S Rank Tier List for FE7

[GreatEclipse]

I thought people here considered contribution to a low turn count was the only relevant measuring stick? Under efficiency premises, a unit that saves 100 turns in a single chapter and then hits the bench is better than someone who saves 3 turns every chapter in a 33 chapter game. I have not played FE10, so I do not know the specifics, but if Edward saves you 13 turns in an early chapter due to only 1 other unit on the field, and Haar saves you 12 from being an awesome mobile tank from start to finish, Edward is absolutely better than Haar under the standards you have agreed to use.

Ranks luckily save us from that debate: as far as Tactics is concerned, every turn saved in chapter 11 is as meaningful as turns saved anywhere else. Matthew killing Wire is certainly possible if you feed him every kill and give him both stat boosters. The more impractical and luck based it is, the greater the worth of Hector.

Anyways, damn you, CATS. You always say exactly what I intended to only better. Do not assume I am illiterate in this stuff based on an official debate where I spewed enough bullshit for a farm. Opportunity cost is absolutely based on the next best alternative, look it up. Replacing Nino with a --/1 Isadora as the EXP unit only changes the magnitude of the trade. It does not change the net loss to EXP because Nino killing instead would still be better purely from the perspective of that rank.

It would be like claiming that the net loss to EXP caused by liberal use of Marcus is smaller than you claim, because in his absence Oswin would be forced to pick up the slack. There are 2 problems with this:

1) I have not proven that this is in fact what will happen. The example you provided did not happen for me. Getting Wil up from nothing to 10/-- in chapter 26 may look drastic, but I determined it to be the best outcome. Would Raven not be accountable for the net loss in EXP had I used him to one-round the Wyverns in the northern section as opposed to dropping them to below 10 HP and finishing with Wil?

2) The fact that killing with Oswin in his absence would be better overall than killing with Eliwood does not change the fact that I could have killed with Eliwood in his absence, and from a strictly EXP standpoint (Maximizing EXP, in your own words) that would be the best option.

[Narga_Rocks]

Absolutely. The problem here is that we're talking in terms of Exp gains and Exp rank, not ranks overall. Letting Sain kill instead of Heath is better for Tactics, worse for Exp, and arguably better for ranks overall. Notice that it's still worse for Exp, in spite of being better for the ranks overall. It's inaccurate to say that, because Tactics is more important than Exp, the action is not detrimental to Exp. It is detrimental to Exp; it's simply that the Exp cost is justified and outweighed by the benefits in other areas. But simply because the benefits outweigh the cost, does not mean that the cost isn't there.

And that's inaccurate. You can't consider only the choices are actually made, for reasons outlined above. If I'm comparing Erk and Oswin, I can't assume that Oswin gets a Speedwing and Erk doesn't, ignoring the possibility for the Speedwing to go to Erk, on the basis that that's what the player will do when actually playing the game. Similarly, you can't ignore the option for Heath or Dart to take a kill over Sain or Raven, on the basis that this isn't what will actually happen during gameplay. This mentality ignores opportunity cost.

Why isn't opportunity cost based on the next best alternative?

This is exactly the same as Narga's point, so my response is the same.

Except your attitude makes Raven's affect on tactics late = Marcus' early.

Here's the deal, Raven is really good. But there are other units that are almost as good for tactics. so you say Raven's affect on tactics is small because you can use them. But then you talk about his affect on exp is massive. Just like Marcus' early. Why? Because you can use Heath or w/e.

You can't do both at once.

Pick one. If you want to talk about Raven's horrible exp, then he is amazing for Tactics because some level 5 nub will suck. If you want to talk about how almost everyone promoted is close to Raven's ability, then he is just fine for exp because all those other options get about the same exp, or maybe 1 or two more.

What is the goal? s rank. You can't ignore one aspect when talking about other aspects. You can't pick and choose. In the case of Marcus, he gets 100% of the credit for tactics because there are no other options that are nearly as good early. That being the case, all the options that are possible would give a ton more exp than he does.

