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

[Tino]

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.

I wouldn't word it the way smash did, but his general idea is rather understandable. How the hell are you going to weigh two shoves against blocking a ledge two times?

[Narga_Rocks]

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.

I wouldn't word it the way smash did, but his general idea is rather understandable. How the hell are you going to weigh two shoves against blocking a ledge two times?

Except that only really matters when you assume there is negative utility for deployment. Otherwise you can just look at what they can do. Like which unit is better with a reasonable (possibly hard to define) amount of effort. With negative utility, Fiona gets 3 free chapters and is thus obviously better than Lyre. Fiona can block ledges in part 3 without forcing somebody out. Lyre can't shove anybody without forcing a better unit to not be deployed. If that isn't a fun way to debate, one should question whether or not to be applying negative utility for deployment.

[Tino]

Don't go argue with me. All I said was that his idea was understandable, not that I necessarily agreed with him.

[cheetah7071]

Presuming this was aimed at me...

you mean net utility instead of gross utility?

No, I don't mean either (or at least I don't think I do). Let's say we somehow come up with a numerical measure of utility. Unit A is available for 2 cchapters, with utility 1 and 3 in those chapters. Unit B is available for 1 chapter, with utility 2. Under the traditional tier system, A is better than B (1+3>2). Under my system, they would be equal (the average of 1 and 3 is 2).

Basically, this just eliminates availibility, and limits Jeigans from auto-topping (because their utility is 0 for many of the chapters they exist) without forcing them below absolute garbage that is only marginally better than an empty slot.

[Defeatist Elitist]

Presuming this was aimed at me...

you mean net utility instead of gross utility?

No, I don't mean either (or at least I don't think I do). Let's say we somehow come up with a numerical measure of utility. Unit A is available for 2 cchapters, with utility 1 and 3 in those chapters. Unit B is available for 1 chapter, with utility 2. Under the traditional tier system, A is better than B (1+3>2). Under my system, they would be equal (the average of 1 and 3 is 2).

Basically, this just eliminates availibility, and limits Jeigans from auto-topping (because their utility is 0 for many of the chapters they exist) without forcing them below absolute garbage that is only marginally better than an empty slot.

But that doesn't really make much sense. That means that if Unit A has Utility 1 throughout the entire game, and at the end of the game, Unit B joins with Utility 1, then they are both equal. Unit A however has contributed much more to the completion of the game, and playing the game without Unit A would be much more difficult than playing without Unit B.

[cheetah7071]

But that doesn't really make much sense. That means that if Unit A has Utility 1 throughout the entire game, and at the end of the game, Unit B joins with Utility 1, then they are both equal. Unit A however has contributed much more to the completion of the game, and playing the game without Unit A would be much more difficult than playing without Unit B.

It does measure something very different from what a normal tier list measures. But, in my opinion, it measures something more useful. I don't care about the fact that unit B wasn't contributing anything while he didn't exist; what I do are about is that when both units exist, they're functionally identical. That's something I can actually use when I'm looking at my deployment screen and deciding whether to use unit A or unit B.

[Interceptor]

It does measure something very different from what a normal tier list measures. But, in my opinion, it measures something more useful. I don't care about the fact that unit B wasn't contributing anything while he didn't exist; what I do are about is that when both units exist, they're functionally identical. That's something I can actually use when I'm looking at my deployment screen and deciding whether to use unit A or unit B.

Using a tier list as a character guide is like using a hammer to put screws into the wall. It'll work, but you have better options, so why make life harder on yourself? Suppose that I was a complete newbie looking for assistance on completing FEDS H5, for example. Which is the better resource: the FEDS tier list, or Colonel M's character ranking thread?

I rest my case. Tier lists are for debates, not advice.

[cheetah7071]

It does measure something very different from what a normal tier list measures. But, in my opinion, it measures something more useful. I don't care about the fact that unit B wasn't contributing anything while he didn't exist; what I do are about is that when both units exist, they're functionally identical. That's something I can actually use when I'm looking at my deployment screen and deciding whether to use unit A or unit B.

Using a tier list as a character guide is like using a hammer to put screws into the wall. It'll work, but you have better options, so why make life harder on yourself? Suppose that I was a complete newbie looking for assistance on completing FEDS H5, for example. Which is the better resource: the FEDS tier list, or Colonel M's character ranking thread?

I rest my case. Tier lists are for debates, not advice.

That's true, but I still feel like my version answers a more interesting question. As I said up above, I'm not really expecting people to agree with me, and I'm definitely not expeting current tier lists to change to this (because it would completely upend pretty much every single list).

[Dat Nick]

Suppose that I was a complete newbie looking for assistance on completing FEDS H5, for example. Which is the better resource: the FEDS tier list, or Colonel M's character ranking thread?

