HoB, couple of astute points there, yes, in regards to metal removal it is only a minute amount necessary if you focus it right at the edge, which a steel does, so if it removed even a little it might produce results, and thus you could argue that when steeling this maybe the effect. However this logic doesn't hold for two reasons :
1) you can see the deformation before and see it gone after, you really can't see metal removed on that scale (x10 mag, bends are fraction of a mm)
2) if it was metal removal you would expect it to continue, that is 10 passes should be more productive than 2, however steeling either produces results right away or does nothing
After the deformation has been fixed, which happens in the first pass or two, the performance does not keep increasing *unless* of course you really lean down and actual work harden and smear the blade out, JJ has pictures of this in his book. This makes the edge really weak however and it will break off.
As for Spyderco, I would have to know exactly what was done, in general though there is a lot of CATRA testing performed and that is highly dependent on wear resistance because the blade ie held rigid and thus lateral forces (which cause ripples) are minimized and the force just goes right into the blade, perpendicular.
Unless you are a robot, you can't use a knife like that, even if you are really skilled there will be lots of side play and thus deformation sets in and the edge blunts. This is one of the reasons that if you keep doing work with a knife the edge retention keeps increasing.
Try cutting rope with the same blade for example, and you will note (if you have not done a lot of it before) that with each sharpening, over time, you tend to be able to cut more rope, this is just due to you getting better and better and getting the force to be perpendicular.
As a really extreme example of that, Buck did CATRA testing on their IonFusion models, it scored retardly high, but in field use, the edge retention could not be matched to nearly anywhere close to the CATRA tests, why, because CATRA tests would induce a "rat's tooth" sharpening effect which will not be productive for a person.
Aside from the theory, Alvin Johnson and Mike Swaim looked at this extensively back in pre-1996 (it was done when I first came on), there was a lot of testing of plain 1095 at 65+ HRC against a lot of knives with high wear resistance and they all failed to last as long as they were a lot softer. However when both were combined (M2) the edge retention when up again.
Specific to Alvin's work, some of the steels like ATS-34 had a problem forming edges at the low angles he used (~3-5 degrees) due to problems with grain size.
-Cliff
