclists upper body from falling forward. Furthermore, if the cyclist is fit using this technique while generating 175 watts at 90 rpm the pedaling forces are less than what they would be at say 175 watts and 70 rpm. The one thing that this method does have going for it is that there is some science to it and it can place the rider much closer to the position that they will feel the most comfortable in compared to a K.O.P.S. derived position. Ultimately, the rider needs to take the bike out on the road and experiment with a few different saddle positions to find the most comfortable one. The bike fitting will get the rider very close to where they feel they are positioned correctly; however it is the rider’s job to take the bike to the road to finalize the fit.

http://tinypic.com/mcekwk.jpg

http://tinypic.com/mcpvnq.jpg
.

“Bicycling is essentially a one legged squat performed over and over again”

Really?

Can you explain this - because I think that this:

http://www.exrx.net/AnimatedEx/GluteusMaximus/BBSingleLegSquat8.gif

… is actually pretty darn different from a pedalling motion…

“The goal of the saddle position is to put the rider in a position that is as natural and as balanced as possible”

I don’t agree with this at all, at least for a racing cyclist.

Screw natural and balanced.

If I can get 5 more watts from an unnatural position that requires me to use my arms to maintain body position, then that’s what I want…

.

So the unicycle is now the fastest?

I try to spin circles (“dancing on the pedals”) and a one legged squat it is not.

Nice rig though.

I was thinking the same thing - can only maintain balance when the cg is over the bb. But of course, a unicycle!

go read your coni manual. you have the right idea…but you’re way out in left field. setback can only be viewed as one element of the totality of custom bike design. when it comes to road bikes it is about balance, but you need to think about balancing the weight of the athlete between the two wheels…if you’re just pushing the seat around and not concentrating on chainstay lengtht, front center stem length and all the other junk that causes a bike to “work”…you might as well just place the saddle in a place that feels comfortable for the athlete that doesn’t throw off the designed in weight balance of the bicycle.

(this is why zero-offset posts suck on bikes designed with traditional european geometry and why “fast-forward” style seatpost anmd aerobar set-ups amake for awkward handling bikes.

jerk

KOPS has been challenged before. If one reads this article it makes me think you may be heading in the right direction.

http://www.sheldonbrown.com/kops.html

You mention center of gravity of the rider which migth get you somewhere and be affected by the saddle position.

But you illustration points out a center of gravity roundabout the person’s torso. This matches more closely to the center of gravity of the torso only.

This moment arm of one’s torso center of gravity will undoubtedly make a large difference in how much work your core has to do. Howver, that it only affected by length and mass of the torso and the angle the torso forms with the horizontal.

None of those three things are afected by the saddle position. You can sit forward or back and your torso will exert the same torque about your hips.

Your method miht have somethign to it, but it has nothing to do with moment of inertia of the center of gravity as you have it represented it.

Jon, I’m new here and have seen what some of the other posters do when someone new posts a strong opinion, but I really don’t care. This is honestly some of the most rediculous jibbersh crap I have ever read about bike fit. It’s great that you are doing something to seperate your shop from the others but you just come off looking like a quack when you say some of the things you do.

“I’ve read just about every document available regarding saddle fore/aft, and I’ve seen this method described in almost every one, but have never seen an adequate explanation for it.”
Google Andy Pruitt.

“The cyclist on the bottom has been rotated forward two degrees about the bottom bracket, note that the limb lengths and frame geometry have been drawn completely arbitrarily”
Is the stick man arbitrarily sitting on the rivet or is that where your system puts most people?

“Furthermore, if the cyclist is fit using this technique while generating 175 watts at 90 rpm the pedaling forces are less than what they would be at say 175 watts and 70 rpm.”
Well goddamn, all this time I thought gear mashing used less force than spinning to put out the same wattage. Has anybody from the Nobel Foundation contcted you yet?

I guess I wasn’t clear as to how the final seat tube angle is chosen that the rider leaves the fit with. The method of the fit is a lot like the F.I.S.T. methodology, in which the rider is fully set up at a progression of seat tube angles, and the rider chooses the one that they feel to be the most comfortable and powerful. The C.O.G. measurements are simply guidelines to give the me a little more understanding of what the cyclist is feeling. One of the hardest things about fitting is getting the information out of the cyclist that the fitter can interperet into a change of body position. This is increasingly more difficult with less experienced, less body aware cyclists. In your case, you would probably prefer the position that places your C.O.G. a bit further in front, however if you did not, it certainly would not be pressed on you.

Actually, on a unicycle the rider’s COG oscillates in front of and behind the COG. It’s the same physics of trying to balance a baseball bat on a finger.

I believe you’ve misinterpreted the way in which we use the data. We not only balance the rider over (or close to the C.O.G.), but we also use a very powerful spreadsheet to modify the chainstay, front center, head tube angle, etc. to give a desireable weight distribution and handling characteristsics based on the type of riding the rider is planning to do. Without the scales and COG measurements, we’d be guessing as to where the COG actually lies.

This moment arm of one’s torso center of gravity will undoubtedly make a large difference in how much work your core has to do. Howver, that it only affected by length and mass of the torso and the angle the torso forms with the horizontal.

