Efficiency formula

[Chiki]

I was wondering if anyone else would be interested in the idea of making a formula to measure efficiency for tier lists and such (though it may have to differ from game to game).

Here's a very basic (and probably very wrong, but I just want to give you guys an idea of what I'm talking about) example:

Number of turns saved x reliability (on a scale of 1 to 10)

____________________________________________

Resources used (on a scale of 1 to 10)

[Darros]

Well, you'd need a couple definitions.

Reliability, for one, could mean a few things. As I'm interpreting it, you're referring to the strategies used to complete a goal. Would this goal be measured on a chapter by chapter basis or used for the whole game? And a scale could be kind of hard to work with, as it's hard to measure reliability. Some people take varying degrees of risk and consider some things as doable which others would not. For example, some people won't engage in battle even if the enemy has a minute chance of a critical, others do this all the time.

Second, what are you referring to as resources? Simply statboosters, or other things as well like weapons that aren't commonly available (A/S/SS rank weapons, hammers & ridersbanes before they're avaiable, etc).?

[nflchamp]

I don't really care how many resources a unit uses in and of itself. What I care about is what other use I could get out of that resource if I didn't give it to whatever unit I gave it to.

Beyond that, measuring turns saved is difficult for any given unit. Units, except for those at the very top, tend to be working with other units at any give time and so how much that particular unit saves is difficult to determine.

[aku chi]

I'm not interested in "making a formula to measure efficiency for tier lists". I believe that such an effort cannot be successful in achieving its goal. I do, however, believe that discussing efficiency in any form can be a useful venture. Therefore, I will happily critique attempts to measure effciency with a formula. My chief goal is to stimulate introspection and discussion, so please interpret my critiques in that light.

What do you mean when you refer to "turns saved"? Consider this hypothetical:

Hypothetical Chapter: Seize Objective

There are 4 units able to be deployed: Unit A, Unit B, Unit C, and Unit D. There are 3 deployment slots. There is one enemy - on the seize square - that does not move.

• Unit A is the only unit that can seize. Unit A cannot damage the enemy. Unit A can reach the seize square in 6 turns by herself. If Unit B shoves Unit A on turn 1, she can reach the seize square on turn 5.
  
• Unit B cannot damage the enemy. Unit B can shove Unit A.
  
• Unit C and Units D are functionally identical. Either can reach and ORKO the enemy with 100% reliability on turn 5. They cannot shove or rescue or in any other way transport Unit A.
  

How many turns, if any, do each of these units save?

[Progenitus]

Turns are overrated. 1 turn in a short, easy chapter like FE10 1-P is not the same as 1 turn in a longer, more difficult chapter like 4-4.

I don't remember who said this, but someone suggested # of actions or decisions to determine the weight of a chapter (although there is no formula, and deriving one will be difficult, particularly because everyone plays the game differently and thus will require different amounts of decisions to beat a given map, it's just the idea that should be incorporated). For example, if I need to make 5 decisions to beat 1-P (with a decision being where to place each unit, what attacks should be made, whether to heal or not, any trade or rescue chains I need, etc), but I need 50 to beat 4-4, obviously each decision becomes more critical in 4-4 as the chapter goes on, and most importantly, a screwup in 4-4 means that I might have to restart, meaning I have to do all of those decisions again, which is a waste of time.

Again, it doesn't really matter if Joe takes 5 decisions to beat 1-P and 50 to beat 4-4, but Bob takes 10 and 100, or if George takes 10 and 30, or if Dave takes 3 and 20, the bottom line is that 4-4 almost always requires more decisions than 1-P and thus should hold more weight (it would be extremely rare to see a player, say, take 10 decisions to beat 1-P but only 5 to beat 4-4). I don't have any suggestion for this magical formula to determine how much more important 4-4 is than 1-P especially when every person plays the game differently, but that's just part of debating; trying to see which subjective factors matter more.

Edited September 13, 2012 by Progenitus