So, this is the formula that is most likely to be accurate. Notice I said most likely. I'm sure you could find a way to break it if you tried. [Edit: Purple Nymph breaks it. Completely. I guess I couldn't account for how much that ability is really worth. The formula has been changed to account for Purpley.]
A+Hm+(S-Sc)-C-3
A is attack. C is cost. Those are pretty understandable.
S is skill worth. Put 0 for a horrible ability. (e.g. Singularity, though of course since that is unobtainable this isn't really a great example) Put 1 for an ability that is usually bad, but can be good. (Ghost in the Past) Put 1.5 for no ability. Put 2 for an ability that is situational, but usually useful. (Psion) Put 3 for a good ability (Pretty much all the cards in the game) Put whatever makes sense for an ability that is 100% awesome. (Purple Nymph would have a worth of 8, antimatter's cost)
Sc is skill cost. 0 for passives and activated abilities, as well as having no skill. 1 for a 1 or 2 cost, 2 for a 3 or 4 cost.
Hm is health modifier. It was originally H/3, but that made gravity seem OP. Then I made it H/8, which made fire seem UP. So I've scaled it accordingly.
Now let's test it on Crimson Dragon.
12+1.5+(1.5-0)-10-3
15-10-3
2
Erm, well, it's supposed to be in the 1 to -1 range... so... I guess crimson dragon is slightly OP...?
Well, to be honest, I would rather use it over most other dragons
If your end result is -1 to 1, your card is very balanced. If your end result is -2 to 2, it's slightly off. -3 to 3 is off. -4 to 4 is very off. -5 to 5 is going to sue you for making cards while on drugs. Above 5 or below -5... ... ... Houston, we have a problem.
