Kontact 

For bikes you don't lean on your elbows to ride, I am a big believer in forward set back limits. I phrase it that way because there really isn't anything terribly wrong with more set back, but moving forward too far rotates the pelvis past the point of working with an approximately level saddle and shifts weight onto the hands. So the maximum forward set back is the point where you are still balanced over the pedals reasonably and your pubic bone isn't burying itself into the nose of the saddle (or your sliding forward from a lowered saddle nose). That forward set back limit should allow the lowest bend to the handlebars, so it is preferred by sporty riders for the aerodynamic benefits. Touring riders who sit up more can be happy further back from that limit.

KOPS is a (now) traditional method of locating a set back for your saddle. In the late '70s, several industry coaches tried to tie bike position to performance, and found that many high performing riders had naturally found very similar stances on the bike. While trying to quantitize these relationships between saddle height, reach, set back and the like; someone noticed that good set back tended to be when the forward knee on horizontal held pedals was over its pedal axle. Lots of refinements to this rule followed, but that's the gist of it - get set back right and your knee will be coincidentally over the forward pedal.

Over time this method gained a life of its own and Dr. Andy Pruit - one of the first sports medicine folks to get into bike fit - created a study that demonstrated that KOPS goes with peak power. That might seem pretty great until you realize that being "over" something is a function of the direction of gravity, which has nothing to do with pedaling. And that's a good thing, or power would drop when we pedaled on an incline or while riding a recumbent. Unsurprisingly, no other bicycle physiologist were able to duplicate Pruit's results. But that didn't stop him from including his version of KOPS as an essential element of the RETUL system.

I trained on RETUL a few years ago, and just a few weeks later ran into the problem: Not everyone's legs are proportioned like 1970s European male racers. I was fitting a Persian woman, and KOPS had her sitting really far back on the bike, exasperating the reach problem many women already deal with. So I moved her saddle forward to a more common location given her seat tube angle and we went from there. It was a good fit. Same but opposite problem with an Asian coworker - KOPS moved him really far forward.

Alternatives:
Fitter Steve Hogg has long advocated for using balance method for finding set back. Essentially, you ride the bike in the drops position, and you try to take your hands off the bars. Too far forward and you can't lift yourself, too far back and you weren't leaned over much. Average them out and there you go.

This is a very reasonable approach. The only thing that I question is that it is all about weight distribution and (to a smaller extent) core strength. Should a guy with a muscular upper body need more set back just because he's top heavy? Should a women with a smallish upper body be moved more forward? What is the effect on pelvic angle? Maybe the big shouldered guy should live with a little extra weight on his muscular arms rather than have to sit back far enough that it limits how aero he can be?

The alternative I've been kicking around for awhile is to ignore the upper body and ignore the leg ratios. Instead, maybe a universal set back angle would be useful - at least as a start. The relationship between your pelvic angle and the location of the pedals comes down to the average pedal location (which is at the BB) and a line running up to the pelvis from the BB. Regardless of what your legs are doing, they average a certain angle, and that average angle determines the corresponding pelvic angle.

Proposal - find the sit bone contact point on people who's set back seems healthy for the pelvic angle and overall fit. Measure that point as the horizontal distance from a 73 degree line originating at the BB. (Why 73? It sits in the middle of most road bike geo charts for seat angle, and several brands have noted that there is no reason smaller or larger bikes should depart from it, as there is no data to suggest taller or shorter people need different proportional set back. So it is a good reference line.) I imagine that we'll find that most people with good fits are perching on sit bones that touch the saddle 2-3cm horizontally behind of that 73 degree line (but that needs verification). And if we start with their saddles around that point and adjust for pubic bone pressure and center of gravity we should quickly arrive at a useful set back that references pelvic angle and balance without falling into the trap of basing everything on shin to femur proportions.

(Keep in mind that every set back adjustment should be followed by a resetting of saddle height. As set back changes and saddle to BB height remains constant, the graph of those two is an arc that curves down and back toward the rear wheel. So they need to be made together.)

