OMG it's a tier list

[Narga_Rocks]

Those are the problems.

Giving BEXP to Laura is a waste of resources, since she's a lot better off at healing than potshotting, and extra levels don't really help her durability. Ditto for Micaiah, especially since Micaiah caps mag so ridiculously fast for physic/fortify/whatever you want.

And that's why you're Micaiah is nothing more than a healbot, and mine is a goddess in 4-E-2 and 4-E-3, and pretty good in 4-E-4, and a rather good attacker for the rest of the game as far as people with no enemy phase and no doubling are concerned. There are plenty like that, and she's better than them all.

As for Laura, little bit of bexp timing and paragon use and she doubles in a couple of chapters and has pretty good hp/def as far as a mage is concerned.

These people aren't always healing, at least when I play, so they might as well be more than wastes of space when they aren't. It doesn't cost much to train them either.

but even if I didn't train them and Micaiah got to, say, level 15 and Laura didn't get that bexp, I see those 3 units, Nolan, Jill, Ilyana, getting to level 20 tops. Aran started much lower leveled than they did, so getting him to level 20 seems unlikely to me.

Funny, because you refuse to even consider Aran with a robe

duh. No endgame, doesn't need a robe to do what he does in part 3, you seem to love his durability in 1-E, so basically I'm spending a seraph robe for 1-5, 1-6, 1-7, and 1-8. Hardly seems like there is a reason to give him a robe for 4 chapters.

Jill sees significant improvement in those same 4 chapters, has canto so she can actually take advantage of that canto now, actually does something important in 1-6-2 (I'll get into that, later), sees improvement in 3-13, if not 3-6, and has actual endgame potential thanks to 34 speed and canto. Think, those dragons only have so many spots available. I'm sure others might also make a pretty good case for the seraph robe, but I don't see a point for Aran getting it.

I'm pretty sure you don't actually need to take 15 turns to beat both 1-6-1 and 1-6-2. I recall people beating it in more than 15 turns and still getting max BEXP, if that's what you're worried about.

Serenes forest says 15, and aside from easy mode I did it in 15 on NM and 14 on HM. Yes, somehow I got faster as I upped the difficulty. Anyway, it was November when I played EM, so I don't remember whether I got full bexp when I took extra turns. I figure unless someone gets VincentASM to change the thing, the tier list should assume 15 turns.

Oh, and I didn't want to quote one of your later posts, but most steel lance pegs have 11 AS to Jill's 15 and with an iron forge she ORKOs them. This is good if you know where the things will show up and have Jill go there before them. She kills the steel lance guy, runs down to where your other people are, and have them switch her to a hand axe. The javelin users aren't doubled anyway, so 14 AS instead of 15 for her won't matter. Makes things much quicker. Also, I think there is a mistake on the HM data site, but I'll go into detail there, not here.

This is largely because I think beating 1-6-2 in 6 turns is REALLY fast, even if you bumrush with prepromos. Tauroneo can't even reach the boss in 6 turns, and if you're blazing ahead that quickly even Volug/Sothe/Zihark can die.

Who needs Tauroneo? Volug goes North, Jill airlifts Sothe and heads north, on turn 2 she drops him and gets out of enemy range, Volug and Sothe destroy the things trying to get to Fiona, and the boss moves. Possibly even before we are actually in his range, I don't remember. Even if not, Volug and Sothe heal themselves once Fiona is safe and waltz over, draw the boss where as few other enemies as possible can get to you, and on player phase say good night gracie/Laverton. It isn't even the only way, Jill can airlift Taur and let him do it, just needs 31 mt, 32 in a thicket. 8 mt weapon, easy. Trouble is finding a position in which Jill won't get attacked, but others say they can do it,

Anyway, there are like, one 5 unit paladin rush and one 4 unit paladin rush around the bridge, and with Taur holding Eddie and a Bronze Lance, and Zihark next to him I think they priority attack the guy carrying something, though I could be remembering wrong. Easy to mop up, and Micaiah takes out the wind edge guy and if there is another it won't be there till the following turn and she can take it. Careful moving north west gets me a couple of shots at the master seal, though obviously not with Sothe. Generally Zihark and Taur (after Eddie removal) taking attempts. The turn 4 reinforcements move on turn 5 and don't even reach, and on turn 6 I take shots at them with anyone close enough and then KO the boss. I could take 7 turns, I suppose, but I don't see a way to finish 1-6-1 in less than 8 to allow me to spend more than 7 turns in 1-6-2. Plus it just isn't safe at that point. 6 paladin reinforcements is just too many when my wall has to be like 4 units wide and there is still an annoying ballista. Unless having Taur hold something really does cause them to attack him rather than a unit they 3HKO, in which case it might be safe.

So suddenly 0% chance of surviving 2 tigers is comparable to having something WAY higher than 0% chance?

Of course anything is WAY higher than 0%. Heck, 1% is infinitely many times greater, and that is still not actually getting there. You were quoting me from my part 3 discussion, I think, so when choosing between a unit I'm going to actually use in part 4 and Aran, I don't see why this guy should be tanking. Sure, we are forced to use him to the best of his ability in the tier list, but since that does not involve him going to part 4 when the GM's just raised plenty of durable units that will never double, I don't see how giving him experience in part 3 is the best way to use him. Thus, he gets dropped, because 2HKO is passable in 3-6. Aran is better when he's 3RKOd, but he doesn't help long term by tanking there, and he isn't guaranteed to be 3RKO'd.

So, this will be complicated, considering thresholds are 33 hp and 21 def, or 35 hp and 20 def, or 37 hp and 19 def, and then it's a bunch of 0 for more hp. We can look in the other direction as well, but that gets ignored because I'm giving him 20/1 and he can't have 22 def. Of course, with 20/2 I have to include the chance of 31hp and 22 def, but that doesn't really help because I also need to calculate 19/2, and I can't with that list.

Anyway, I'll look at 19/1, 20/1, and 20/2 and calculate his chances of survival. I might do it for Boyd, but that depends on when I get done with the rest. (Oh, and those levels are only possible with a Micaiah doomed to be a healbot and a Laura doomed to be dropped after part 3 making Oliver and Bastian your next best 4-E healers if you want more than just Micaiah. Not only that, we aren't even likely to keep using all 3 tier 1 units we raised when part 4 comes around, so why not just raise Jill and Nolan if they are going to endgame, or possibly not even them if we aren't? No real reason not to get Micaiah to level 20. I suppose if I do this for Nolan and Aran, for example, I can actually get Aran to a reasonable level, but I wonder if three additional 19/1 units in 3-6 helps more than two additional 20/1 units and one additional 14/1 unit. But I digress.)