Int, I don't get what you're implying. Are you saying that the character ranking thread is more helpful to a complete noob, or the direct opposite? Because I don't understand why Colonel's rating topic would help more than a tier list.

Edited November 17, 2009 by laws b122

[Narga_Rocks]

Suppose that I was a complete newbie looking for assistance on completing FEDS H5, for example. Which is the better resource: the FEDS tier list, or Colonel M's character ranking thread?

Int, I don't get what you're implying. Are you saying that the character ranking thread is more helpful to a complete noob, or the direct opposite? Because I don't understand why Colonel's rating topic would help more than a tier list.

Well, you could argue a person could read the thousands of posts in a tier list and learn much about what to do, but I think the idea is that a character guide has concise summaries of how to best use a given unit and compares the power of each and gives recommendations about whether or not to use them.

A tier list is a list of a bunch of units ordered from 1 to the number of units in the game. It doesn't tell you to give speedwings to Haar and Titania if you want to obliterate, it doesn't let you know that Mia taking Ike makes a good pair for map destruction without reasonable fear of death, it doesn't let you know that Gatrie needs a crown if he wants significant doubling in part 3. It doesn't let you know Sothe is so high for his footprint on turncount in part 1 and his talents in 3 and that using him as a heavy combat unit in part 4 (or attempting to train him into one) is a bad idea. If you pick units like Titania and Gatrie and Mia because of how high they are and don't use them appropriately, chances are you will be disappointed with all of them. Other units could very easily outperform them, or at least have similar performance. This is doubly true with Sothe in part 4.

A character guide should say:

Titania gets wing

Gatrie gets crown

Mia gets Ike support and she gets Adept for a while

Sothe doesn't get significant training and doesn't do much offensively in part 4

Tormod is helpful in part 1 but don't go out of your way to train him since he's leaving soon

Stuff like that, which a tier list does not have near the front.

Edited November 17, 2009 by Narga_Rocks

[Interceptor]

Int, I don't get what you're implying. Are you saying that the character ranking thread is more helpful to a complete noob, or the direct opposite? Because I don't understand why Colonel's rating topic would help more than a tier list.

That's exactly what I'm saying, and I'm a little taken aback that you don't see Colonel's thread as vastly more helpful than the tier list. I mean, it contains class path suggestions and specifically spells out information such as how units perform, with specifics.

Tier lists have zero of that stuff. As a matter of fact, some of the rankings rely on some pretty esoteric information that's never actually shared in the list, such as an early promotion, or a specific class, etc.

EDIT: you win this round, Moriarty.

Edited November 17, 2009 by Interceptor

[Dat Nick]

what's with the whole Moriarty thing anyway

does it have something to do with Narga being canadian

[Interceptor]

No, it's just a reference to Sherlock Holmes. Professor Moriarty was his arch-nemesis in one of the more famous books. smash likes to call me "Sherlock Holmes without the intelligence", betraying his utter lack of knowledge about Holmes. smash's ignorance is something I find funny. So, I've adopted the title and extended it into a catchphrase for rivals, for my own amusement.

And now, naturally, I've killed all of the humor by explaining it.

Edited November 17, 2009 by Interceptor

[Progenitus]

Character rankings should be reflected in tier lists, and vice versa.

If a unit who is top tier only gets an 8/10 and is lower than several other units, then one of the two (or possibly both) is wrong.

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.

Rofl shoves a few times, Lyre runs off to be cannon fodder, Leo makes a potshot here and there for several chapters, etc.

Do you see what happens to debates between mid/low tiers? It ends up being a bunch of stuff any unit can do, but these guys end up doing little to nothing because if they try to fight or something, they just get in the way.

No, I don't mean either (or at least I don't think I do). Let's say we somehow come up with a numerical measure of utility. Unit A is available for 2 cchapters, with utility 1 and 3 in those chapters. Unit B is available for 1 chapter, with utility 2. Under the traditional tier system, A is better than B (1+3>2). Under my system, they would be equal (the average of 1 and 3 is 2).

Basically, this just eliminates availibility, and limits Jeigans from auto-topping (because their utility is 0 for many of the chapters they exist) without forcing them below absolute garbage that is only marginally better than an empty slot.

Your idea sounds like net utility over gross utility.

[Dat Nick]

Rofl shoves a few times, Lyre runs off to be cannon fodder, Leo makes a potshot here and there for several chapters, etc.

Do you see what happens to debates between mid/low tiers? It ends up being a bunch of stuff any unit can do, but these guys end up doing little to nothing because if they try to fight or something, they just get in the way.

Define "getting in the way". If Rofl chips some guy for like 10-12% of his HP or something then that obviously doesn't count for much, but how did it get in the way of anything? If we are working with a Mid/Low tier unit, their contributions are obviously not as much as high tier units but it doesn't mean they don't contribute. Yeah, there's absolute sucking which pretty much does get in the way [ie, Astrid/Lyre/Fiona/Meg/Arran/Lolenz/Wendy/Sophia bad] just because those guys can't do ANYTHING, but since this is strictly speaking about bottom of the barrel, I don't see why the "bunch of shit anybody can do" or whatever would even come into play. Them doing little to nothing is already reflected by them being low/bottom tier anyway.