Also, how many promoted units are you running around with lategame? Let's say you have 7 promoted units that are good. You need 5 of them for a map. The cost of using 2 more of them is -14. -7 each. Are you going to say that all 7 of those promoted units each have a -7 to Experience? I'd suggest they have a -2 each. You are only dropping two of them, not all of them. Dropping all of them isn't an option because you will fail to s rank. If you give a -7 to each one of them, that's a -49 total, but that -49 isn't realistic. That's also why I said earlier that the cost of exp is distributed over all your promoted units since while you could drop Raven to get that extra exp, you could also drop some other guy. Why must Raven incur the full cost? Like I said, if all promoted units incur the full cost of that exp, then you must be expecting to s rank while deploying zero promoted units lategame. Either that or you are calculating more total cost than there actually is.

solid is suggesting to see what happens if Raven doesn't exist. Any time you would deploy Raven, you now deploy someone else. Now, either a: this unit is bad but gets a ton of exp or b: this unit is good but gets about as much as Raven would anyway.

This unit will not be a unit that is both good and gets a ton of exp, so one of two complaints against Raven is invalid. I don't actually care which one you choose to make invalid. Or you can merge the two into a unit that is lower leveled but not that much lower leveled and less powerful but not that much less powerful. In any case, Raven isn't looking as bad as you are making him out to be.

I don't get what your (both of you) issue is with solid's argument. There is more than one parameter to consider before determining the "next best alternative". (x, y). Consider each unit to have an x and a y. The "next best alternative" needs to be based off of this rather than solely the x parameter or solely the y parameter. You are trying to s rank, not merely 5 star exp. Once you have determined the "next best alternative" based on a combination of those two parameters, you can only then determine Raven's cost to exp. And I can assure you that the "next best alternative" will not be something like base Nino.

Edited May 2, 2010 by Narga_Rocks

[CATS]

Except your attitude makes Raven's affect on tactics late = Marcus' early.

Here's the deal, Raven is really good. But there are other units that are almost as good for tactics. so you say Raven's affect on tactics is small because you can use them. But then you talk about his affect on exp is massive. Just like Marcus' early. Why? Because you can use Heath or w/e.

You can't do both at once.

Pick one. If you want to talk about Raven's horrible exp, then he is amazing for Tactics because some level 5 nub will suck. If you want to talk about how almost everyone promoted is close to Raven's ability, then he is just fine for exp because all those other options get about the same exp, or maybe 1 or two more.

What is the goal? s rank. You can't ignore one aspect when talking about other aspects. You can't pick and choose. In the case of Marcus, he gets 100% of the credit for tactics because there are no other options that are nearly as good early. That being the case, all the options that are possible would give a ton more exp than he does.

When I'm talking purely in terms of Exp gains, yes, I can and should ignore things like Tactics. Saving turns doesn't impact your Exp figures. Finishing the chapter in fewer turns won't cause my units to gain more Exp per kill in the process or anything like that. Now, I agree that all ranks need to be considered, but that's not relevant to the point at hand. "You have to consider Tactics too" is true, but that fact does not change the way in which Raven affects the Exp rank, which is the main point in contention here.

Switching over to considering Tactics without considering Exp, the basis of the Marcus > Raven argument is made obvious. In terms of Tactics contributions and the ability to lower your turncount, there is a much larger gap between early Marcus and the next best alternative than there is between late Raven and the next best alternative. If Tactics is more important than Exp (as is implied with the constant assumption that Raven's kills are going to other promoted units even if he's not there), this ends the argument immediately. If Tactics is not more important than Exp, we move on to considering how each unit affects Exp rank. We see that the gap between each unit and the next best alternative is similar. If you try to maximize your Exp gains, there is no significant difference between the two. Marcus has greater Tactics contributions, while Exp contributions are similar.

Your position blurs the line between these two factors, or maybe eliminates it entirely, I'm not sure. Anyways, the fact that you can't do both at once is exactly why Marcus is arguably above Raven. You're either using more powerful units to improve Tactics in exchange for lowering Exp, or vice versa. You don't have the option to do both simultaneously. When you're doing the former, Marcus is more valuable than Raven. When you're doing the latter, neither one is useful.