So, using this:

http://www.feplanet.net/games-10-characters-averages/8

http://www.feplanet.net/index.php?fep=game...;id=8&key=1

http://www.feplanet.net/index.php?fep=game...;id=8&key=7

Reader's digest version

at 19/1 he's got a 20.72% chance of being 3RKOd by 39 mt tigers,

at 20/1 he's got a 34.94% chance of being 3RKOd by 39 mt tigers,

at 20/2 he's got a 65.35% chance of being 3RKOd by 39 mt tigers.

It is a monumental jump from 20/1 to 20/2, but a lot of people don't think he'll even have level 19 by 3-6, and 19/2 is likely a fair bit less than 40% but more than 34.94%, because def is uncapped but otherwise there is really no change between 19/2 and 20/1.

For a 39mt tiger and a 41 mt tiger,

at 20/2 he's got a 34.71% chance of not being 2RKOd.

For two 41 mt tigers,

at 20/2 he's got a 10.62% chance of not being 2RKOd.

On to the calculations:

Anyway, I had a stupid math idea before where I was double counting stuff, but I figured out the proper way, so deleted about an hour of worthless calculations. Oh well. This is the correct way:

so, at 19/1, his chance of having:

>= 33 hp is: .39 and 21 def is: .49

>= 35 hp is: .07 and 20 def is: .23

>= 37 hp is: 0 and 19 def is: irrelevant

So, to survive he needs at least 33hp and at least 21 def, or at least 35 hp and at least 20 def.

That's .39 x .49 + .07 + .23,

The total for being 3RKOd at 19/1 is 20.72%.

This is from 39 mt tigers while having an A support, or B Leo/Ed/Jill.

Note that of those who only need a B support, Ed/Leo will likely not be deployed in 1-7, 1-8, 1-E, and Jill is now restricted to staying near Aran and not flying off on the river where she can help without being attacked. Also restricts her in 3-12 and 3-13, and means if they don't both go to endgame or don't both not go to endgame, she needs to rebuild a support, or he does.

So he needs A support in 1-3, 1-4, 1-5, 1-6, 1-7, 1-8, 1-E.

Doable, but likely only with units that start in 1-4 or before. B Jill is also likely simple enough, if not for Jill carrying Sothe making it much easier to save Fiona and friends, Jill's tendency to fly to ledges before pegs appear, Jill not being in 1-8, and starting to build in 1-6. Ouch. Likely not happening either.

So, just the guys in 1-4, meaning Sothe, Nolan, and Micaiah, and Laura, unless you want Ed, Leo, and Meg taking a slot away from their betters in 1-7, 1-8, and 1-E. No wonder people always give him Laura in comparisons.

Next up, 20/1.

>= 33 hp is: .5 and 21 def is: .65

>= 35 hp is: .13 and 20 def is: .18

>= 37 hp is: .01 and 19 def is: .10

So, to survive he needs at least 33hp and at least 21 def, or at least 35 hp and at least 20 def, or at least 37 hp and at least 21 def.

That's .5 x .65 + .13 x .18 + .01 x .10,

The total for being 3RKOd at 20/1 is 34.94%.

This calculation, by the way, is equivalent to:

% for 33 hp X % for 21 def + % for 34 hp X % for 21 def + % for 35 hp X % for 21 def + ... + % for 37 hp x % for 21 def

+ % for 35 hp X % for 20 def + ... + % for 37 hp X % for 20 def

+ % for 37 hp X % for 19 def

Since the formula can be factored to:

(% for 33 hp + ... + % for 37 hp) x % for 21 def

+ (% for 35 hp + ... + % for 37 hp) X % for 20 def

+ + % for 37 hp X % for 19 def

Which is what I'm calculating when I use the numbers listed above.

That said, time for the estimate of 20/2.

So, to survive he needs at least 31 hp and 22 def, or at least 33hp and 21 def, or at least 35 hp and 20 def, or at least 37 hp and 19 def.

So now for a more exhaustive calculation, simply because I typed this before figuring out the best way to do it, then went above here and fixed it. I don't want to type the short way when the long way is already there.

31hp/22def: .16 x .46 = .0736

32hp/22def: .18 x .46 = .0828

33hp/22def: .21 x .46 = .0966

34hp/22def: .18 x .46 = .0828

35hp/22def: .12 x .46 = .0552

36hp/22def: .06 x .46 = .0276

37hp/22def: .02 x .46 = .0092

38hp/22def: .01 x .46 = .0046

sum= .4324

33hp/21def: .21 x .32 = .0672

34hp/21def: .18 x .32 = .0576

35hp/21def: .12 x .32 = .0384

36hp/21def: .06 x .32 = .0192

37hp/21def: .02 x .32 = .0064

38hp/21def: .01 x .32 = .0032

sum= .192

35hp/20def: .12 x .13 = .0156

36hp/20def: .06 x .13 = .0078

37hp/20def: .02 x .13 = .0026

38hp/20def: .01 x .13 = .0013

sum= .0273

37hp/19def: .02 x .06 = .0012

38hp/19def: .01 x .06 = .0006

sum= .0018

total=.6535

So, to survive he needs one of the combos listed above,

The total for being 3RKOd at 20/2 is 65.35%.

Also, for surviving a 41 mt tiger + a 39 mt tiger, he needs >=33hp and 22 def, or >=35hp and 21 def, or >=37 hp and 20 def.

I'm just going to look at 20/2, because I'm sure the lower level options are just sad.

33hp/22def: .21 x .46 = .0966

34hp/22def: .18 x .46 = .0828

35hp/22def: .12 x .46 = .0552

36hp/22def: .06 x .46 = .0276

37hp/22def: .02 x .46 = .0092

38hp/22def: .01 x .46 = .0046

sum: .276

35hp/21def: .12 x .32 = .0384

36hp/21def: .06 x .32 = .0192

37hp/21def: .02 x .32 = .0064

38hp/21def: .01 x .32 = .0032

sum: .0672

37hp/20def: .02 x .13 = .0026

38hp/20def: .01 x .13 = .0013

sum: .0039

For a total of: .3471

So even at 20/2, he's only looking at a 34.71% chance of not being 2RKOd by a 39 mt tiger plus a 41 mt tiger.

This will just be low, and not relevant since not many others have a chance here either, but:

Chance of being 3RKOd by 41 mt tigers at 20/2:

Needs >=35 hp and 22 def, or >=37 hp and 21 def.