I think my approach might be useful in part because saddles don't come in sizes for horizontal length and sit bone location is closely tied to hip joint location. So the size or sex of the person shouldn't cause the horizontal relationship between angle and sit bone location to change much.


If any of that was unclear, please ask. I welcome any thoughts.

elcruxio

I've found that setback is over rated. Typically issues people try to solve with setback corrections are directly related to a saddle that's too high.

Also the last thing cycling needs is another useless formula...

When you have your saddle height set, push around 90% VO2MAX or at FTP for five to ten minutes. If you feel strain mostly in the quads, move saddle back. If you feel strain mostly in the hamstring area or back of the leg, move saddle forward.

It's actually easier to start out with the saddle pushed back as far as it'll go. That way you'll start noticing the balanced strain between quads and hamstrings and you'll notice when you'll start getting unstable / falling forward. But with the correct (low enough) saddle height people can handle quite a large spread of setback positions.

The prerequisites for all of that to work is a saddle that's stable for you at your chosen pelvic angle and not too much reach. Funnily enough too much reach (which is also really common) causes same sorts of issues as too much saddle height.

Kontact 

Thank you for adding another useless formula, based on a change to the saddle height formula - that you don't reveal.

elcruxio

Well that thing I described there wasn't actually a formula. It's a method. A minute yet significant difference.

as to saddle height, there really isn't a good formula for that either. Too many variables to make one work. However there is some evidence that if you want maximum power, around 103% inseam seat to pedal gets you there. But that's a big ask in terms of mobility so definitely not for everyone. And I'm not sure how cleat fore aft, foot length or shoe model are controlled in the studies which got said conclusion.

So no. No formula for seat height.

The METHOD I use is the heel to pedal method. Heel on pedal and if I can pedal without my hips rocking, I'm close to golden. There's some caveats however. With an aft cleat position or a mid foot cleat position the seat needs to come down from the heel to pedal height. Also shoe sole and heel thickness play a part.

But the heel to pedal gets you pretty close and probably not way too high. If it leaves you a bit low, well that's totally fine. A little low is good. A little high means injuries and pain.

79pmooney

I learned a long time ago that I need a position I can ride for hours that is aero and I can do with little loss in power. That leaning forward and closing up the angle between thighs and torso improves aero but limits deep diaphragm breathing and hence sustained power. So, what works is rotating my entire position forward. Seat forward, bars forward and down, more seat tilt. More weight on my hands. I can lift up with no hands but it is marginal (and I have good core strength).

Now, I have to pay close attention to my hand positions or I'll struggle with ongoing injuries. But when I have that right, I get to ride bikes that are a joy to ride, even if I have to spend a few hours straining my little VW bug engine pushing my lightweight and not remotely aero body upwind. For that , I will happily ride bikes that I get told regularly aren't set up properly.

So I ride 6 different bikes, all with similar positions only different rotations. The most rearward (or upright) is on my Raleigh Competition with it's CX brake levers and nice, cushy ride. My fix gears are more aggressive. The commuter a real step more aggressive than the Raleigh. Roughly the same on my Mooney. My avatar photo bike is my most aggressive. Ultimately my '80s pure race bike will match the avatar when I get a stem with a shorter quill that doesn't hit the steerer butt.

And seat setback? It is whatever it is. I measure another bike and copy to set up the first time. (Actually I draw my bikes on a CAD program and compare it to my others - mostly to see what I need for seatpost seatback and stems to get into the ballpark.) And after that, it is riding with the wrenches and stopping as needed to dial in.

Kontact 

Your formula would bring down my saddle height by 5cm, which is crazy low for knee angle and knee damage. The usual number is 109%, which is still a tad low.

elcruxio

Yeah so too low a saddle doesn't cause injuries. A too high a saddle on the other hand... Well you can imagine what happens when you put repeated strain on a joint when the joint itself gets destabilized during every strain cycle.

Extreme end of too low can become an issue but there's usually several centimeters to play with before you get to that point. You'll lose power way before that.