I don't get what your (both of you) issue is with solid's argument. There is more than one parameter to consider before determining the "next best alternative". (x, y). Consider each unit to have an x and a y. The "next best alternative" needs to be based off of this rather than solely the x parameter or solely the y parameter. You are trying to s rank, not merely 5 star exp. Once you have determined the "next best alternative" based on a combination of those two parameters, you can only then determine Raven's cost to exp. And I can assure you that the "next best alternative" will not be something like base Nino.

You're determining the next best alternative in terms of overall ranks, then using it to determine Raven's cost to Exp specifically. If you are going to measure Exp cost specifically, you need to find the next best alternative in terms of Exp gains, and likewise if you want to measure ranks overall.

It seems like we're just repeating irreconcilable viewpoints at each other here, so I'm thinking we should just agree to disagree.

[Narga_Rocks]

No response to this?

Also, how many promoted units are you running around with lategame? Let's say you have 7 promoted units that are good. You need 5 of them for a map. The cost of using 2 more of them is -14. -7 each. Are you going to say that all 7 of those promoted units each have a -7 to Experience? I'd suggest they have a -2 each. You are only dropping two of them, not all of them. Dropping all of them isn't an option because you will fail to s rank. If you give a -7 to each one of them, that's a -49 total, but that -49 isn't realistic. That's also why I said earlier that the cost of exp is distributed over all your promoted units since while you could drop Raven to get that extra exp, you could also drop some other guy. Why must Raven incur the full cost? Like I said, if all promoted units incur the full cost of that exp, then you must be expecting to s rank while deploying zero promoted units lategame. Either that or you are calculating more total cost than there actually is.

You are getting -49 then when you should be getting -14. At least, assuming you aren't pulling a double standard against Raven by not penalizing all the other promoted units at that point. Either way, I don't get why Raven bears full responsibility for the cost. Unless you are planning to use zero promoted units, your reasoning fails miserably. As soon as you are carrying around even one promoted unit, that unit could be any promoted unit. Thus, none of them bear the full -7. They each have a reduced cost. You could drop any of them. As such, none of them take the full penalty. There are no alternatives to dropping Marcus. There isn't a second promoted unit in use that you can drop instead to cut his cost in half.

I can only accept your idea of Raven's cost to exp being equal to Marcus' if you want to tell me that you will never attack anything that Raven could have attacked with some other promoted unit. Every single action that Marcus ever takes, if you drop him you are guaranteed to be using some unpromoted unit. This is not the case with Raven. If even one action he takes would've been taken by another promoted unit had Raven not existed, your argument fails.

As for your response to the "next best alternative" thing, this is why I said earlier to widen your scope. You can't narrow it down and look at one aspect while ignoring all others and then turn around and do the same with some other aspect. It doesn't make sense because it makes the list impractical. If you can't accomplish s ranking then what's the point of maximizing exp? You make decisions towards s ranking, and then determine the costs based off that. You can't just ignore the other ranks when determining costs. You won't be making those decisions if you want to s rank. Thus, they are not the next best alternative.

Edited May 3, 2010 by Narga_Rocks

[Crimson_Edge]

Nah, I'm not combining EXP and Tactics. I'm only talking about what happens to the EXP Rank in the event of Raven's absence. Absolutely maximizing EXP has nothing to do with my point.

Could you maximize EXP in Raven's absence? Sure. Are you necessarily trying to? No. In Marcus' case, you don't have to try to specifically maximize EXP--removing Marcus should do no matter what your goals actually are. Easier to accomplish. Marcus incurs more of the penalty.

Or, given three candy bars: a Twix, a Snickers, an Almond Joy. The Snickers is the largest, followed by the Twix. But one condition of the choice is Almond Joy will be chosen over Twix, just as a rule. So you pick the Snickers, right? What's the opportunity cost of the Snickers? Well, the Twix isn't really an option you'll be exercising. In the absence of the Snickers, you'll be picking the Almond Joy, according to the rules. So the actual opportunity cost of the Snickers in terms of size is compared to the Almond Joy, not to the Twix, and is thus greater. Etc. etc.

Or, on the other point, to put it in Reikken's words, "Using Nino is entirely independent of using Raven."

This is more of a credit to using Nino than a negative to using Raven.

I'm assuming no kind of illiteracy on your part, GE. I'm only saying that you're wrong. Once again, I'm willing to go into much greater detail on this, but from here it seems that nobody is willing to change his mind. It seems to be a definitional argument, if anything.