I absolutely agree, and it is for this reason that I believe that novice cyclists have preferred positions that place their overall COG just behind the bottom bracket. If hip angle is maintained and seat tube angle is steepened, then as you mentioned, the core has to do a lot of work to keep the position comfortable. I think that novice cyclists either don’t have the strength for this or simply don’t feel safe or comfortable supporting themselfs in this way. One of the thing’s that I’ve noticed while experimenting with handle bar drop with novice cyclists is that when they aren’t loving a steeper STA, they often won’t tolerate a lower drop, even if the angle is more obtuse than they preferred in a slacker position. This further proves your point.

Thanks for the kind words.

I’ve read a lot of Dr. Pruitt’s work, and none of it gives a real explanation of the reasoning behind KOPS. I recommend that you read Mr. Empfields writing on seat tube angle regarding tri positions and also the Kevin Bontrager Article that Cerveloguy posted a link to. A lot of people say that the knee should be directly above the pedal spindle at 3 o clock because this is when peak pedaling forces occur. Say that you set someone up with KOPS, and it just so happens that their STA is at 73.5 degrees at this position. Now put them at 74 degrees and raise their seat accordingly to maintain saddle height. Their peak pedaling forces are now occuring 1/2 degree further into the pedal stroke than they were before. Put someone at 78 degrees, as in a tri position and the diferential becomes more extreme. Furthermore, if you do look at a pedaling force analysis for any given road rider, peak pedaling forces usually occur around the 4 o’clock position, but I’ve never seen anyone do KOPS around there.

As for the rivet comment, no not at all. I don’t have all the numbers worked out, but on average taller people prefer slacker seat tube angles and shorter people prefer steeper STAs. A lot of bikes steepen their STAs as you get into the smaller sizes, so it seems to me that either the bike companies are onto this, or it gives similar results to the afformentioned more popular method.

“Furthermore, if the cyclist is fit using this technique while generating 175 watts at 90 rpm the pedaling forces are less than what they would be at say 175 watts and 70 rpm.”
Well goddamn, all this time I thought gear mashing used less force than spinning to put out the same wattage. Has anybody from the Nobel Foundation contcted you yet?

Actually yes, the ceremony is going to be in a few months! The article is attuned to a less knowledgable audience, and I was simply trying to explain that a bit further as well as point out one of the flaws of this method. Ask another fitter to explain their saddle fore/aft methodology and see if they can give you a reasoning for their method, let alone be humble enough to bring up a flaw of it.

not sure if Jon is on to anything, but it certainly is an interesting approach. it may affect bike handling in a positive way. who knows?

but speaking of quackery, the KOPS method is a perfect example of bike quackery.

since you are a fan of KOPS, find me just one decent study that shows the KOPS method verifiably does ANYthing.

Jon, you’re not measuring the COG of the TORSO, you are measuring and adjusting the COG of the rider as a whole. It’s a bit of a leap to say that moving the COG of the rider has some effect on the torso.

You can check your CG with two bathroom scales and someone to balance you over the scales. I believe there is less roll resistance / your faster . I also check that " I try to have my bars in the same sight line as the front axle" while looking down.

Picture rotating a rider about the bottom bracket while keeping the max and mins of the hip angle constant throughout the continuum, just like what is done to the extreme in a tri fit. As you rotate one forward, the back ‘flattens’ out. The Torso’s COG is more or less a fixed point somewhere in the rider’s stomach. The amount of torque (core strength) it takes to support the torso could be calculated by knowing the weight of the torso, the radial arm that goes from the Torso’s CG to the hip and the angle that this radial arm makes with the horizontal plane. If you rotate one forward, the radius from the torso to the hip stays the same, but the angle this line makes with the horizontal plane approaches 1.

Torque = torso weight * radius * cos (angle)

Rotate the whole body forward about the CG and this angle approaches 1, thus creating a stronger demand on the rider’s core strength.

Oh yeah! Well I can do it with ONE scale and a trainer (which we all have, right?). ;^)

I’m thinking that this fit of COG over the BB is going to produce some VERY long chainstays in order to get a good balance on the bike and handling that is anything short of semi-like…

“As you rotate one forward, the back ‘flattens’ out.”

No it doesn’t, unless you lengthen the top tube.

“The Torso’s COG is more or less a fixed point somewhere in the rider’s stomach”

if by more or less, you mean constantly changing as the torso angle changes…

“The amount of torque (core strength) it takes to support the torso could be calculated by knowing the weight of the torso, the radial arm that goes from the Torso’s CG to the hip and the angle that this radial arm makes with the horizontal plane”

Nonsense, torque is the completely wrong term, who cares, and that’s why we have arms, isn’t it?

Could you please, for a moment, step away from the lever analogy and attempt to explain how the methodology you are proposing deals with the rider’s musculature and issues of economy, output and efficiency?

I am pretty uninterested in any fit methodology that is based on minimizing core fatigue; I don’t think I have ever heard a single rider, in the hundreds of fits I have done, complain about this. How do your position tweaks affect the activation of different muscle systems? What is the performance advantage?

.