35hp/22def: .12 x .46 = .0552

36hp/22def: .06 x .46 = .0276

37hp/22def: .02 x .46 = .0092

38hp/22def: .01 x .46 = .0046

37hp/21def: .02 x .32 = .0064

38hp/21def: .01 x .32 = .0032

Let's just go with total:

.1062

Or 10.62% chance of being 3RKOd by 41 mt tigers at 20/2.

So 20/2 means a 65.35% chance of being 3RKOd by 39 mt tigers, but even so, for any other combination of tigers he's got a >65% chance of being 2RKOd.

Considering we generally look at 65% hit and go bleh when deciding whether to lay a unit's life on the line hoping for a hit, it is better than anyone who is 2RKOd no matter what, and slightly better than a unit that needs resources to get there.

The thing is, 65% is also much to high for giving him stat boosters. Think about it, if you give him a seraph robe he's looking at a 35% chance of swallowing it up in part 1 and never getting a benefit from it in part 3. Well, except against really weak cats.

This 20/2 Aran has a 64% chance of having 16 or less speed, as well, so he's getting doubled by cats most of the time and will likely be killed by even a cat if you give him a seraph robe, because it alone won't get him enough hp to survive anything after taking two tigers.

So to comment on your comment about me not offering him a seraph robe, uh, cause it's a waste? The %ages will only increase after 20/2, just look at the improvement from 20/1 to 20/2, so by turn 5 in 3-6 the robe is going to waste, if you somehow got him to 20/1 before 3-6.

Since you like arguing for Boyd, I'll give you his % chances of having his average speed or more, just to show you that Aran's chances for hitting those HP/def/whatever averages are high.

http://www.feplanet.net/games-10-characters-averages/49

For example, Boyd has a 64% chance of having 22 or more spd by 17/0.

Yet you have no qualms assuming Boyd will always have average speed, even though if he's even slightly spd screwed he just became a giant waste of exp.

I don't know, that 64% chance seems pretty close to the 65.35% chance that Aran is pulling to be 3RKOd by 39 mt tigers at a level that most people don't think he'll reach until late 3-6 at best. If he manages, then he's looking at matching Oscar's speed in 3-11 with just a speedwing. If Aran is going to be great in 3-6 with his 65.35%, then what about Boyd? Now, while you could say giving Aran a seraph robe would fix the 34.65% of the time he isn't cutting the mustard, and since Boyd is getting a speedwing why can't Aran get something. I got into that earlier, but really if you want Aran to get a seraph robe it seems like a waste in part 3 two thirds of the time and in part 4 he'd rather have been given a speedwing than a seraph robe. You are effectively dumping a seraph robe for making him better in 4 chapters of part 1, namely 5, 6, 7 and 8. As if other units can't get more use out of one.

So one extra unit can be 3HKO'd and suddenly Aran is useless? wtf?

Yes. Well, not useless. Just redundant. You've got Volug and crown Tauroneo, and you only need two guys to block. Heck, if you want Volug to go on a resolve rampage or if he's got double earth and you wanna take a bit of a risk, you now have Nolan and possibly Jill that can survive. Unless you want to argue Aran should be used in part 4, I hardly see the benefit of letting him gather experience rather than someone who can continue to be used. And if you really want to use someone who can't continue to be used, then Tauroneo didn't need to be raised throughout part 1 in order to get to 3RKO, he just needed a crown.

I see this as perfectly reasonable for upper mid, because he is one of our options. I'm not trying to make him neutral for the fact that Volug, Tauroneo, Nolan, and Seraph Robe Jill are all likely 3RKOd now, especially since if Nolan or Jill get a bit screwed they might not make it at the beginning of the chapter. I'm just saying, when you tie a bunch of people, even if it's near the top, and this is supposed to be your crowning moment, then it isn't a very high crowning moment. Frankly I'm not sure whether Aran > Boyd or Boyd > Aran,. Basically, though, when you have no endgame and need to be raised with middle of the pack offence, and never really get much higher than that since you never double, tying the people around you for a few chapters does not do a lot to make up for it. Especially since using Aran makes Tauroneo redundant, but Aran took work to get up there and Tauroneo didn't. With your whole Rolf A vs. Rolf B argument where Rolf A needs to be raised from 3-P and Rolf B starts at 20/15 in 3-11 or something and never needed to be raised, and Rolf B was supposedly clearly better than Rolf A, isn't crowned Tauroneo clearly better than Aran? At least for 3-13? And obviously 1-6. And Nailah and BK didn't take work and they are better than him in 1-E. And so on.

less than half the map? wtf? He's recruited on turn 3. You finish 1-3 in 5 turns?

That was typed before I thought about it. I looked back and decided it works anyway since a bunch of units are killed before recruiting Aran, he won't face the right half if we go left, nor the left half if we go right, and we are likely drawing the boss with Sothe then Thani bombing it. Frankly, he is facing a few enemies on the left if we use him, and he is 3RKOd by most, though the 25mt and 24 mt guys can combine with any 22 mt guy to 2RKO him if he isn't on a thicket. The point is, he isn't exactly tanking 3 or 4 on enemy phase, so he is limited when there can be that many, which is generally the first 2 turns after he's recruited, then there aren't too many enemies as we go to the escape square. A person posted 4 hit and 4 kill to get him a level, but I'm not sure that will happen. With the enemy concentrations here, he can get 2 kills soon after recruitment while Sothe and Nolan are tanking and other units are finishing kills, but then it's almost clear sailing to the escape square and I don't think he'd get much. If we face the reinforcements, sure, but there's a reason they aren't even listed on the hm data. It is because we shouldn't be around where they show up by then.

So because exactly two people can do it better than Aran, that makes Aran worthless.

When we need precisely two people to do it, and they cause more damage making it easier to finish off the leftovers? Yes, very much so. You didn't even address the idea I sent out that letting Aran wall there rather than Nolan or Sothe do it would prevent finishing in 10 turns. You'd rather say this:

I guess that means only the top 10 units in the game are only ever played, because we never need to use anyone else.

One difference between what you are saying the logic leads to and what I'm doing is that we are forced to take them along. Now, you did make an attempt:

Nevermind that we might want Sothe to jump ahead of the frontlines and take care of an annoying enemy, like the boss who moves or an archer that can attack my tanks uncountered unless they're using crappy 1-2 range weapons.