[CATS]

As for your response to the "next best alternative" thing, this is why I said earlier to widen your scope. You can't narrow it down and look at one aspect while ignoring all others and then turn around and do the same with some other aspect. It doesn't make sense because it makes the list impractical. If you can't accomplish s ranking then what's the point of maximizing exp? You make decisions towards s ranking, and then determine the costs based off that. You can't just ignore the other ranks when determining costs. You won't be making those decisions if you want to s rank. Thus, they are not the next best alternative.

I think you can measure them individually so long as you combine the measurements for an overall value. I also think we might just be seeing two sides of the same coin, as both lines of reasoning should lead to Marcus > Raven. Me, for example, I would see that a unit like Marcus contributes -30 towards Exp but +100 towards Tactics, and would combine the two for a simple +70 overall. It seems like you're just seeing the +70 overall. "Raven doesn't hurt Exp," but that's because it's preferable to give his kills to other promoted units instead of low-level unpromoted units, i.e. continuing to favor Tactics over Exp. Marcus has a much stronger Tactics contribution, so with Tactics > Exp, Marcus > Raven.

No response to this?

I responded to that earlier in the topic:

The other side of the coin is that there's alot more kills to give up on Raven's side. Lategame chapters have higher enemy counts, and with mostly promoted units on the field, they'll be taking a higher percentage of the chapter's kills than just Marcus is taking early on. If there's 35 kills taken by 5 promoted units in a lategame chapter, and 7 kills taken by Marcus in an earlygame chapter, each of the 5 promoted units hurt Exp rank to a similar extent as Marcus in the earlygame chapter. Now if it was something like 7 kills between 5 promoted units during the lategame chapter, I would agree with you, but I don't think that will be the case. And even if it is, you almost certainly sacrificed Tactics if you held your good units back like that.

[Crimson_Edge]

"Raven doesn't hurt Exp," but that's because it's preferable to give his kills to other promoted units instead of low-level unpromoted units, i.e. continuing to favor Tactics over Exp. Marcus has a much stronger Tactics contribution, so with Tactics > Exp, Marcus > Raven.

Not preferable. But it may happen. There being more alternatives matters just as much as it matters that more alternatives can boost Tactics in Raven's stead.

It matters that replacing Raven with ~20 combat units hurts Tactics. It matters that replacing Raven with ~10 combat units doesn't hurt Tactics.

It matters that replacing Raven with ~15 combat units doesn't do anything to EXP. It matters that replacing Raven with the other ~15 combat units helps EXP.

Or whatever arbitrary numbers you fill in. Raven is compared to all possible alternatives, not just the best one, with each given weight according to how likely it is to be used. We're tiering characters, not rank contribution methods.

Definition of opportunity cost uses the words "mutually exclusive." This is beyond the scope of simple, two-dimensional opportunity cost.

[Rewjeo]

I think CATS is assuming that we're ONLY talking about EXP rank, ignoring the others completely, while CE is still taking them into account, but only discussing the effect on the EXP.

[Narga_Rocks]

I think CATS is assuming that we're ONLY talking about EXP rank, ignoring the others completely, while CE is still taking them into account, but only discussing the effect on the EXP.

Well, you do have to s rank, you know. Hence, while doing that, what goes on with the exp rank? That's a better question than "if we ignore the other ranks and focus solely on maximizing exp then what happens?" The problem is that you aren't ignoring the other ranks and focusing solely on maximizing exp so why would you even bother asking/answering that question?

I responded to that earlier in the topic:

The other side of the coin is that there's alot more kills to give up on Raven's side. Lategame chapters have higher enemy counts, and with mostly promoted units on the field, they'll be taking a higher percentage of the chapter's kills than just Marcus is taking early on. If there's 35 kills taken by 5 promoted units in a lategame chapter, and 7 kills taken by Marcus in an earlygame chapter, each of the 5 promoted units hurt Exp rank to a similar extent as Marcus in the earlygame chapter. Now if it was something like 7 kills between 5 promoted units during the lategame chapter, I would agree with you, but I don't think that will be the case. And even if it is, you almost certainly sacrificed Tactics if you held your good units back like that.

