We need to figure out what logic we're using when deploying units

[Progenitus]

There seems to be two extreme views on how units are fielded.

1) Some people say that units should/can be dropped once they hit their period of suck, or outlive their usefulness. For example, Marcus/Jeigan aren't fielded/killed off after they outlive their usefulness so they don't drag down their own usefulness with their suck. Basically, everyone is used as efficiently as possible and won't be used if it won't generate any efficiency.

2) Another view is that units are used every chapter. We can't drop Marcus/Jeigan after they outlive their usefulness.

I have problems with both views. For the first one, it basically turns the thing into a maximum efficiency playthrough. In other words, even mid tiers are no longer fielded, as they are not efficient. Then as you get to the bottom tiers, debates become stupidly boring. like, Unit A does a couple rescues or whatever and then run off to be cannon fodder vs unit B does a couple potshots or whatever and then run off to be cannon fodder.

But the other extreme is bad, because Marcus/Jeigan/etc. end up building up an unnecessary amount of suck when it's completely avoidable if you simply didn't field them or had them hide in a corner. It'd be like throwing a dancer on the frontlines and complaining that they can't attack.

Basically, the first one forces you to conform to a certain playstyle (it tells you how to play), but the second is unrealistic.

There needs to be a sort of middle ground, or at least consistent logic. This topic has sprung up in various tier list topics, but I don't ever think a centralized topic was made for it.

Personally, as of right now, I like Mekkah's post in the FE8 tier the most. Pretty sure it can ironed out better though, or scraped entirely if a good argument is made.

WRT +/- utility: performance over the whole game is weighed, as otherwise we are reflecting a 100% efficiency playthrough, and not the actual value of a unit throughout the game. However, a time where you are bad weighs less against a unit if your good part is before your bad part.

Basically, peeps like Marcus are not penalized as heavily for being sucky during the time period that they can avoid. I'd still prefer if Marcus and peeps are not penalized at all, but that would deviate towards the logic 1 extreme where we go for maximum efficiency.

There's more I'd like to add, but I think I'll leave it at this for now.

Edited November 9, 2009 by 8========================D

[Florete]

I agree. In the FE10 tier list, for comparison purposes, I always note that for someone like Tormod almost anyone is doing better than him at Endgame but it holds little weight. Not nothing, though, or he'd probably be a bit higher. What Mekkah said is basically my view on things; look at performance for the whole game, but put less/little weight to a bad part if it comes after the good.

[CATS]

If forced to choose, I'd prefer the first option because it's the one that makes sense. Tier lists are based on efficiency, plain and simple. This is the reason why units aren't assumed to get supports with low tier units--the tier list is based on the assumption that the player is making efficient choices.

The point about boring low tier debaets is quite true, though. If it were up to me, I'd go with the first option when units being compared are able to create some sort of positive utility, and go with the second if the units are not able to make any significant contributions at all.

In other words, if the two units in question are both only ever able to achieve negative utility (by being fielded) or zero utility (by not being fielded), then you move on to assuming they're fielded anyway and seeing which one is less negative, as opposed to the option of not fielding them and giving both zero utility, which indeed shuts down discussion. In the case of someone like FE6 Marcus, he still gets his due, as he is able to create a significant amount of positive utility when used a certain way. But then when you have something like Lyn vs Fiora, you can still have a discussion about it.

This approach doesn't adhere to strict consistency, but if you want absolute consistency, you have to go with the first option all the way. Either that, or change the premise of the tier list entirely. I'd rather try and eliminate the major flaw of the first option while retaining as much consistency with the tier list's premise as possible.

[Meteor]

Would it be fair to agree not to penalize a unit for a single deployment spot? I believe many members have been suggesting something like this, and it makes a lot of sense to me. Assuming every other spot is filled by efficient units, decide how well the character in question stacks up. If said character needs other inferior units for supports, then he/she will noticeably start to drag down efficiency, but a single slot seems fair/expendable.

I somewhat disagree with giving more weight to early +utility just for the sake of it being early. Rather, Marcus is hugely helpful because he is good when he's irreplaceable. The majority of the time, yes, this will probably happen because of early join time, but it's not outside the realm of possibility that some unit could be vital mid- or lategame.

Edited November 9, 2009 by Meteor

[Colonel M]

I think that for all utility units their good period should always be weighed more than their bad period. Then again there aren't a whole lot of said "utility units" that follow this barring units such as FE6 Marcus, FE8 Dozla perhaps, FE10, and FEDS.