Enemy density. There are two armors blocking on the left. Once you kill them, the boss comes as well as a couple of other things because if you get close enough to take on the armors (that don't move, btw), you are in their range. So yeah, I'd rather attack with Sothe and Nolan, considering when I attack the things, I'm looking at it being very unlikely I'm able to send Sothe in before the boss moves. One thani bomb later, a few of my guys grouping up on that archer and I'm pressing forward and the next time I need to wall it's one square and Sothe is far more likely to be the one put on it. At least the first turn we use it. Afterwards it could likely be Nolan or Aran if we try to get to the coin, but otherwise we are just escaping at this point.

50ish% chance to be able to survive a huge number of combinations of enemies is actually pretty damn high for debates.

Now to 1-4, right?

Remember, 5 cats only? And there is only once in the entire map in which there are 2 cats that will attack at once, otherwise it is cat/tiger combos? That one is pretty minor, since it could happen once, and it requires moving slowly to prevent them from attacking two different units instead of focusing on one (because if we go quick, we can put our guys side by side).

Now, for the other 57.75 % chance of being 2RKOd, lets take a look:

5/17 cats

4/17 24 mt tigers.

You realize the probability of facing two in a row is low, right? I don't remember positioning, so going off pure probability, we are looking at 5/17 x 4/16 + 4/17x5/16, at least I think so. The reasoning is, he can face a cat first then a tiger, or a tiger first then a cat, and that once he has faced an enemy, there are only 16 left that can attack.

So multiply it out:

5/68+5/68,

which is 5/34.

Which is 7.35% chance of facing that exact combination. Now, obviously as the chapter goes on it changes, and I ignored how part of that 17 is 4 reinforcements, and enemy positioning, but that also means I ignored how the chapter starts with just 3 cats.

Still, that is a pretty low probability of actually facing this other combination that has a 57.75% chance of not killing him.

Every single other combination that could be faced, ie: the vast majority of combinations likely to go against him, is pulling a 65% chance of 2RKOing him.

For example, let's say a unit has 95% true hit on enemies. After just 10 attacks, you actually have ~40% chance of missing one of your attacks. Is 95 true hit suddenly not reliable?

Yes, over time I'd rather have 100%. Whenever possible. I'll take a 95% if I don't have a choice, or if it's near the beginning of the chapter, or if I have another unit available to kill the thing if my idiot misses, or if I want to roll the dice, but I'd rather have 100%.

Btw, to continue this fun with random statistics, you have a 64.15% chance to miss at least once with 95% true hit after 20 attempts. Pretty high. Another fun one, after 30 attacks I'm looking at a 78.5% chance to miss at least once.

There's a reason it happens to people. Getting to the last enemy on a map with only one unit left I can use and having only a 95% true hit and having the guy able to kill a unit of mine either on a counter or on enemy phase means that I might have to redo 10 turns, or whatever.

Fun fact, if 1000 people go through a playthrough of a game like this with 30 chapters, and do something silly like on every single chapter attacking on the last turn with 95% true hit when missing means death, then most likely over 700 people are going to be whining about RNG screwage, and the other ~50 people it happened to understand statistics and know they had a pretty good chance of it happening to them because of what they were doing. Of course, these ~50 people who understand statistics likely didn't do something silly like that.

I can randomly bring up statistics, too.

Funny, because you only talked about Tormod.

You didn't talk about Muarim/Vika, probably because you knew they had transformation gauges and can't even tank half the time, especially Vika who has 38 HP/7 def untransformed.

No, I didn't talk about anybody else because I thought Tormod was the only one for whom it wasn't obvious that he does better. Muarim and Vika are likely to be assaulting the enemies near the bottom right room after Tormod takes out a mage on a ledge (possible 6 attacks between player phase/enemy phase/player phase and needing to hit twice with elfire, no problem). They can then revert and head towards the enemies in the top left, and will retransform before getting there, and slaughter them, too. Meanwhile, the rest of your guys go around opening doors and Micaiah and one other unit head to the boss. Assuming this other unit is Sothe or Volug, Micaiah finishes up the 2RKO combination with a thani bomb, and once the last soldier leaves, end of chapter. Trust me, their gauge is irrelevant. And who needs tanking when very few enemies actually move?

The sad thing is that Neph isn't far behind Boyd either, and Soren is actually comparable.

You do know that Neph is better than him, right? She starts 2 or 3HKOd by enemies, but will double and with a forge is 3 or 4HKOing, so 2RKOing. She doesn't even need help to improve on that, and will soon have a solid 3HKO on everything she doesn't ORKO, except Swordmasters are 2HKO without doubling and generals aren't as great for her. A little loss in durability against some enemies isn't so bad.

The difference between Boyd and Aran is that Boyd is getting massively OUTCLASSED by more than 5 guys in part 3, which apparently makes Aran worthless in every part 1 chapter because he's losing to like 2 guys in 1-3 and 1-4, and then like 3-5 guys for the other part 1 chapters.

List of d00ds Boyd loses to in part 3...

Haar, Ike, Titania, Shinon, Gatrie, Mia, Oscar, Ranulf, Ulki, Janaff, Mordy

Oh, and Reyson counts too, since when you consider second actions every turn, my uber guys can extend their leads by even more (for example, if unit A is hitting enemies for 100% of their HP while unit B is hitting for 50%, this means in one attack unit A is doing 50% more HP damage, but after 2, it increases to 100% more HP damage). Rafiel, on the other hand, for the DB, only really does that in 1-E, since in 1-8 he's in the middle of nowhere and is stuck chanting Nailah/Volug.

But apparently, losing to this many people can still net Boyd neutral utility, yet Aran losing to a few people in part 1 suddenly means he isn't?

Never said Aran is't giving at least neutral utility, he gives positive or neutral in most of part 1, just not ever a lot of positive, because of the gods.

Since you can't see Boyd as neutral utility,

It's time for Narga's Economics Lectures part 2.

(disclaimer: I'm a math major, not economics, and have only taken 3 econ courses, so any cringing econ majors can give corrections if I did anything silly)

Today I'm discussing the law of diminishing marginal returns.

The idea starts with a random factory, that for some reason has 4 machines, each of which can be operated by one person, and we will call these capital, and K = 4.

Then we have the option of hiring employees, who will use these 4 machines to make a product.

The idea is, as we hire more employees up to 4, we get pretty big gains. Then when you add a 5th, you still get a pretty good gain, but not as much as the first 4 gave us. The 6th employee is now giving even less.

For random numbers, lets say

5

6

7

8

5

2

2

Now, this is the marginal return from each extra employee.

With 4 employees, we made 26 things.

With 5 employees, we made 31 things.

With 6 employees, we made 33 things.