Now the reason why I don't like formulae is because they're unspecific and don't account for pretty dang important variables, like cleat position, shoe model, foot length, etc.. And they should also change with the times. The 109% formula was developed when people were still using toe clips. That makes a big difference in saddle height when compared to clipless. Also do you measure it with shoes on or off? And if one over the other, why?

So formulae are pretty much worthless. Optimally everyone would have a friendly neighbourhood fitter who could record footage and see when the knee joint stays i tension over the full pedal stroke or if it goes slack when saddle is too high. But failing that using mobility as an indicator works almost as well.

Kontact 

I completely disagree. Low saddles cause knee strain because the joint never gets to unload by approaching full extention.

Toe clips and clipless put the foot within mm of the same height above the axle. Your 103% makes changes that are more than 10 times that.

Carbonfiberboy 

Oh boy! Another fit method discussion! I get height by heel-on-pedal. Problem is that's not really precise because it depends on being able to detect hip rocking. My method is super precise - I set my saddle height by being able to pull back hard at the bottom of the pedal stroke. Which sound totally imprecise, but everyone's different in the way their muscles are attached and how they work, so I think it's as good as anything. For first heel-pedal and then climbing hard at a moderate cadence and trying to produce power at the bottom of the stroke. My impression is that, at least for me, that's a very narrow range, maybe 2mm.

Setback, I like my hands to be fairly light on the bars, but some weight in them. That has me with my saddle all the way back, but I think that's because it's 52cm and there's something funny with the seat tube vs. toe strike.

Reach - I like a lot of reach, my upper arms at 90° to my torso with my forearms horizontal when on the hoods.

So I go setback - height - reach. Using this method means my thighs will hit my lower ribs when I'm all the way down, which limits my aero, but OTOH it's comfortable for a 24 hour ride and I'd rather have the latter.

No formula, but oddly enough I'm exactly at KOPS. Whatever.

elcruxio

Toe clips tend to have the foot further back from the pedal axle than clipless and that's a big factor for saddle height.

Now I don't actually use the 103 %, because formulae are stupid. For whatever reason you latched on to that number. But it does apparently give peak power so whatever. I'm at 102% I think, but that's with shoes off. Wasn't a figure I was aiming for but rather I checked it out of curiosity.
Actually it's kinda complicated since I use three pairs of shoes and they all have wildly varying sole thicknesses. And come to think of it the cleat fore aft might be different too since I've just pushed the cleats as far back as they can go.

Letting the joint unload at the bottom of the stroke just seems like a recipe for pain.

Firstly the loaded joint and the whole leg serves a purpose. After the main power stroke the bottom leg needs to support the rider UNTIL the upper leg travels over the peak and begins its own power stroke. If your bottom leg isn't supporting the rider you'll experience pelvic instability, which in turn causes hip pain, saddle sores, numb genitals, back pain, shoulder and neck pain, arm pain, hand pain, hand numbness etc. Some of those are caused by compensating btw.

Secondly (analogies are also kinda stupid in this context but I'll give it a shot anyway), If you unload so much as to let the muscles around the joint relax It's kinda like throwing a ball and letting the arm go completely slack at the follow through. That'll jolt and rip the muscles and tendons and after a while it'll start breaking stuff. If your saddle is too high, that same thing will happen at the knee albeit at a lesser rate. If the saddle is high enough to allow for unloading the leg there's this observable acceleration of the knee joint at the bottom of the stroke and that really isn't good for you long term. Even if the effect is slight, there's so many repetitions that over time it's going to start taking a toll.

I believe the general agreement is that riders should have their knee flexion at 25-35 degrees at the bottom of the stroke. The 109 % formula leaves over half of riders and closer to two thirds outside that value. Since the 109 % is a pretty high saddle height to begin with that'll almost automatically mean that almost two thirds of riders setup with the 109 % end up with a smaller than 25 degree knee flexion at the bottom of the pedal stroke. You'll need some freaky mobility to be able to pull that off with an intact pedal stroke. I suspect that riders who fell within the specified range had their cleats quite far forward on the shoe.

