I guess this is a question primarily for Slowman, but figured there’d be enough of you with thoughts/opinions on it that I’m posting on the forum.
Per Slowman’s latest and greatest article on the topic, some measurements are taken in cartesian (ie x/y, stack/reach) coordinates, and others are taken in polar coordinates (angle, distance). Examples:
cartesian: frame stack, frame reach, saddle fore/aft,
polar: saddle height, cockpit
In what seems to be an extra level of complexity (for me, at least!), some are derived from a combination - for instance, handlebar elevation (which requires an angular measurement plus an x-y measurement to determine the ‘Y-coordinate’ of the saddle, and a second x-y measurement to place the ‘Y-coordinate’ of the handlebar.
What’s the reasoning behind using the different coordinate systems for different measurements? Are certain components easier to fit using one coordinate system vs another?
Just curious, as I translated my fit from one bike to another using cartesian coordinates exclusively (ie saddle stack, saddle reach, pad stack, pad reach, shifter stack, shifter reach), etc)…
interesting!
I suppose one issue is if the question is - “what frame will support my position?”
then to answer it most easily, you want to know your effective seat tube angle, because bikes just so happen to accomodate massive ranges of seat stack and reach, but a much smaller variance of seat tube angle.
So if you know the STA, stack, reach, you have 3 numbers that tell you if a bike works or not
If you used saddle stack and saddle reach, each bike would need to specify a 2 dimensional chart to relate all of the supported saddle reaches at any given saddle stack setting.
Just curious, as I translated my fit from one bike to another using cartesian coordinates exclusively (ie saddle stack, saddle reach, pad stack, pad reach, shifter stack, shifter reach), etc)…
I do exactly the same thing and just use X and Y coordinates centered on the BB.
To quote an old engineer friend, “Pick a coordinate system and stick with it!” 
… because bikes just so happen to accomodate massive ranges of seat stack and reach, but a much smaller variance of seat tube angle…
Huh?
I dunno how to clarify it any more.
seat can go up infinity high, or really low, and the reach changes as you change the height (on many bikes)
Seat tube angle is a particularly tricky one to define, actually - because the effective angle from the BB to the center of the saddle really is subject to a lot of variables - including the actual seat tube angle, but also the seatpost insertion (in the case of non-axial seattubes like on the P3), seat clamp position (in the case of most modern TT/Tri bikes), and saddle rail length.
But, to Tom A’s point… picking a coordinate system and sticking with it makes a lot of sense… so why do we mix and match?
Seat tube angle is a particularly tricky one to define, actually - because the effective angle from the BB to the center of the saddle really is subject to a lot of variables - including the actual seat tube angle, but also the seatpost insertion (in the case of non-axial seattubes like on the P3), seat clamp position (in the case of most modern TT/Tri bikes), and saddle rail length.
But, to Tom A’s point… picking a coordinate system and sticking with it makes a lot of sense… so why do we mix and match?
Because they are used to seats being attached to a tube that’s inclined from vertical :-/
As you point out, seats don’t necessarily have to be attached/located that way. So, each frame should have a “point cloud” of possible seat locations, much like we’ve seen lately with bar/pad locations for frames with integrated front ends.
But, to Tom A’s point… picking a coordinate system and sticking with it makes a lot of sense… so why do we mix and match?
I’d say mostly because Cartesian is easiest, but saddle height is sacrosanct and carries over with little change between road and tri. We try to make up for that with a saddle setback from bb to convert it back to the Cartesian world.
But, to Tom A’s point… picking a coordinate system and sticking with it makes a lot of sense… so why do we mix and match?
I’d say mostly because Cartesian is easiest, but saddle height is sacrosanct and carries over with little change between road and tri. We try to make up for that with a saddle setback from bb to convert it back to the Cartesian world.
I don’t look at saddle height as “sacrosanct”, but more as a “quality check” for the saddle X-Ys.
And once you start using point clouds, the average consumer’s eyes will glaze over =)
Seat tube angle is a particularly tricky one to define, actually - because the effective angle from the BB to the center of the saddle really is subject to a lot of variables - including the actual seat tube angle, but also the seatpost insertion (in the case of non-axial seattubes like on the P3), seat clamp position (in the case of most modern TT/Tri bikes), and saddle rail length.
But, to Tom A’s point… picking a coordinate system and sticking with it makes a lot of sense… so why do we mix and match?
Because they are used to seats being attached to a tube that’s inclined from vertical :-/
As you point out, seats don’t necessarily have to be attached/located that way. So, each frame should have a “point cloud” of possible seat locations, much like we’ve seen lately with bar/pad locations for frames with integrated front ends.
And once you start using point clouds, the average consumer’s eyes will glaze over =)
This is Slowtwitch. Nobody here would ever admit to being ‘average’.
And once you start using point clouds, the average consumer’s eyes will glaze over =)
Seat tube angle is a particularly tricky one to define, actually - because the effective angle from the BB to the center of the saddle really is subject to a lot of variables - including the actual seat tube angle, but also the seatpost insertion (in the case of non-axial seattubes like on the P3), seat clamp position (in the case of most modern TT/Tri bikes), and saddle rail length.
But, to Tom A’s point… picking a coordinate system and sticking with it makes a lot of sense… so why do we mix and match?
Because they are used to seats being attached to a tube that’s inclined from vertical :-/
As you point out, seats don’t necessarily have to be attached/located that way. So, each frame should have a “point cloud” of possible seat locations, much like we’ve seen lately with bar/pad locations for frames with integrated front ends.
Well shit…then we might as well just go back to sizing bikes by top tube length then, huh? :-/
PLEASE NO NOT THAT! haha
stack and reach and STA is a great compromise.
Pretty rare that listed STA doesn’t tell you enough, given the range of adjustment available on most bikes.
So we could get extremely pedantic about it, but to very little pragmatic effect.
PLEASE NO NOT THAT! haha
stack and reach and STA is a great compromise.
Pretty rare that listed STA doesn’t tell you enough, given the range of adjustment available on most bikes.
So we could get extremely pedantic about it, but to very little pragmatic effect.
To be honest, I rarely ever worry about STA on bikes when picking a frame since they all pretty much have a fairly wide range of adjustment (and are basically centered on 73d for road bikes), and if I’m setting up something with the saddle really far forward like for a TT or Tri bike, I’ll probably be using an Adamo anyway, which can get me in a proper TT postion even on an S5 road bike.
I pick a frame based on head tube stack and reach and trust that the saddle will be a non-issue as long as the designer doesn’t do anything super odd back there…
Well shit…then we might as well just go back to sizing bikes by top tube length then, huh? :-/
Yeah - my point exactly. If stack/reach is such a great system for comparing frames (which it is) by locating the top of the head-tube (and therefore the cockpit point-cloud), then surely it ought to be a good way to describe the saddle ‘point-cloud’…