Wild take (Strawman Units)

Strong stuff doesnt cause stalemates, it breaks them. #IWinButton

Even if everything else is strong it becomes an outrange/outspotting game.

Ano no, the strongest in zk is juet the one whos effective apm is closest to an anthill in rl.

Then go play BAR, that's the game you described.

I'll stick to Z-K

Thinking about the "real" version of the unit is useful on some level because there is a certain amount of suspension of disbelief for unit designs. It makes sense for units to be designed to be good at the things they are good at. Otherwise, you can end up with units that seem to be deliberately misdesigned from an in-world R&D perspective, and there is some threshold where this makes the unit feel bad.

I don't really agree with the OP though. You could just as easily say that "real" ZK had 1000s more units, but the units which were too ineffective to be useful were removed, and we've been left with the strongest ones. There isn't really anything distinguishing this from the claim that the ZK units are the weak ones. I don't get the comment about unit costs, it seems clear to me that a unit that is more capable is going to cost more, so how are costs an arbitrary weakness?

The arbitrariness of a weakness is often up to the details of unit art and implicit lore. Why can't Fencer fire while moving? Because the missiles have to be unstowed to fire, and presumably the truck is too unstable if it moves without stowing. Why is Skuttle so bad? Whatever it is that enables such a deadly explosion must be heavy and emit a bunch of radiation that makes it difficult to cloak. For game balance to make some sort of realistic sense, you should be able to come up with a rough estimate the weight, cost, and vulnerability (ie difficulty of applying armour) of each weapon and chassis. This will explain things like "Why don't we mount a Lance on a Flea?".

SEE ISTROLID FOR PRO UNITS DESIGN!
Every part costs money, you can't make value/money higher by bigger unit, but to choose which value/money is high and which is low.
Different values has different efficiency when cooking each other, e.g. high range cooks low speed, beam cooks high speed.
Normally, a unit with even values are weaker than any kind of unit with pros and cons, see leggy nuke.
Also, a unit with super bias may win easily (even with high value differ) vs some units, while also lost easily vs some, see moderator, flea.
Pros and cons and advantage between them are where strategies of units grow.
They are not fully correct in zk, due to some not pretty features (1hp full weapon force, aoe disaster), while this is general rule of units design in all games.
istrolid live when
BUT, batteries(ammo, stardust heat) really makes units different.
Play istrolid, don't try to cover energy use by reactors, you'll find some of your units becomes much stronger than before.
If your reactor are not generating energy, you are wasting, while some of your units may just shot 1s and ded.
Then see zk, imaging that glaives are using 1x1 reactors instead of battery, while if they shots 10s only they can be 45m cheap (if not for raiding).
Some units in zk is weaker version of the batteried!

i miss istrolid :( ive played that game all the way up until a few months ago. Great game with an absolutely terrible developer.