[Narga_Rocks]

If forced to choose, I'd prefer the first option because it's the one that makes sense. Tier lists are based on efficiency, plain and simple. This is the reason why units aren't assumed to get supports with low tier units--the tier list is based on the assumption that the player is making efficient choices.

The point about boring low tier debaets is quite true, though. If it were up to me, I'd go with the first option when units being compared are able to create some sort of positive utility, and go with the second if the units are not able to make any significant contributions at all.

In other words, if the two units in question are both only ever able to achieve negative utility (by being fielded) or zero utility (by not being fielded), then you move on to assuming they're fielded anyway and seeing which one is less negative, as opposed to the option of not fielding them and giving both zero utility, which indeed shuts down discussion. In the case of someone like FE6 Marcus, he still gets his due, as he is able to create a significant amount of positive utility when used a certain way. But then when you have something like Lyn vs Fiora, you can still have a discussion about it.

This approach doesn't adhere to strict consistency, but if you want absolute consistency, you have to go with the first option all the way. Either that, or change the premise of the tier list entirely. I'd rather try and eliminate the major flaw of the first option while retaining as much consistency with the tier list's premise as possible.

I mostly agree with you. It is basically a tie breaker idea. When units can build positive utility (if we go with the existence of +/- utility as it relates to deployment slots) then they are fielded. As soon as they can no longer build positive utility, and are just a drag on the team, they are dropped. Those units that never build positive utility and are always a drag, even after training, are then fielded anyway to get a useful assessment of how bad they are.

Basically, while it may not sound consistent, it has its own form of consistency. Also it is more of a middle ground between smash's 2 extremes.

I'd still rather not apply opportunity cost to deployment, but it does make some comparisons more complex than they would otherwise be.

[Progenitus]

If forced to choose, I'd prefer the first option because it's the one that makes sense. Tier lists are based on efficiency, plain and simple. This is the reason why units aren't assumed to get supports with low tier units--the tier list is based on the assumption that the player is making efficient choices.

The point about boring low tier debaets is quite true, though. If it were up to me, I'd go with the first option when units being compared are able to create some sort of positive utility, and go with the second if the units are not able to make any significant contributions at all.

In other words, if the two units in question are both only ever able to achieve negative utility (by being fielded) or zero utility (by not being fielded), then you move on to assuming they're fielded anyway and seeing which one is less negative, as opposed to the option of not fielding them and giving both zero utility, which indeed shuts down discussion. In the case of someone like FE6 Marcus, he still gets his due, as he is able to create a significant amount of positive utility when used a certain way. But then when you have something like Lyn vs Fiora, you can still have a discussion about it.

This approach doesn't adhere to strict consistency, but if you want absolute consistency, you have to go with the first option all the way. Either that, or change the premise of the tier list entirely. I'd rather try and eliminate the major flaw of the first option while retaining as much consistency with the tier list's premise as possible.

This could work, but there's one thing that bugs me, and that is the issue of forced units who don't take up a deployment slot. Can we allow them to simply make minor contributions during the time period they're forced and never use them again?

For example, liek... Wolt vs Treck. Wolt is awful, but he's forced into several earlygame chapters and can do some minor chip damage, and then we can drop him after he's no longer forced. Treck is going to be better than Wolt in the chapters they both exist in, but Treck is probably taking up a unit slot, meaning Treck isn't efficient. yet I would still put Treck over Wolt (and I'm sure Treck is over Wolt in the tier list), since Treck is much easier to bring up to par than Wolt.

That might've been a pointless and obvious example, but there are many within FE to pick from. For example... Bartre vs Fiora. Would Bartre's forced period where he can make minor contributions beat out Fiora, who's probably better than Bartre in the chapters they both exist in but would be taking up a unit slot? FE10 has tons of these, since there are plenty of bad characters forced into chapters who can make minor contributions while averageish units aren't forced.

I mean, sometimes I'll side with the forced unit, if the unit who is better when they both join is still god awful. Like... maybe Wolt vs Dorothy; wolt's forced chapters vs Dorothy being better when they both exist but she's still pretty bad (I haven't really looked into FE6 for a long time though, so don't kill me if I'm underestimating Dorothy). But when do you switch off? When would the later joiner win?

[Paperblade]

I somewhat disagree with giving more weight to early +utility just for the sake of it being early. Rather, Marcus is hugely helpful because he is good when he's irreplaceable. The majority of the time, yes, this will probably happen because of early join time, but it's not outside the realm of possibility that some unit could be vital mid- or lategame.