[Interceptor]

Rofl shoves a few times, Lyre runs off to be cannon fodder, Leo makes a potshot here and there for several chapters, etc.

Do you see what happens to debates between mid/low tiers? It ends up being a bunch of stuff any unit can do, but these guys end up doing little to nothing because if they try to fight or something, they just get in the way.

Mid/Low tiers? The premise of your post holds no water. There are only four characters in Radiant Dawn who are actually so bad that they end up doing silly things like shoving: Astrid, Meg, Lyre, and Fiona. This is why they are in Bottom.

Everyone else, including those in the bottom of Low, have the capability to be at least somewhat useful when they are around. Your Rolf and Leo examples are pretty out of place, both of those units can contribute measurably to an efficient playthrough when they are deployed. Leo in particular has some tricks up his sleeve that helps speed things along.

So I have no idea what the heck you're talking about. RD has 70+ characters in it. I can't speak for anyone else, but I am OK with 5% of them being so lame that we can't have an in-depth argument about them.

[Rodykitty]

Suppose that I was a complete newbie looking for assistance on completing FEDS H5, for example. Which is the better resource: the FEDS tier list, or Colonel M's character ranking thread?

Correct answer: http://serenesforest.net/forums/index.php?showtopic=9857

But Colonel M's ranking topic is still good to archive reasoning towards characters generally being good or not good (may or may not be the exact reason for a character's placement on the tier list, but can solve general questions as to why <x> is so high or <y> being so low).

[Interceptor]

Suppose that I was a complete newbie looking for assistance on completing FEDS H5, for example. Which is the better resource: the FEDS tier list, or Colonel M's character ranking thread?

Correct answer: http://serenesforest.net/forums/index.php?showtopic=9857

But Colonel M's ranking topic is still good to archive reasoning towards characters generally being good or not good (may or may not be the exact reason for a character's placement on the tier list, but can solve general questions as to why <x> is so high or <y> being so low).

I summarized my argument a little poorly. I didn't mean to put Colonel M's thread as the be-all end-all resource, just something that would specifically address what cheetah was looking for. My understanding of his view: a tier list would be a resource he could use to figure out whether or not he should use a unit at any particular moment in time, aka he's at Ch. 11 in H5 and wants information on whether or not he should add Jake to his army. Etc.

I think that the H5 tier list can't answer this question, but I also think that it shouldn't try to, because it's a tier list and not a character guide.

[-Cynthia-]

I agree with the general premise that time where you're bad isn't as important as time where you're good. Let's take an example like Tormod :

Tormod is good for 3 Part 1 chapters than is well below average for 3 chapters in Part 4.

So someone could say that Tormod's 3 chapters (4-E is about 2 chapters worth) of good cancels out his 3 chapters of his bad and thus is completely neutrsl. However, we could easily just kill Tormod in 4-4 and he'll still have his Part 1 contributions, the negative cancelling out the positive doesn't make much sense.

This is why I have problems with people liek Lehran in smash's tier list being rated higher than Kieran/Geoffrey/Tormod. Yes, intuitively Lehran should be above average since he only gathers minor positive utility. However, this doesn't mean Lehran is automatically in the upper half of characters, since most characters are more positive than negative. Lehran's presence probably won't even save us a single turn overall, since 4-E(5) is generally 2 turned regardless of his presence. However, people like Tormod, and definitely Kieran and Geoffrey, make completing their various chapters much easier, and they have more chapters which are longer. Even people like Edward and Lucia (not countign possible Part 4 performance) save us more turns than Lehran in 1-P and 2-2 respectively. Negative utility just doesn't really make sense.

Edited November 18, 2009 by -Cynthia-

[Progenitus]

bumping this topic because it's important

[Colonel M]

bumping this topic because it's important

Because...?

I mean I can understand that it can be important, but you need to state a reason.

[Anouleth]

3 months old counts as necroposting right?

Well, my 0.02$ is that tier list debates aren't very fun when people assume the default to be 12 Top Tiers and any variation is automatically a negative. For example, fielding FE8 Natasha means Moulder won't get as much staff experience - Moulder is Top Tier, so he's assumed to be used and thus, Natasha is a negative influence on the team. Thus, Natasha goes down because pretty much everyone in FE8 is a 'positive'.

Of course, FE8 is easy enough that Moulder's presence isn't that necessary - so it's very possible to have a team without Moulder. But if Moulder was necessary?

I don't think there are many cases like this, where you have two units fulfilling a specific niche role, one obviously better than the other. Volke/Sothe could be described that way, and Moulder/Natasha.