I can't comment on your numbers. But what I can comment on is this seems to miss the point. If Raven disappears, whatever kills he may have made will be split between promoted and not-promoted units. Were Marcus to cease to exist, that won't happen. Hence, Marcus has a bigger impact on exp than Raven because in Marcus' absence the increase in exp is much greater than the increase in Raven's absence. Which makes it exceedingly difficult for me to believe that they have the same impact on exp.

The point of me constantly bringing up the other promoted units is that Raven can't bear the responsibility on his own. If you were to drop Raven and suddenly all his kills are going to non-promoted units, then you clearly had too many promoted units in the first place. And so now you have x-1 promoted units. Why can't Raven be one of those x-1? And so now he is, because otherwise you are sandbagging. And then repeat the process. If you can tell me each time that removing Raven gives all his kills to non-promoted units, then I can repeat that argument to drop another promoted unit that isn't Raven. And if you continue your argument until the end, there are now no promoted units at all. One step before the end, Raven is the only promoted unit. He now has the same impact on exp because removing Raven will remove that hit on exp. But he now also has the same impact on tactics as Marcus does, because removing Raven will kill your tactics. But that clearly won't work for s-ranking, anyway, so at some point, it should be obvious that when you reach the desired number of promoted units in a map, removing Raven will split the exp between promoted units and non-promoted units. And at this point it becomes obvious that Marcus has a bigger effect on the exp rank than Raven does.

Edited May 3, 2010 by Narga_Rocks

[CATS]

I can't comment on your numbers. But what I can comment on is this seems to miss the point. If Raven disappears, whatever kills he may have made will be split between promoted and not-promoted units. Were Marcus to cease to exist, that won't happen. Hence, Marcus has a bigger impact on exp than Raven because in Marcus' absence the increase in exp is much greater than the increase in Raven's absence. Which makes it exceedingly difficult for me to believe that they have the same impact on exp.

It just boils down to whether you think the choice not made is relevant, or if you think that only the results of the decision that is actually made are relevant. Clearly, we have a fundamental difference on this issue.

[Dat Nick]

. I have not played FE10, so I do not know the specifics, but if Edward saves you 13 turns in an early chapter due to only 1 other unit on the field

It's the freaking prologue chapter. Auto-Topping Edward just for what he does in the prologue and totally hand-waving how bad he is the rest of the game is stupid as hell.

[CATS]

It sounds pretty similar to auto-topping Matthew because of the silver card, tbh. Matthew's poor contributions later in the game are "hand-waved" by the magnitude of his actions during a single chapter.

[Florete]

It sounds pretty similar to auto-topping Matthew because of the silver card, tbh. Matthew's poor contributions later in the game are "hand-waved" by the magnitude of his actions during a single chapter.

Matthew's position is based on his entire early game performance, not just the Silver Card. If all he had was the Silver Card, I'd also be arguing for him to drop. I actually did argue that back when I thought he was only up there for the Silver Card.

[Crimson_Edge]

The implications of pure, two-dimensional opportunity cost are pretty severe.

The issue is the other side of this Raven vs. Marcus as related to EXP argument is ignoring other factors, and assigning the blame entirely to Raven when there are alternatives who do the same that Raven does, or worse (Marcus being one of them in the lategame, if he's used, ironically).

Anyway. Behold. The opportunity cost of getting Lalum is not getting Elphin. Lalum gives you a dancer but loses you a dancer. Net value = 0. The opportunity cost of getting Elphin is not getting Lalum. Elphin gives you a dancer but loses you a dancer. Net value = 0. This is what happens when you try to assign an individual unit the penalty shared by multiple units, that could have been caused by multiple units, and which could not have been removed by removing any given unit (the player may still have gone Elphin's Route despite not using Elphin himself). This is your logic in a nutshell. Correct me if I am mistaken.

Otherwise, Elphin/Lalum to lower mid pls.

[Dat Nick]

The Silver Card isn't even THAT awesome in Efficiency. The main thing is just three promotion items and a red gem. Even though all three of the items he gets you are pretty well contested for I'm not really sure it's an auto-top.

