A thought about points systems

I'm thinking mostly of FT as I write this, but it could apply just as
well anywhere else.

It seems to have been fairly comprehensively established that the
combat-effectiveness and construction/training cost of a unit are not
simply related. Construction costs make sense when one's running an
extended campaign with replacement units coming in; but what are
combat-effectiveness points really good for? SG2 doesn't have them at
all; DS2 showed how hard they can be to get right. Why bother?

The standard approach seems to be "to balance pick-up games". But this
is intrinsically unrealistic; if the forces were really even, why would
either commander have attacked? From what I've read, most commanders
given the choice to attack would rather have at least a 2:1 force
advantage, so unless we assume that every power has ESU-style commissars
forcing its field officers into attacks for which they're
under-resourced...

No. A "balanced fight" makes no sense historically or psychologically,
for all it's convenient for a game. Can we instead find a way of
creating realistic battles as pick-up games? And can
combat-effectiveness points be a useful tool in that process?

I think so; but it needs a change of attitude. Rather than "let's put
our forces on the table, slug it out and see who wins", the approach
would be "here's a situation, see who can make the best of it".

Assume there's a basic ration of victory points for the scenario,
derived from specific objectives: a freighter that's part of a convoy
(which might be something both sides wanted control of, or might be
something one side wanted to destroy), a location that you have to have
troops in when the other guy doesn't (basic attack/defence), or
whatever. That part of the game is zero-sum: there are only so many
points to be won, and they will be won by one side or the other.

Now, compensate for force sizes. Each side's score multiplier is
something along the lines of:

(surviving points) * (enemy's original points) / (original points)^2

So. You go in with 500 points against a 200-point defence. You lose 200
points during the attack. Your score multiplier is
(300*200/500/500)=0.24.
If the enemy had inflicted those losses on you with a 100-point defence,
it would be 0.12; if you hadn't taken any loses at all, it would be 0.4.
This gets multiplied by your basic victory points to assess how well you
did; whoever scores higher is the winner. It's entirely possible to take
all the objectives, but to use up so much manpower and materiel doing so
that you end up the loser...

Advantages:

- encourages preservation of own forces
- doesn't directly reward destruction of enemy, unless it helps gain
basic victory points
- gives resonably sensible results with very unbalanced sides

Does this make sense? Has anybody tried anything of this sort?