Well, considering that most of those posting here have many thousands of hours of successful flight experience behind them, one has to assume that there is more than one way to skin the cat, so-to-speak.
One difference that perhaps migrates into our collective experience lies with propeller slipstream effects, or lack thereof. I believe there was one poster who reflected this in his past experience; it is highlighted by both Webb and Davies in their books when addressing this topic. Abzug, in Airplane Stability and Control, explains that propeller slipstream, in addition to providing nearly instant lift increase with zero speed change, can also influence longitudinal stability.
Abzug also points out that low aspect ratio wings, such as those used by carrier borne aircraft, are notoriously difficult to manage when power is used to control airspeed. The problem lies with the drag curve and where one is on that curve during the approach. In carrier ops, the approach speed is below the minimum drag speed; in air carrier ops with a much larger aspect ratio, it is not. This is one of the reasons that naval aviation opts for a firm policy of using a constant angle of attack to establish airspeed, with power used to control altitude. In fact, the power control issue is so important that throttle control systems must meet specific requirements for carrier operations. Ultimately, the problem of carrier operations seems to have led Lockheed to develop direct lift control for the S3 Viking, which of course later appeared on the L-1011.
An aspect of this issue that I have not seen mentioned here is the problem of pilot gain, or ability to track and respond to deviations. Gain is influenced by a whole range of things, including instrument response, mass and momentum, the relationship of the thrust line to the center of gravity, and the different thrust responses from constant speed turboprops to pure jets to high bypass jets. All of these things are considered, in contemporary designs anyway, in order to keep the pilot “in the loop” and able to make small but precise changes.
Some have said that this is not that complicated; actually I believe that it is extremely complicated, but fortunately most of the complication is removed by the designer. The result of a lot of good engineering is that most of us never fly anything with nasty characteristics, allowing us to employ a range of strategies as we have discussed here.
That said, it would seem prudent to be cautious when comparing manual flight techniques to autoflight principles. Depending on aircraft and autoflight generation, the autoflight system is capable of a much higher gain than the pilot. It is not uncommon for the autoflight system on the 757/767 that I fly, when properly maintained, to detect and respond to a deviation before it can even be seen through the flight director system. This high gain allows the autoflight system to employ one strategy to keep the four forces in balance, while the pilot might get better results using a different strategy.
It truly is unfortunate that our primary education on this tends to be lacking, leading us to import one “ironclad” axiom from one class of aircraft into another class for which it might not be appropriate. I suspect that between the carrier pilots, the large turboprop pilots, the pilots who learned on pure jets and the high bypass crowd, we have an amalgam of ideas and techniques that get passed around without the necessary differentiation.