Optimization choices - strategy?

Just my personal preference, but for ships I always go miniaturization (or increase my ship's capacity). I haven't quite done the math, but I'm fairly certain it's usually the best option. 

Everything else is "it depends".

With the limit of 1 ship/shipyard/turn and 1 improvement/turn, cost reduction techs quickly become useless - hence I end up going for the "improved performance" techs (since maintenance is trivial).  For engines & sensors, I go with mass reduction - filling a ship with low-end drives + mass reduction bonuses is actually cheaper and more effective than filling the same ship with high-end drives using cost reduction.

Weapons have the only real decision to make - choosing range allows superior forces to wipe enemy fleets sooner, minimizing damage, but choosing mass reduction gives your ships more long term firepower.

Due to the ineffective use of military resources by the AI, mass reduction is the indisputable king of optimizations.  Anything you can do to increase the efficiency of your logistics points will pay off in spades.  

Yep, we can easily make planets which can produce one unit per turn, so making that one unit better is the best option by a long shot. Most of these specialization choices are like this. There always seems to be one options which is superior.

There is no such thing as a civilization which is poor at manufacturing. The biggest malus is -25 production. This hurts colony start up and research/economy planets (ironically perhaps). Actual production planets don't even flinch at this penalty.

Priority is always

1. Miniaturization or Capacity

2. Food or population cap then growth. 

3. Improved raw production or manufacturing

4. Improved logistics

5. Improved morale

I rarely play on insane maps but even on huge maps high morale is easy to obtain and it's effectively a fifty percent production bonus. 

Later on you might want to consider that

1. Carriers with fighters can do with improved weapons over miniaturization as the fighters are limited in how many modules they can add but will at least add one module of each weapon type. 

2. Lasers  can always use more range. 

3  kinetics can always use more accuracy. 

For ships, I go for more speed and then space. If you want more power, you can always add another weapon.  I also just go for missiles and no defenses. Missiles are a great standoff weapon.  If I have space for a shield or armor, I figure one more missile should fit there.  

Until I can get carriers. Then it's all carriers. 

That's a big problem I have with the tech tree.  The concept of making you choose between mutually exclusive options is great.  I like being forced to make strategic choices....but the problem is they aren't strategic choices.  As people have mentioned in almost all cases there is a clearly superior choice.  In many instances you can even prove it mathematically.  It's not a choice it's just a trap for those who don't know what is best.  And it's usually obvious which is best.

Doesn't space = speed? More space = more engines which means more speed, right? The choice that comes to mind here is right after Ion engines. Do you take 10% speed, or 20% reduced mass? A player looking to optimize speed here would pick the later option right?

BTW I totally do the no armor, all missiles approach to early ships. I've fitted 7 harpoons on an early military ship, and was just like "Come and get me!" I find it pretty strong so far.

I really hope we see balance improvements soon, more then anything else. This game is all about choices: Race options, tech trees, idealogy, moral decisions, planetary structures. Yet, despite this, I am never actually weighing the choices I face. Their is either a clear winner, or the effects are too insignificant for me to care. (my planet has a -10% research penalty..oh nooo...). I think if I started to care about my choices more I'd be able to get a lot more into the game.