Most of this could be applied to any strategy game (C&C etc), for example being able to build efficiently and Micromanage.
I would say that games like TA/Sup Com would seem more "strategical" since they tend to be bigger in the sense that they have bigger maps and bigger , more complex tech trees. You also have more options in terms of setting up queues and automated behaviours for units (patrolling,ferrying etc) which means that you don't have to worry about revisiting your base as often because you have already queued up everything you want built for the next 5-10 minutes. It also allows options such as going into a resource deficit to produce units you cannot afford (but will be built more slowly).
I suppose this could be said to reduce the importance of the "micromanagement" aspects of the game (although these can still be important) but I'm not sure if this is necessarily a bad thing in itself. Professional SC looks to me more like it is being played on the basis of quick reflexes and drilling build orders quickly in a sense closer to an action game where you control a lot of units rather than "strategy" as such, perhaps tactics is a better word.
To me it would seem that as technology advances strategy games should become less about micromanagement, since in a real war the General does not usually micromanage the actions of individual soldiers on the scale of "move left, now fire you gun" etc. Of course a general may make high level decisions such as how to train their soldiers which will affect how they behave in the field etc and this could be simulated with AI.
One of the things he mentions about SC1, if I remember correctly is the limited number of units. If the possible composition of your army increases, more ways of using the army becomes available. This becomes a detriment because players spend more time exploring composition than actual strategy.
With regard to your point about automating behaviour, I think this is a more philosophical question. How much automation should an RTS have/allow? Given SC1's history as a mechanically intensive game, SC2 attempted to lower the basic skill level needed to reach a broader audience, but this demand is in constant struggle with the professional level of SC2.
On micromanagement, the SC2 community seems to regard this as the easiest way for an e-sports/SC2 newcomer to identify skill. Long-term strategies and the steps/timings that go into setting them up are fairly guarded secrets among professional players. While some can be particularly obvious, building a specific army at a specific time for a specific reason can be difficult to explain to a newcomer.
SC2 and SC1 certainly do not reflect a "real war," because experience seems to indicate an e-sports RTS requires elements that would not belong in a "real war." I think this is a philosophical decision and you might disagree with it. SC2 was an attempt to capture the popularity of SC1 for a broader audience, not just S. Korea. To do so, the developers attempted to incorporate the aspects that they thought would be most popular both to play and to watch.
> Professional SC looks to me more like it is being played on the basis of quick reflexes and drilling build orders quickly in a sense closer to an action game where you control a lot of units rather than "strategy" as such, perhaps tactics is a better word.
That's a naive view. Flash, the best Starcraft player ever, was not known for being a particularly fast player. What makes him so good is his decision making. It's not just how fast you play, but WHAT you spend your limited attention on that separates a champion/pro-player from your average top-tier player. Also, the fast-paced nature of Starcraft is incredibly engaging for the viewers--there's always something to focus on, from the large big battles, to a small harassment play, to armies simply jockeying for position.
I would say that games like TA/Sup Com would seem more "strategical" since they tend to be bigger in the sense that they have bigger maps and bigger , more complex tech trees. You also have more options in terms of setting up queues and automated behaviours for units (patrolling,ferrying etc) which means that you don't have to worry about revisiting your base as often because you have already queued up everything you want built for the next 5-10 minutes. It also allows options such as going into a resource deficit to produce units you cannot afford (but will be built more slowly).
I suppose this could be said to reduce the importance of the "micromanagement" aspects of the game (although these can still be important) but I'm not sure if this is necessarily a bad thing in itself. Professional SC looks to me more like it is being played on the basis of quick reflexes and drilling build orders quickly in a sense closer to an action game where you control a lot of units rather than "strategy" as such, perhaps tactics is a better word.
To me it would seem that as technology advances strategy games should become less about micromanagement, since in a real war the General does not usually micromanage the actions of individual soldiers on the scale of "move left, now fire you gun" etc. Of course a general may make high level decisions such as how to train their soldiers which will affect how they behave in the field etc and this could be simulated with AI.