With 7 employees, we made 35 things.

Now, I've never figured out what number 6 and 7 are actually doing, in all the lectures I've been to. I figure employee 5 is filling in on other people's breaks and stuff, so that is an improvement, but with only 4 people and 8 hour days it is unlikely they are on break for 2 hours each in those 8 hours so that's why the 5th employee's marginal return is a 5 not more, because it doesn't have access to the machine as often.

I'm assuming employee 6 and 7 are basically just gophers and clean up and various things that is boosting efficiency by not much and so only improves the total production by a very small amount.

The fun thing here, is that it doesn't actually matter whether employee 6 and 7 is as skilled as the first 5 or not, because they never actually need to go on the machines and do stuff. They are just getting coffee, cutting paper, cleaning up garbage, whatever.

The same logic can be applied to Fire Emblem.

If you have 7 units who all ORKO 90% of the time and are 10RKOd in return, but the chapters are mostly set up in ways in which only 4 of those units can actually have enemy phases each turn, and there aren't enough enemies nearby for player phases where all 7 units need to be able to ORKO, then it again doesn't matter whether we replace those 6th and 7th unit with a couple of guys that aren't as good. They will be able to clean up, grab coffee, cut paper just as well as the original 6th and 7th unit.

By the way, only 7 of those units are in 3-2, and we can deploy more than that.

And whatever happened to the smash that was complaining about phantom Narga suggesting you always bring only high and top tiers? I even agreed with you that it wasn't something that should be required. So if we aren't supposed to always bring high and top tiers, then isn't Boyd pretty good for the spot we leave open for him?

Anyway, the idea behind neutral utility for Boyd, is that he can clean up just as well as anyone else, and he isn't going to be a liability because at most he should face one or two guys if he cleans up. And since it is apparently bad that those 11 units are better, then if he hits a couple of things on enemy phase, can't they kill them? 5 of the guys you mentioned have canto, and all should be able to finish off something Boyd hit, unless he hit a general. All but Oscar should be pretty close to OHKO on whatever's left. The only one who might have issues sometimes would be a dropless Ulki.

Having 7 or 8 guys better than him, when only 4 or 5 of those will typically be able to fully exercise their greatness, will mean on most turns, Boyd can do exactly what the other 2 or 3 guys are doing, at almost no detriment to speed, efficiency, or whatever. So that's why he can have neutral utility. Guys like Soren, Ilyana, Rolf, Lyre, Kyza, Lethe, eventually Astrid and a couple others, aren't able to do exactly what the other 2 or 3 guys are doing. Mostly because they either don't always have a player phase, or they don't have as much freedom of choice in where they attack from, or if they have an enemy phase they are suddenly causing bad enemy phase damage and some of the high tiers even won't be able to OHKO the weakened thing. Some of those guys I mentioned can pull out of the hole they are in, like Soren and Rolf, and eventually Ilyana, though having 2 range helps all 3 of them a bit. Others can't, because of other issues. The point is, Boyd is neutral at worst until he becomes good. Then he's no longer neutral. Plus, since he actually has a 34 speed cap, there is a goal in mind of training him, unlike Aran, who is effectively gaining experience in part 3 and not being used much in part 4, so why get so much experience in part 3? Unless of course you drop all the dbers by part 4.

And that neutrality only lasts until he starts doubling paladins and generals and other things. At this point, his offence is typically better than Oscar's.

All it takes is a speedwing, and that's for getting better than neutrality. Without a speedwing he's still neutral.

I have to stop now, so I'll finish up some other day.

I left out important things you said and was hoping to get to them, but I can't. Needless to say, I haven't even attempted to justify the speedwing thing yet, but I will later.

Edited July 15, 2009 by Narga_Rocks

[Interceptor]

~6.2/10, overall. You lost a fair amount of points because I'm not sure I see the turncount reduction. It might be possible I'm not seeing something, though.

Uh, thank you for responding, but I wasn't looking to have my rant rated for thoroughness and accuracy, I was trying to start a discussion about tier philosophy. I'll address some of the things that you brought up (like reducing turn count) in another post. I was just hoping that someone besides smash would bite, because it's not really easy to have a reasonable discussion with him.

Where would you want to see him moved, then?

My feeling on his placement depends on how things are weighted. See my earlier rant about the DB.

Except he can’t pair both resolve and pavise, and the former’s effect is much greater (25+ avo jump that is compounded by the 2RN system).

Here are two more bullet points for you to focus on to the exclusion of the rest of this post:

1) Resolve is only improving his avoid when he's one hit from death. In short, his chance to get hit becomes his chance to die. I don't think that this is necessarily superior to what Pavise could do, since that skill is also helping on the first hit (both Enemy Phase and on counters). Pavise reduces Zihark's chance to have to get healed, Resolve just reduces his chance to die. It's important that you understand this distinction.

2) Pavise is more or less a purely defensive skill. Resolve is dual-purpose. The salient point here is that sending Pavise to the GMs is much sillier than sending Resolve to the GMs, since Resolve can also be used for boosting offense. The DB is not terribly efficient at using Resolve on offense other than limited Player Phase activity, while the GMs have higher chapter availbility and several good Resolve candidates.

And who here is denying that 3-6 is Zihark’s roughest chapter?

Who here accused them of it? I said that people downplay the difficulty, I didn't say that they flat-out claimed that 3-6 wasn't his hardest chapter.

Vykan, I am very precise with word choice. Whle I don't always succeed at it, I make an effort to ensure that I am not stating things that I don't want to claim, and also that my real point is properly articulated. It really annoys me when you hamfist your way through one of my posts, quote a bunch of things out of context, and then either misrepresent my point or infer something that I never implied. If I was making a post and said "I don't like M&Ms", you would come back with one that said "So what are you saying, that you don't like candy?", and I would /facepalm.

i love how int always says my arguments are strawmen. [...] Do you actually know what strawmen are?

Not all of your arguments are straw men, and not even most of them, but I'd say you're in the top 3 offenders amongst the regulars here. It's not something you'll ever admit, but that's only because you are unable to recognize how often that you "refute" positions that are only superficially similar to the actual positions held by the people you are arguing with. Naturally, you think that your interpretations are the correct ones, and brook no criticism to the contrary, but I'm hardly the only person who has noticed this behavior.

As for whether I know what strawmen are, I'd say that I'm at least as knowledgeable about them as you are about the definition of "trolling". Your move. Mr. Bond.

the fact that Int is a dumbass[...]

You'd have to be a dunce like int[...]

just to show that int is a tool[...]