Being good before being bad is better because a unit that's good and then becomes worse can be dropped to avoid their bad period (if it's as bad as say, Jeigan's or Marcus etc.), so I can still field a better unit in their place and not have to worry about them sucking up my team (see: Option 1).

Meanwhile, a unit that sucks before they get better (such as, say, Rebecca or Thany), must be fielded when they're bad in order to get the EXP and levels for them to improve.

It doesn't help that in most FEs, the more difficult portion of the game is the early game. Why is this? You have fewer units, so a lower chance of being able to field a bunch of units that just stomp all over everything. Enemy growths are usually lower than PC growths, so our units that are sucking it up early are going to outgrow enemies, especially since in a lot of FEs they level faster than enemies do (lolfe7).

A larger cast also makes it much less likely for there to be a "clutch" mid or late game character, simply because you're more likely to have someone that can also do the job out of a team of 30 than out of a team of 8. Compounded with the previous paragraph, there simply aren't going to be "vital" mid-lategame units in most FEs. Ones that are head and shoulders above others, sure (lolike, lolhaar). But vital the way that Marcus is? Unlikely.

The only real exception I can think of where a unit's bad but then gets better but's still liek the best are Sedgar and Wolf, who are pretty much as close as you can get to being vital mid-late game.

Edited November 9, 2009 by Paperblade

[Interceptor]

No offense to Mekkah/RFoF and others, but I'm firmly against wishy-washy shit rules like have been suggested here. It's not sufficient to say that you're going to weigh FE10 Tormod's late chapters "less", because it's impossible to quantify what "less" is. Narga gave voice to the only way that such a rule would be useful: if everything else was equal, and it was just a tiebreaker. It's pretty easy, after all, to show how one unit is less crappy than another, so all other things being equal we can pull it off. The problem is, things rarely work out that way.

CATS has the most consequential observation here. TBH, the thread ended as soon as he posted, the rest of you just haven't noticed it yet. The answer to the smash's naive question in the OP is not one of logic, it's one of purity. We're talking about efficiency tier list here, and the logic only takes us in one direction as far as that's concerned: option #1. Anything else is corrupting the goal of the tiering process.

Now, I'm fine with coming up with "house rules" to facilitate tiering debates, but those are game-specific. aka tier list specific. For example, in FE10 HM, there are really only 4 units in the cast of 70+ that are actually so horrifically bad that fielding them at all is auto-negative utility (they are in Bottom, for that reason). So FE10 has no need for this logic-busting bullshit, in my opinion, because it's not needlessly constraining debate.

In other words, this thread hasn't broken new ground, and I doubt that the problem can be solved generically because of the environmental factors unique to every FE.

[NinjaMonkey]

For the most part, I only deploy units if I'm actually going to use them later on in the game.

Take Shinon and Rhys in FE10 for example: Shinon is my archer for endgame, so I make sure that he is always fielded. Rhys on the other hand, isn't going to endgame, and therefore isn't very likely to be fielded (depending on how many deployment slots I have availiable).

[CATS]

This could work, but there's one thing that bugs me, and that is the issue of forced units who don't take up a deployment slot. Can we allow them to simply make minor contributions during the time period they're forced and never use them again?

For example, liek... Wolt vs Treck. Wolt is awful, but he's forced into several earlygame chapters and can do some minor chip damage, and then we can drop him after he's no longer forced. Treck is going to be better than Wolt in the chapters they both exist in, but Treck is probably taking up a unit slot, meaning Treck isn't efficient. yet I would still put Treck over Wolt (and I'm sure Treck is over Wolt in the tier list), since Treck is much easier to bring up to par than Wolt.

That might've been a pointless and obvious example, but there are many within FE to pick from. For example... Bartre vs Fiora. Would Bartre's forced period where he can make minor contributions beat out Fiora, who's probably better than Bartre in the chapters they both exist in but would be taking up a unit slot? FE10 has tons of these, since there are plenty of bad characters forced into chapters who can make minor contributions while averageish units aren't forced.

I mean, sometimes I'll side with the forced unit, if the unit who is better when they both join is still god awful. Like... maybe Wolt vs Dorothy; wolt's forced chapters vs Dorothy being better when they both exist but she's still pretty bad (I haven't really looked into FE6 for a long time though, so don't kill me if I'm underestimating Dorothy). But when do you switch off? When would the later joiner win?