Personally I feel that aiming for 25 - 35 degrees knee flexion is a bit tricky as people'll try to achieve middle of those values when many riders would function better near the 35 degrees, especially if you work sitting down. But of course even degrees are stupid in isolation as they do not account for mobility, cleat position, pedal stroke, ankle flexion, or how the rider is feeling.

Kontact 

109% is barely 35 degrees. I did this for a living and checked the math many times.

I wrote the OP because I was hoping for a discussion of whether the way I was thinking about set back was considered useful or not. Instead you decided to post that nearly everything I do as a fitter is garbage and people should ride bikes like Sean Kelly. I apologize for being brusque, but I really wasn't looking to throw out everything I've been doing for years in favor of squatting low and a long way from the handlebars.

elcruxio

You probably shouldn't post if you take disagreements personally.

I just had to verify the math on that 109% figure. Using my own measurements I got 32 degrees knee flexion if my foot was pointing 45 degrees heel up. With a more reasonable 25 degrees heel up I couldn't reach the pedal. With 30 degrees heel up I got a crisp 10 degrees knee flexion. I once tried 15 degrees and boy that wasn't fun.

perhaps that 109% figure works with the cleat all the way forward on the shoe right under the toes with the rider ankling down like crazy, but I wouldn't call that comfortable, efficient or stable. Personally I prefer far less heel rise and cleats placed aft. That's more stable and more energy efficient as there's far less calf usage.

High saddle also isn't very aerodynamic so there's incentive to try to go lower.

using the same numbers with the 103% I got 35 degrees knee flexion with an even more reasonable 20 degrees heel up. Effectively that gives me space to drop my heel pretty far down on climbs or when grinding is otherwise necessary.

Oh btw, my knee issues disappeared after I stopped using formulae and just dropped my saddle where it's comfortable. And I can now ride centuries which is a nice bonus.

Kontact 

I doubt you are using the measurements correctly. There is no way that a 32 degree knee bend at a 45 degree heel angle goes to not being able to reach the pedals when you lower the heel 20 more degrees. You would still have about 20 degrees knee angle.

If your technique for measuring this sort of thing is not done right, your conclusions about how things work aren't very useful.

elcruxio

I have pretty large feet. The difference between 45 degrees and 20 degrees is quite a few centimeters. The 109% formula puts my seat around 6 centimeters too high. It's actually three centimeters higher that the Lemond method which is also too high.

Also using the measurements is just trigonometry. There's nothing mystical about it. I could have used made up values or just split the inseam value in two and the results would have been similar.

Kontact 

So your cycling inseam is 99cm?

elcruxio

Pretty close. 96.5cm. But with shoes on it's fairly close to that 99cm mark.

PeteHski

I've experimented quite a bit with saddle setback over the last year since I got my Kickr Bike - as it's easy to change very quickly.

I've tried Steve Hogg's balance method (which appears to be quite popular among UK fitters), but that didn't really nail it down to any specific position i.e. I feel "balanced" anywhere over at least a 50 mm range. So while I like the concept, ultimately I didn't find it all that useful in setting a specific saddle setback.

What I have found is that if I go too far rearward (saddle height and reach corrected) then my pedal stroke feels more laboured. I find spinning a relatively high cadence with power easier and more natural with a relatively forward position. This reminds me that I haven't actually found my limit yet on moving the saddle forward. I still have plenty of scope to move at least another 10-15 mm on the rails. This is all relative. My saddle setback is still quite rearward compared to the current race trend of zero setback posts and saddles slammed forward.

I can't really comment much on your proposed formula. It sounds tricky to measure and I'm not convinced it's going to give a "Universal" answer. There was a long thread somewhere on another forum (more race-oriented) discussing how much saddle setback riders actually use. They were all over the place, with no obvious pattern. Although measuring setback from saddle nose across a wide-range of saddles is not going to be apples vs apples. The OP tried to get people to measure setback relative to their sit-bone contact point, rather than saddle nose, but you can imagine the response.