Int is just a mindless Aran hater[...]

are you fucking retarded or something?

Fascinating arguments, these.

Edited July 15, 2009 by Interceptor

[Don Draper]

No further comment on the debate, but smashfanatic really needs to look up the term "ad hominem attack".

Edited July 15, 2009 by IOS

[Interceptor]

OK, I think I'll take a stab at smash's post. There ought to be enough red meat amongst the insults and straw men for me to actually find something to respond to. I have a feeling that I'll be doing a lot of "no, that's not what I said", though. Let us find out.

First, he claims that "magnitude" is the only factor when it comes to "ass-whipping".

Well, that didn't take long. No, smash, that's not what I said. The word "only" does appear in my post, but not with respect to the magnitude of ass-whipping. AKA, I am not asserting that my list was exaustive. Length was not germane to my point (because of the second issue that I brought up), so I didn't bother to mention it. I'll comment on it now, though.

It's not, he's also ignoring how LONG they're ass-whipping. Tauroneo/BK/Nailah are only around for 1-2 chapters (and even for Nailah, one of those chapters is 1-8, where they're on opposite sides of the map and not really competing against each other in anything). The people beating the shit out of Boyd, however, are people like Ike and Haar, who are around for basically FOREVER.

You know, Nailah has 9MV and flies through thickets like they weren't even there. For you to assert that they are starting on "opposite sides of the map" (which is wrong: the boss is on the opposite side of the map realative to Nailah. The DB starts in the center) is to ignore that Nailah can basically be standing on top of the DB on Turn 1. So much for geography.

The BK/Nailah/Tauroneo are not the only units that are wrecking Aran. I simply brought them up because they are not just superior to him, they are incredibly superior to him, and therefore illustrated my point about magnitude. Aran is still getting beaten pretty badly in many of the chapters where these god-mode units don't exist.

Next, he greatly exaggerates the leads the prepromos have over Aran.

I did not attempt to accurately measure Ike's lead over Boyd, I just back-of-the-napkin'ed it just to make a general point. You make a few leaps here that I think are really bad:

1) You're ignoring that "never dying" is better than "takes a lot of pepper to die". Tauroneo basically never has to be healed by anyone (no healer even in danger around him), and frankly never even needs to give up a Player Phase to pop a heal. You took the worst case (T is 8HKO'ed) and assumed it was the average case (it isn't: I picked out the baddest unit I could find in 1-6-1. Tauroneo is actually 10-13HKO'ed by a lot of things, where he is not infinity-hit-KO'ed due to taking 0 damage, which is pretty ridiculous since his avoid is also good).

2) The ability to OHKO or to ORKO while being invincible also translates into the ability to ignore counters, which is a luxury that I assure you Aran does not have.

3) You're either overestimating Aran's level, or underestimating the cost of getting him to that level, take your pick. I'll let dondon continue to fight that one out with you. I was hoping that you'd get into the fight last time it came up, but now you're forced to, so I will sit here and enjoy it.

I'll give you credit for acknowledging the 2-range advantage that T and the BK have over Aran, which is surprising.

The bottom line here is that these units are infinity better than Aran on paper, and many, many times better than him in practice (reality only allows you to do so much).

So, can we conclude that Boyd gets shitstomped by the top tier GMs far harder than Aran gets shitstomped by the top tier DB? The answer is an astounding yes.

I disagree with your conclusion. I can manipulate lots of levers in Boyd's case to make him improve, and my Ike estimation was intentionally high (I don't think that he's 4x better than Boyd). I can't do much for Aran except give him a forge (which you already assumed, given the #HKO number you cited) and make it a bit harder for him to die.

Now, not only that, in previous arguments, int claims that Boyd getting 3-4HKO'd is perfectly serviceable. Which would imply that durability leads beyond 3-4HKO are largely superfluous, since 3-4HKO is okay.

If that's what you inferred, you are wrong. Serviceable means that I can do something with it. It does not imply that there is little or nothing to be gained by being even more durable. I think that is it very sad that you misrepresentation of this point caused you to waste so many keystrokes on a silly tangent.

Why is it bad for Aran to be getting outclassed when he's getting 3-4HKO'd, but it's fine for someone like Boyd to get 3-4HKO'd when he's getting outclassed?

Put simply, the cost to train Boyd is smaller, and he has better prospects than Aran once trained. And again, it's not "bad" that Aran gets 3-4HKO'ed any more than it's "fine" for Boyd to face something simliar: it's the amount of effort it takes to train Aran compared to what you get out of it.

Now before anyone strawmans me, I do think Aran getting 3-4HKO'd in 1-6 could be better when you have Tauroneo godmoding, and Volug beating him, and Sothe doing about the same (I think they'll have similar HP/def, and Sothe has avoid). But the number of people beating him is far less than the number of people beating Boyd's 3-4HKO, which is why Aran's durability is superior to Boyd's relative to the team.

This is basically a non sequitur. You make the jump from "fewer people are beating Aran" to "Aran's durability is therefore superior to Boyd's relative to the team". You are missing a bridge. Boyd's durability gap matters less than Aran's does.

I can generally beat part 1 within the turn limit for max BEXP for all chapters except for 1-2 (to get the chests) and 1-4 (get the hidden items) and still get the DB I want to use to close to level 20 by the end of 1-E (Nolan was actually close to level 20 by the START of 1-E so I threw an early seal on him), so I was going pretty fast through the chapters already. While I do say I could have gone faster if I had my prepromos solo maps instead of my usual "avoid using them if possible", I seriously doubt I could have gone so much faster in part 1 to the point where it will override making part 3 and 4 easier and faster for me because I don't have underleveled pieces of crap. To give you an idea of how fast I was already going in 1-6, 1-7, 1-8 and 1-E, I generally had Micaiah use her full move every single turn to move forward, and so you can't get THAT much faster than that other than "Nailah rescues Micaiah and blasts through 1-E and wins", which is pretty retarded since now you not only lose out on CEXP, but the items in the chests too, and it's only possible in 1-E.

I understand your feelings, but you are wrong. Let me explain on a chapter-by-chapter rundown:

1-2: Laura arrives. Sothe is a god, it's easy to get her to the tile ASAP once you can use him on Turn 3. You can choose to extend the turn count to gather all of the items, but it does not really help you beat the game faster to do a full clear.

1-3: You can either go left or right. I prefer right so that you can backdoor the enemies and avoid recruiting Aran, but in either case it's straightforward to have Sothe godmode his way down there and escape on the exit tile.

