Formula for card strength

[Abominate Scourge]

Most mechanics of every game have a background formula which takes playable elements into sum and the goal is to have them all around equal strength number. The nicest visible example of that is in Elder Scrolls Construction Kit where they measure the power of base character points in sum of (if I remember correctly) 250. If you go above, it's overpowered. If you go below, it's underpowered.

Similarly, Magic the Gathering has exact logic of card power. If it does this, then you have to pay 1 more, if it does this, then caster has to take damage, or alike.

So I wonder, has anyone tried (or knows) to figure out how exactly to measure the card's power? When we read the card we all have this feeling of how strong or weak it is, mostly, is it balanced. I am also not talking about it being used in game, since it depends from situation to situation, but exactly about the base design.

For example, let's say that all cards have to be around 100. And then we count. The more energies you have to use to attack, the number reduces. The stronger the attack is, the number grows. If it has effects, it grows, if it backfires, it reduces, and so forth, ultimately concluding that for example each energy cost reduces power of card by 7.5, each 10 damage raises by 10, in proportion to energy card increases by 1/2, so for 1 energy and damage 20 it doesn't value 20, but 25, and then reducing HP from 50 for example reduces by 15, increasing over 90 increases by 1.25, and so forth.

Has anyone tried to figure that out? Or would you like to try?

 

[Dork Void]

There is stuff like synergy and pokepowers and pokebodies, which would be extremely hard to compute with a formula. Your best bet would be to simply try out the card with many others against as many different successful decks as possible and see if its worth it or not.

 

[Abominate Scourge]

There is stuff like synergy and pokepowers and pokebodies, which would be extremely hard to compute with a formula. Your best bet would be to simply try out the card with many others against as many different successful decks as possible and see if its worth it or not.

Yes, but like I said, that is situational. There, however, must be the basic concept. Why exactly that poke-power and exactly those effects? You don't need to test a card in a game if you see that it has no retreat cost, high hp and poke power that prevents evolving. It's obvious it's overpowered. I highly doubt that card developers don't have rules for crafting a card.

 

[Zyflair]

Try calculating Yanmega in comparison to other meta decks and tell me how the analysis goes (and explain the procedure used).

But seriously, in Pokemon TCG, it's mainly all about matchups, not card strength.

 

[Abominate Scourge]

But seriously, in Pokemon TCG, it's mainly all about matchups, not card strength.

You mean how you combine the cards? You don't think that, for example, making Base Set Kangaskhan with only 1 reatreat would make it overpowered in any matchup?

 

[Zyflair]

I'm assuming you mean Jungle Kangaskhan, but please explain in what particular matchup changing the RC from 3 to 1 would change in unlimited. Changing the RC to smaller numbers definitely help, yes, but if a particular matchup isn't based on how well you can retreat, RC makes no difference. A mere 60 damage looks like nothing until you realize that one of the metagame decks is weak to it. Simple numbers have to take into account of hte situation.

 

[scizorlicious]

There is simply no formula for determining card strength.

Take Reshiram: 120 damage for RRC and it discards 2 fire energy. Look at that card alone and it's really not that great. Donphan can do 60 for 1 with less of a drawback and a nice PokePower. But you have to consider other cards: Typhlosion, maybe Emboar. They make Reshiram work well.

Yanmega is a lot less powerful without Judge and Copycat. Heck, how do you determine the value of a Pokemon's PokePower or Pokebody? Arbitrarily? There's not really any formula you can use. And how much better is Poison than Burn? Any better? Worse? The same? And what about Confusion; the odds a Pokemon can retreat out of it, etc.?

Even attacks would be tough - is 2 for, say, 70 better than 3 for 90? With 2 for 70, you're getting more damage per energy. But is the time you lose with 3 energy worth the extra 20 damage?

And that's when one stat depends on another - Pokemon with low HP will require low-energy attacks to be effective, for example.

And Weakness/Resistance - that really depends on how good other decks are. It even depends on area.

It'd be nice to have a formula to determine card strength. There just isn't any way to.

 

[Celebi23]

Clearly there's not a formula, because 90% of the cards we have are not balanced and we rarely have good metagames. Also, this game is more about matchups, playing skill, deck consistency and luck than what card is the most powerful. Clearly a card like Blastoise should be ranked pretty high on that power list but it's not even that playable.

 

[Zero]

There used to be something closer to a formula. For example, attack damage would be calculated by adding 10 damage for each [C] and 15 damage for each [X] (energy type). Adding in effects will mess with this formula a bit, but that is about how it went. Now for current, there really is nothing close to a power formula. The card that deviate from anything close to a power formula are what are amazing/broken. If we had a formula, or a few rules to follow, certain mistakes would not have been made. Even so, we don't, so we will continue to repeat history.

 

[JEKyl]

Assuming someone wants to do this, averaging damage would be best served by measure it in different energy classes. I would structure such a system as follows:

*0
*1
*2
*3
*4

(Five energy attacks are unnecessary to measure, since they're so rare.)

Each would have two sub-classifications: colorless and non-colorless attacks. Additionally, you would exclude all non-damaging attacks from measurement, as well as all "unlimited damage" effects (sans "flip a coin until you get tails" attacks, which can be counted as their printed damage amount due to expected utility being roughly equal). Finally, I would include damage counter attacks.

 

[Zyflair]

Classifcation would be difficult and almost pointless since it's still so metagame reliant.

 

[JEKyl]

Classifcation would be difficult and almost pointless since it's still so metagame reliant.

Oops, should have included my other assumption: that this only pertains to the HGSS-on format.