So my thoughts are that a formula may not be applicable to saddle setback. Maybe it's just one of those things that requires individual trial and error. From my own experience it isn't a particularly sensitive parameter for me anyway. For me it's very much second order to saddle height and it's hard enough coming up with a "Universal" formula for that.

GhostRider62

Because they can use the same mold on the rear triangle for all frames and lie to us that it does not matter.

At 6'3 , frame setback is always a really big issue for me on a 73 degree frame. 74 degree STA on a road bike? No chance, the rails would be slammed and I'd need a 150 stem and my weight would still not be balanced over the pedals.

WRT 0.883 or 109% ratios, they are stupid. Crank lengths vary. Foot lengths vary. Pedaling techniques vary as does cleat mounting position. A huge variable is the measurement from the crotch to the floor. Who sits on their tailbone? Tailbones vary a lot. Ischial tuberosity to floor is how it should be measured.

Balance and knee angle are probably better approaches with hand comfort a major consideration for club riders. So, your approach tends to resonate with me but I doubt any formula will work.

I just did a $450 fit. They did not measure a single thing on me. Nothing. Just eyeballs and video analysis.

oldbobcat

If you're getting a usable range of 5 centimeters (that would be about two inches), you're probably not doing it right. What is your cruising torso angle, or what is your handlebar height in relation to the saddle?

Carbonfiberboy 

I don't use formulas either, but rather the balance method. I try to move my saddle back until, pedaling normally, I can briefly take my hands off the hoods without changing my forward lean or sliding forward on a level saddle. That's only about reducing hand pressure and thus hand and arm fatigue, really only an issue with long rides, but that's what I like to do.

I find that pedaling issues are a function of training, not position. Obviously, one can become efficient wherever the pedals are or recumbents would be unusable, a quickie reductio argument.

Kontact 

To address your statements in order:
That is not how bike frames are generally made. Many carbon frames clearly have different stay lengths and seat stay angles. Moreover, the vast majority of frames do have variable seat tube angles. I named the only two exceptions I'm aware of. So I'm unsure what you're talking about.

I'm 5'4" and 74 STA also locates the seat on the front of the rails to get reasonable set back. That was my point - few seem to benefit from steep seat angles. Some folks need even more relaxed than 73, but 73 is a good average angle across side ranges.

The formulas can't be perfect, but it is important to remember that while crank lengths vary, the center of rotation doesn't. It is the average of all pedal positions. Lowering your saddle to accommodate a long crank arm means that you have doubled the hip and knee bend at the top of the stroke. There is no data to suggest there is only one right way of setting a saddle height for varying crank lengths. And, foot length is only a factor if you pedal toe down or have feet dramatically out of proportion to leg length. Otherwise, foot length and angle are part of the formulas - since they were derived from observation of cyclists. And the saddle to crotch measure is not to your tailbone. It is a measure to the soft tissue adjacent to the hip joints, and that relationship to where you actually sit is contained in the formula. Most people do not have radical difference in the location of those two spots on the groin, so the formula is taking that into account.

Fitters who don't measure angles are still measuring angles - they're just doing it from long familiarity. I don't do it because I like to confirm that what I'm observing matches a metric and I'm not fooling myself, but I don't have a decade of experience.

I can certainly see why pat formulas don't sit well with anyone. But I do think that we should acknowledge where the formulas come from (observation), and that the point of bicycle geometry is to get people sitting more alike than different. My thought is that we get wrapped around the axle with rules of thumb and kind of forget that you can make things too bespoke. Saddle height formulas provide good starting points, and maybe such an approach would work with set back.

79pmooney

I take my setback very seriously. Have two custom posts with 65mm of it. And that SR MTB post with a mile of adjustable setback. The lugged steel Nitto post with quite a lot. So I could take it personally being told it is over-rated!