1-4: Sothe can solo one side, Nolan + healer can solo the other side, Ilyana/Micaiah/Leo can take the top by breaking the wall on the first turn and having Eddie or Aran or Meg or whomever attract the boss for a Turn 2 kill. Sothe should go left for Beastfoe, probably.

1-5: Turn-limited, and Volug is doing all of the hard work.

1-6-1: Tauroneo, Sothe, Zihark, and Volug are doing all of the heavy lifting here. Four prepromos are more than enough for a Rout, especially if you use Jill (who does not need to be trained) to deal with the pegs.

1-6-2: Jill can air-drop Tauroneo on the boss island by Turn 2 (I swear to god I'm like the fourth person who has pointed this out to you now). This map is just a Boss Kill, so you just have to save Fiona and maybe get the Red Gem or something while T is going Rambo on everything with a javelin.

1-7: Just a Micaiah Arrive, and the LEA helpfully fills in the gap left in your prepromo line by T leaving you. If you care about Turn counts, you don't care about the prisoner BEXP here. It's a nice leisurely run to the boss while your super-units obliterate everything.

1-8: A rout with the same group of badasses from 1-7, plus Nailah and Rafiel. Say goodbye to your face. Vika can cheat-whore her transform gauge.

1-9: BK clear without Micaiah even getting attacked. She could be level 1 and you'd be fine.

1-E: Another Arrive. Now you get every badass in the whole Part in the same chapter. Nailah can Pass + ferry the BK and the two of them obliterate everything in sight. Micaiah heads for Arrive, Sothe picks out whatever chests you want (I'd get the Speedwing, the other stuff is crap).

3-6: Easy Button BK clear. I think this is 13 turns with no problem. Slightly less if Sothe helps.

3-12: ZiharkxVolug support, with T-crown/Sothe helping and Laura/Micaiah healing, plus a ramp that only requires a whole two units, fighting tier 2 beorcs. Yawn. Resolve possibly justified in this scenario because of the effect on Turn Count.

3-13: Two meaty units at the stairs, scrubs everywhere else, heal-spam, manually control ballistas with Roam/Halt, try not to fall asleep before the 12-turn limit is up. I guess if you are feeling lucky you could try for a quick clear by air-dropping T in front of Ike and hoping for a Luna.

It's not really required to have DB units for Part 4. You only need your best 10 for Endgame, and the game gives you plenty of pretty hax units for 4-P through 4-5. Also, you have the GMs, which can be trained for this purpose. I guess if you really cared, Zihark or Volug might be salvagable with resources, and Tauroneo is better than a kick in the nuts as long as he stays out of Endgame.

Also, the idea that the DB have no one worth training and thus we don't need them to bolster our ranks is idiotic beyond belief. Let me make a comparison with one of int's favorite units.

Sorry, but that's not what I said. I wasn't claiming that it's not worth training the DB (obviously whether it's worth it depends on your goals), but instead that training the DB costs more than training the GMs under the criteria that I outlined. Your NolanxMia comparison is a great example of this: Getting Mia to Trueblade is a helluva lot easier on my turn counts than getting Nolan to 20/15/1 Reaver, because getting Nolan to 15 before Part 4 basically requires that I slow down Part 1 to focus on him, and then use Part 3 to train him on laguz. Mia can get to TB just by taking kills as the GMs roll Part 3. I mean, she only needs to level up 13 times, plus promote. I could even pull the same crowning BS that you did on Nolan and promo her early.

Your other comparisons as more of the same. Some require less effort than Nolan, some require more.

But nah. the DB are just a means to an end. They aren't worth training, right?

You're half right: the DB is a means to an end, at least when you're talking about minimizing turn counts. Whether they are worth training depends on whether you're trying to keep turn counts minimized or not. I posit that the DB is not really seriously involved in the faster clears. Ergo, if you were to frame the tier list around this idea, basically everyone in that army would drop (except for units like Volug, naturally).

The difference between upper mid tiers and mid tiers is that the upper mid tiers can actually do something without given any favoritism, or if they require favoritism, they better be damn good. Boyd is not the former (you can't honestly say never doubling and not having that spectacular durability with only average mobility is anything special), and even with favoritism, he's not the latter, because with a speedwing he's not "damn good" in part 3. He's still like one of your worst units possible for a huge chunk of part 3, which is a LONG time for someone to catch up.

See, you just drew a line in the sand. I said that people are tiered with their peers, you add to that by arbitrarily throwing out that the difference between upper and mid hinges on what the units do with favoritism. This is a silly point in my opinion. Favoritism is obviously always going to be a factor, because different units respond to it differently, and different resources have varying opportunity costs, but I wouldn't go as far to say that it's the primary difference between those tiers.

Edited July 15, 2009 by Interceptor

[Progenitus]

No further comment on the debate, but smashfanatic really needs to look up the term "ad hominem attack".

Haha, do you even know what "ad hominem" is?

http://www.nizkor.org/features/fallacies/ad-hominem.html

Ad hominem is when I try to pass an insult/flame AS my argument.

e.g. You're wrong because you're an idiot.

That's not what I've been doing, because I also backed my argument up with other things. Something like this is what I've actually been doing.

e.g. You're an idiot because of this and this.

It's amazing how people claim I'm using logical fallacies when I'm really not.

[Florete]

I'm going to have to remind you of Shuuda's earlier warning. I know debates can be pretty intense, but the flames need to stop now or I'll be handing out warns.

In other news, I was going to make an argument for Tanith vs. Skrimir, but since too much shit happens in this thread (only one person even responded to my mention of Sanaki a few pages back, so I think I'll just make the move), I ended up forgetting about it and so I just moved her down for now.

[Colonel M]

I forgot about Sanaki. Maybe above Oliver would be the highest, maybe lower pending on the situation.

[Vykan12]

There's no way Sanaki is jumping > Lethe. Part 2 use >>>> major chip damage at the price of needing serious protection.

[Florete]

I forgot about Sanaki. Maybe above Oliver would be the highest, maybe lower pending on the situation.

There's no way Sanaki is jumping > Lethe. Part 2 use >>>> major chip damage at the price of needing serious protection.

Well, I got someone to respond with that. Okay, so she's not > Lethe. However, I still believe she should be > Bottom tier.

[-Cynthia-]

Well, I got someone to respond with that. Okay, so she's not > Lethe. However, I still believe she should be > Bottom tier.

I think Lethe could rise actually, she has 2 chapters of being sort of averageish. It's probably better than what Kyza ever does, Pelleas is debatable since he at least has staves eventually.