Yeah, I thought about that and it's definitely the biggest issue with my proposal. Ultimately it would need to be decided on a case-by-case basis; you can't really lay down a single, universal cutoff point, as circumstances differ in each case. It would just be something else to discuss when addressing the tiering of those units; a Bartre vs Fiora debaet might indeed depend on whether or not Bartre's contributions during early chapters can be considered significant.

Conversely, if you want to avoid the issue entirely, you can just say that forced chapters are ignored as a general rule for units below upper mid or something similar. Personally, I know I don't care that much about whatever Bartre does in the first few chapters.

Edited November 10, 2009 by CATS

[Dat Nick]

I don't believe forced time should be ignored simply because they are free unit slots and you will be using them and they will be contributing positively [Just you'll be giving Bors/Walt less kills than usual if you're not using them] but I do agree that not a lot of weight should be put on it if you suck during the period where you are forced.

I also don't think "lolavailability" accounts for much if unit B is still available for a notable amount of time and shitstomps unit A when both exists, and it's even stronger in Unit B's case if Unit A barely contributes anything due to his overall suck. So no, I don't think Bartre>Fiora, at all.

[nflchamp]

I don't think people here realize that being bad while you're forced makes you even worse. I mean, if I have two units that suck terribly, but only one is forced, the forced one is worse solely on the fact that I can't choose to get rid of him and his suck.

[Rewjeo]

But he also has no negative because you aren't losing anything for deploying him.

[Florete]

I don't think people here realize that being bad while you're forced makes you even worse. I mean, if I have two units that suck terribly, but only one is forced, the forced one is worse solely on the fact that I can't choose to get rid of him and his suck.

And what about the event where you toss said sucky unit into a corner? I can't think of a single chapter in any FE I've played where a sucky character has to get in the way.

[Rewjeo]

Or that sucky character could hide behind your main forces and run up and do something if it's helpful, thus making them>any unit you didn't deploy for that chapter.

[Dat Nick]

If you have no intention of using said sucky character and this is the only chapter he's forced you can just kamikaze him.

[Florete]

Or that sucky character could hide behind your main forces and run up and do something if it's helpful, thus making them>any unit you didn't deploy for that chapter.

If you have no intention of using said sucky character and this is the only chapter he's forced you can just kamikaze him.

Those work as well. The kamikaze one works very well in 4-4 of RD with Tormod, Vika, and Muarim to run out the enemy Sleep Staff.

[Progenitus]

If the unit is so crappy that the best thing you can do with them is to have them hide in a corner or suicide into enemies, that's a problem. EVERYONE can do that.

[Dat Nick]

Obviously it's a problem, and obviously a unit won't be ranked that high if the most we can do with them is suicide, but it still isn't negative utility.

[cheetah7071]

This is why I think (and I bet nobody will agree with me) tier lists should measure average utility, not total utility. The question tier lists should answer should not be "I'm planning a new runthrough, who should I use?" but rather "I just got this new unit, should I deploy him instead of my previously trained guys?"

[Progenitus]

you mean net utility instead of gross utility?

Obviously it's a problem, and obviously a unit won't be ranked that high if the most we can do with them is suicide, but it still isn't negative utility.

Well of course, it's always possible to use units in such a way that they are never negative utility. For example, Lyre can do a couple of shoves, Fiona can block ledges, etc. But that's not a very interesting premise to debate under.

Edited November 15, 2009 by 8========================D

[Florete]

Well of course, it's always possible to use units in such a way that they are never negative utility. For example, Lyre can do a couple of shoves, Fiona can block ledges, etc. But that's not a very interesting premise to debate under.

Considering this is stemming from bad units who are forced, you could easily say Fiona > Lyre in this situation for having 3 maps (3-6, 3-12, 3-13) of free deployment while Lyre can't exist without tossing someone else out. I don't completely agree with this logic as a method of tiering myself, but it can definitely work.

[Zkirsche]

Well of course, it's always possible to use units in such a way that they are never negative utility. For example, Lyre can do a couple of shoves, Fiona can block ledges, etc. But that's not a very interesting premise to debate under.

And this matters how? A tier list is based off effiency, not play which makes it most interesting to talk about. If it is more efficient to have Rofl to shove a unit than fight then so be it.

[Rodykitty]

I like to use Ronan to force enemies into capturing him.

Auto-Mid Tier.