My reason is the bikes that have the best weight balance between the wheels for me have short chainstays. Long or even medium chainstayed bikes get light in back when I corner on rough downhill descents. If I want anything like tire clearance on the bikes that handle well under me when I have the bike fitted for my best position, I need a bike built with a steep seat tube angle (or a curved seat tube which I might ask for if I get another custom). So, with that 74 or 75 seat tube angle, I need real setback to keep from slamming ti rails (my preferred seats all have them) all the way back just to get a "normal" setback. I've already broken at least one slammed ti seat slamming the seat on a "regular" post.

In fact, I really like seatposts where the clamp is roughly centered on the rails. Now Ican do mid-ride seat placement adjusts and not ever hit or even get close to the limits.

As I said in an earlier post, I measure the setback off one of my other bikes to get the ballpark when I set up a new bike, then go ride with the wrenches. When I buy the frame (built up or nor) I draw it up on my CAD program and see what stem and post I need to but bars and seat where I need the. Set both, the bars and levers using the other bikes or the drawing and go for that ride. (No handlebar tape yet. Levers almost certainly have to move.)

I guess I should start feeling sorry for myself, leading this life with so many and such large setbacks but I'm committed to mustering on.

elcruxio

I went from the setback being the most important bikefit metric and still having constant issues and injuries to it being of almost no consequence, no issues and no injuries with added mileage.

I almost went and got a custom frame with a 70 degree seatpost.

My former touring bike had 72 degree seatpost angle, a 25mm setback seatpost and all the saddles I tried slammed as far back as they would go. With Selle SMP's that's quite far back.

My current road bike has a seatpost angle of 73 degrees and I'd honestly need a zero setback seatpost, because the saddle is slammed forward. But when I got the current seatpost there were only 18mm setback posts in stock so...

Now setback for me isn't as much a body balance issue anymore. It's more related to kinematics and muscle strain balancing. But the functional setback area I can use is wide indeed.

And the main changes I made was lowering the saddle height from what it should have been according to all the formulae of yore. Then I slammed my cleats all the way back and had to drop my seat a bit more.

I'm starting to think I could easily ride road bikes with even 74 degree seat angles. Luckily I have my dream bike so no upgrades necessary.

PeteHski

My back angle is 46 deg. Bar drop a modest 50 mm to the tops. Basically an endurance-minded fit. I've done this balance test many times and just find it too vague to define a specific position. I have quite a slim upper body and a fairly strong core, so maybe that's a factor? I had better results defining my setback based on how smoothly I pedal. My starting point was simply the middle of the saddle rail with the stock seatpost. My seatpost happens to have an offset flip-head so you can flip it over changing the post setback from 25 - 13 mm. I marginally prefer the 13 mm position with the saddle rails still centred. But it is not a big deal for me either way. When riding, my hands feel comfortable on long rides and the bike feels well balanced. I don't find myself edging forward or backward on the saddle, so I guess it's all good.

GhostRider62

Interesting comments, unfortunately most of them are made to statements that I did not make. For instance, I did not say, "Fitters who don't measure angles are still measuring angles" ....I said, "They did not measure a single thing on me...." meaning they did not care about static measurements like distance from tailbone to floor or length or arms, etc. Not a single static measurement.

WRT foot length, toe angle, and cleat mounting position, it most certainly matters at the top of the stroke although not as much at the bottom of the stroke. Like most fitters, you are wrong.

I'm glad you have a decade of experience.

You are 5'4''. When you get to be more like 6'4'', you might appreciate why 73 degree can be and is often a bad starting point. Pinarello and Trek for instance do understand that, they have more like 72 degrees on big bikes. That was my point. Obviously, some prefer to be more over the pedals and might have a shorter torso.

You are telling me the back half is different from size to size? Interesting. The STA is identical as are the chainstays on all three of these very popular bikes. Are you not aware how costly molds are or that using the same mold for the back half saves a lot of money? Or that manufacturers don't like to save money.

https://www.canyon.com/en-us/product...etry/?pid=2893

https://www.cervelo.com/en-US/bikes/s5

https://www.bmc-switzerland.com/intl...c-grey-23.html