You’re an ME right? Do a quick set of drawings in CAD (2D is more than fine for this) with a few different stem lengths (say 80 mm versus 120 mm) while holding the elbow/hand position fixed (relative to your headset pivot point, do it at 20 cm c-c width and 40 cm c-c to model aerobars vs handlebars too) and then tell us how much the stem length affects things. Perhaps a difference between 0 and 15 degrees steer angle? I’d love to, but I no longer have access to the software to demo this, shouldn’t take more than 5 minutes to draw up.
I’m betting it makes no difference, because the true handling effect is between where you weight your hands/elbows and the pivot point drive everything. One may need a short/long stem in order to put ones hands/elbows in the right spot, but the actual stem length is pretty much immaterial.
I missed this by a week or two, but as I believe I previously posted, I would suspect that stems within a certain length range won’t have a tremendously noticeable effect on the bike’s handling, but that’s qualitative stuff right there. I was AeroE, but I think I still have access to and ability to use Solidworks were it warranted here–it is a pretty simple geometry problem though, and Dan actually already did an article on this a while back. The stem length does affect the length of the steering lever arm, but if you presuppose as you do that the hand position does not change relative to the steering axis (a somewhat silly constraint…the rider’s hands are intended to be fitted on the hoods, so as the stem length changes, the distance between the hoods and the steerer tube changes) then the effect would basically be negligible, since it’s the distance between the hands and the steering axis that matters. The problem is that your supposition is flawed…the person will not maintain the same hand position (in coordinate space) relative to the steerer tube as the stem length changes–the hands will either move closer or further, but only by changing the grip on the bar (a somewhat irrelevant consideration since you would never fit someone such that their hands rested inches behind the intended position on the hoods) will the hands not move when the stem length is changed. Basically, the stem length definitely has an effect on the length of the steering lever arm, which in turn changes the way the bike handles a bit. Whether or not the effect will be noticeable (or positive or negative, if it is noticeable) is much more rider-specific. Dan has attempted to quantify the factors that will determine how well the bike handles, so when a stem length change is made, the change can be viewed in that context and it can be primitively determined whether or not the change will result in noticeably better or noticeably worse handling. As incredible as this may be, Dan described it much less verbosely than I did in the article I link below.
Look at the pictures…you can tell that the the stem length changes things. It is easy to determine how big of a change in geometry there is given some “example stems,” but the resulting “handling” of the bike is probably best described in the qualitative realm. As for the angle of the stem, the only thing that really matters is the projection of all the angles into the body-centered x-y plane (x axis projects out the front of the bike, parallel to the ground, y axis is perpendicular to the x-axis and points out to the right side of the bike). Intuitively, a stem with a very high angle has a very short effective length if you “flatten it out” by looking at it directly from above and only view it within the context of my so-called xy plane). The effect of changes in the stem angle on the effective stem length changes dramatically with angles between say 30 and 70 degrees. In other words, the effective length difference between a 5deg stem and a 10 deg stem is much less than the effective length difference between a 50deg stem and a 45 deg stem |(cos(10)-cos(5)| << |cos(50)-cos(45)|. Even though the difference in angle between the two stem pairs is 5degrees, there is a factor of 5 difference in the magnitude of the change in effective length.
Dan’s articles:
Steering Levers
Less relevant to our discussion but still a good read.
Steering Geometry