[Narga_Rocks]

It just boils down to whether you think the choice not made is relevant, or if you think that only the results of the decision that is actually made are relevant. Clearly, we have a fundamental difference on this issue.

I assume that in the event that Raven is missing the choice made is the next best alternative and determine the cost of using him accordingly. I see no reason to consider various bad ideas that aren't the next best alternative when determining Raven's cost to the exp rank. What's the point?

Our difference seems to be in determining what actually is the next best alternative. I think that since the goal of the tier list is to s rank the game that the next best alternative would be whatever best helps us to do that. You seem to think that the next best alternative is whatever maximizes exp. To this point I still can't understand why. But yes, this seems be a fundamental difference. Or whatever you wish to call it when I am trying to s rank and you are trying to 5 star experience alone.

[Anouleth]

The implications of pure, two-dimensional opportunity cost are pretty severe.

The issue is the other side of this Raven vs. Marcus as related to EXP argument is ignoring other factors, and assigning the blame entirely to Raven when there are alternatives who do the same that Raven does, or worse (Marcus being one of them in the lategame, if he's used, ironically).

Anyway. Behold. The opportunity cost of getting Lalum is not getting Elphin. Lalum gives you a dancer but loses you a dancer. Net value = 0. The opportunity cost of getting Elphin is not getting Lalum. Elphin gives you a dancer but loses you a dancer. Net value = 0. This is what happens when you try to assign an individual unit the penalty shared by multiple units, that could have been caused by multiple units, and which could not have been removed by removing any given unit (the player may still have gone Elphin's Route despite not using Elphin himself). This is your logic in a nutshell. Correct me if I am mistaken.

Otherwise, Elphin/Lalum to lower mid pls.

That's incorrect logic. Lalum/Elphin have no opportunity cost, they are both recruited automatically. How can a character that joins automatically have an opportunity cost?

[Narga_Rocks]

That's incorrect logic. Lalum/Elphin have no opportunity cost, they are both recruited automatically. How can a character that joins automatically have an opportunity cost?

You must give one up to get the other. You could argue they then have x-x = 0 worth.

[WeaponsofMassConstruction]

So is Gonzo also counted off for preventing you from recruiting Gonzo?

[Dat Nick]

The cost of B route Gonzo existing is that you don't get the superior Gonzales, so that's not completely atrocious logic.

[Life]

Doesn't this logic also mean that either Raven, Guy, Bartre and Dorcas should move up while Erk, Serra, Prissy and Lucius should move down based on the cost of getting Kenneth's map over Jerme's? I find that Jerme's map allows you to gain much bigger Tactics and Exp boosts than Kenneth's due to the number of turns needed for the map and the extra amount of enemies present.

[Crimson_Edge]

The point is that Raven doesn't get 100% of the blame for the lategame EXP penalty (or even his own EXP contributions) because pure opportunity cost doesn't work with more than one non-exclusive variable. Other non-Raven factors motivate that EXP penalty. Raven's absence does not remove it.

In the same vein, Lalum's absence does not remove the possibility that you are going to A Route anyway, and thus will miss out on Elphin.

In order for the current Marcus > Raven argument to be valid, the other side must be willing to accept these implications, and all sorts of arbitrary notions that will follow.

[Life]

I don't understand why GE's penalizing Raven for doing what every other endgame combat unit does. Doesn't promotion give Raven an extra 19 levels that he can achieve which blows Marcus out of the water in terms of Exp gain?

Seriously, I'm lost. I'm not seeing Marcus > Raven because I don't understand why Raven's getting penalized for shit that never comes up in other debates. Like Lyn vs. Isadora for example.

[nflchamp]

I don't understand why GE's penalizing Raven for doing what every other endgame combat unit does. Doesn't promotion give Raven an extra 19 levels that he can achieve which blows Marcus out of the water in terms of Exp gain?

Seriously, I'm lost. I'm not seeing Marcus > Raven because I don't understand why Raven's getting penalized for shit that never comes up in other debates. Like Lyn vs. Isadora for example.

No one here gets it, don't worry.

[Crimson_Edge]

I think no one would be penalizing Marcus more if there were some magic unit in earlygame who gained +30 EXP over every other unit. That unit would simply be given greater credit.