How To Control The Eager Beaver

No, you are correct, RB, there should be no skating force with a tangential tracker.  The only side force would be to overcome any friction in the carriage, and there is always some of that, which is why the best LT tonearms in my opinion float the carriage on an air bearing.

And you've answered the question.  Warps throw the balance of skating force vs AS out of whack, momentarily. I'd never thought to worry about that, so I won't start now.

The issue with tangential tracking arms of the air bearing or ultra low friction variety (e.g., Clearaudio) is that the arm does not have the mechanical advantage of a pivot/fulcrum, so considerable force is imparted on the stylus/cantilever to pull the arm assembly sideways across the record. 

In some cases that force is much higher than skating forces and there has been reported damage to cartridges.  There are ways to make pivoted tangential tracking arms.  The earliest used a sensor to detect when the arm is no longer tangent to the grooves and a motor to then move the arm/pivot assembly to again achieve tangency.

There are modern versions based on Thales circles, that has a conventional arm attached to a base that rotates to move the whole arm into a new position to achieve tangency (Reed T5, Shroeder LTA), these arms achieve tangency and virtually no skating forces while minimizing other adverse consequences.  

There are also straight arms that have not offset angle at the headshell.  The tracking geometry is WAY off with these arms, but, skating force is substantially less than that of arms with an offset angle.  I have heard some examples and I was quite surprised that the lack of any attempt to get close to tangency of the cantilever to the groove did not seem to hurt performance much.

Larry, That is an interesting idea, but I don't get it. The force applied to the stylus/cantilever would be to counter the force of friction between stylus and groove, and here, with a tangential arm, we have a cantilever that is always tangent to the groove across the entire LP. So the friction force would only vary according to groove tortuosity, ideally.  The carriage end that holds on to the arm wand has the same job as the pivot bearing does in a conventional pivoted overhung tonearm, to resist friction or stylus drag. For much of the time (except for two moments when the stylus passes through the two null points on the surface of an LP, if your pivoted tonearm is set up properly), with a conventional arm, the cantilever is not tangent to the groove, which imparts a side force on the cantilever. I would think that's "worse" than with a tangential arm.

Properly designed tonearms that have zero headshell offset also underhang the pivot. That's a huge distinction from tonearms with an offset headshell that overhang the pivot. In order to minimize TAE, Lofgren, Bearwald, Stevenson, et al, introduced the overhung tonearm and headshell offset, regardless of the effect on skating force. You are correct that the resulting high TAE of an underhung straight tonearm does not result in audible distortion. But you say it doesn't hurt performance "much".  I'd say it does not hurt performance at all, and in fact there is much to be said about the at least equally excellent performance of underhung tonearms with zero offset. (And I've said it elsewhere so won't bore you with my rationale.)

An air bearing tangential arm is not subject to  skating force.  But, in order for the stylus/cantilever to drag the arm along the air bearing, even with little friction, a lot more force is required, in part because of the high horizontal inertial mass of such arms, but many because of the loss of mechanical advantage of a fulcrum/pivot arrangement.  Whether or not this was the cause, I know of a couple people who insisted that they snapped cantilevers on some delicate cartridges when using a Walker table with air bearing arms.

I use the term "much" in describing the effect of the tracking angle error of the straight arms, because I could not discern any obvious negative sonic attribute to such design, even though calculations show that there is higher distortion levels from that geometry, but I could not say for certain that there was not "any" negative attributes.  I should have been clearer on this.  I am impressed by such arms.  If I were to get another arm, I  would seriously consider this type of arm, specifically the Viv arm.  It sounded very dynamic and lively as compared to conventional arms.