[Interceptor]

Sanaki is not bad enough to qualified for Bottom tier. Her potshots, while hilarious, are not at all useless, and the usual bankrupt argument of taking up a deployment slots does not even apply to her. She hits pretty hard with her PRF weapon, and she's capable at fighting with siege magic. Siege weighs her down considerably, but it's not like she's doubling anyway with her still-inadequate-even-in-tier-2 SPD of 23, and her Aran-level awful SPD growth. Even the threat of being doubled doesn't matter, since the threat of being breathed on is already enough to keep her from being hit.

[charmander6000]

From a quick look I don't think Sanaki is better than Oliver/Pelleas/Kyza/Lethe though I do believe that she's closer to them than the characters in bottom tier though. On a similar note how does Meg going up with Sanaki sound? Sure she's bad, but Astrid/Lyre/Fiona are really in a tier of their own.

[Interceptor]

Not this Meg bullshit again.

What about her strikes you as being anywhere near the people in Low? Low tier is full of shit units that come pre-levelled, but most of them can do SOMETHING when they are around. Meg is a shit unit that you have to grow like a Chia Pet, except when she's done growing she's still shit.

[cheetah7071]

Taking a look at moving Lethe up (possibly because Sanaki might have an easier time against somebody else), I think she might stand a shot against Kyza. First off, she has part 2 availability. When she rejoins alongside Kyza in part 3, he has 4 more attack, but doubles pretty much just mages, against Lethe doubling warriors, halberdiers, and snipers as well. Neither is turning any heads offensively, honestly. Lethe remains a better mage killer throughout the game (10 more res), but Kyza has a few more points in def and hp. Their growths are pretty similar, except Lethe has str tied for 3rd highest growth, whereas it's fourth on Kyza, so Lethe can bexp her str up if she ever manages to get within a reasonable throwing distance of a bexp levelup (gaining cexp as a 20/20/2 beorc isn't fun in early part 3). Neither of them ever ORKOs anything (barely missing against mages).

So basically Kyza vs Lethe is a better meter vs part 2. I think Lethe has a shot at winning that.

Though having Kyza at the bottom of low doesn't really help Sanaki, which was the original point...

Maybe Sanaki can beat Pelleas?

[Florete]

I think Vykan proved Kyza > Lethe a while back, so I'm not quite ready to make that switch. For now, I'm moving Sanaki out of Bottom.

[Vykan12]

That only applies if you consider their part 3+4 performance.

[Florete]

That only applies if you consider their part 3+4 performance.

If we don't, then Kyza might as well not be on the list...or he should go down to Bottom. I think what you mean is that it depends how much weight you put to their part 3+4?

[Vykan12]

Kyza's not the only unit who's sub-par his entire existence so we can still compare him to other people in his tier besides Lethe.

[Florete]

Kyza's not the only unit who's sub-par his entire existence so we can still compare him to other people in his tier besides Lethe.

I wasn't actually suggesting moving him to Bottom...perhaps that was quite confusing though.

[cheetah7071]

On the subject of low tier, why is Lehran so high? The Ashera staff is cool and all but he isn't even the only one who can use it (if we have Micaiah on sole healing duty in endgame like people seem to be saying, and she enters with A rank staves, then she's only 9 uses of recover (exactly enough to counter the status staves in 4-E-1) and 10 uses of physic away from SS), and we still have a fortify+hammerne if we really need the multitarget healing in 4-E-5. Surely somebody like Pelleas is better than Lehran. He can pop a crown to get immediate healing for 7 chapters as opposed to Lehran's one. Sure, it needs a crown, but it's still better than what Lehran's doing in 4-2 through 4-E-4 (i.e. nothing).

[Narga_Rocks]

Uh, thank you for responding, but I wasn't looking to have my rant rated for thoroughness and accuracy, I was trying to start a discussion about tier philosophy. I'll address some of the things that you brought up (like reducing turn count) in another post. I was just hoping that someone besides smash would bite, because it's not really easy to have a reasonable discussion with him.

Sorry. I warned everyone reading what I said there would be a waste of time. I couldn't come up with anything major. I did put in the few points that I found with it. I added a bit of stuff like ratings and inane comments, but that's pretty much just because I thought my post was rather bland without it. Rather meaningless with it, too, but I felt better about typing it. Plus I was more focused on the Boyd thing and just wanted to push something out in response to your rant. It took a long time to type all the Boyd stuff.

It's just tough to say anything against it when it is logically constructed and has its own flow. I don't agree with the conclusions entirely, but I can't find where your logic breaks, or if it does.

But, I really don't think it is all that much work to get Nolan and Jill to level 18 by the end of 1-E. I don't think it pads turncounts significantly, so unless we are speedrunning there's no actual reason to 2 or 3 turn 1-6-2, for example. If we just look at max bexp, there isn't really a downside to training them. I see I was wrong about the higher turncount thing, because it seems as long as you are using a couple of units in the right way it is possible to do things faster than if we train the first tiers.

That said, assuming lvl 18 or 19 by the end of 1-E with training, we have paragon if we don't send it with Ilyana so Jill can get it in 3-6 (seal if necessary) and Nolan in 3-12 and 3-13. It's worse of a situation if you are trying to raise 3 units in addition to the units you are using anyway (Volug, Micaiah, Sothe, Laura), so that is bad, but it's not necessary to train more than 2.

Still, I agree that there could be a bit of change to tier philosophy to the db.

Basically, most will disagree with the idea, but raising one unit in part 1 means not raising another. If we assume that (in addition to Sothe, Volug, Micaiah, and Laura) 3 somewhat high-leveled tier 1 units can be raised to tier 2 by part 3, and only Nolan, Jill, Ilyana is high enough to make it there if we raise 3, then to raise a different unit up to level 19 instead of one of those 3 would mean only 2 tier 1 units will make it to tier 2, and another might reach level 14 or 15.

Well, we could waste turns above the counts to train others instead and still get 3, but that doesn't generally meet the bexp thing. Unless you want to leave Micaiah underleveled, possibly have Volug not reach S strike until part 3, and have Laura never reach tier 2 without really early sealing.

Further, now that you have raised a certain number of units to tier 2, the next question is what credit to give their part 3 performance. The problem now is are we attempting to raise any dbers to be useful in endgame. If yes, then the only units it's worth allowing to fight in part 3 are the ones with a good part 4 and endgame. Anyone else, in order to get credit for being good, they have to fight to make use of that goodness. Once they are fighting, they are gaining 4 or 5 levels that they'll never use in part 4 and a unit with a good part 4 could've used. Since it is easily doable to finish 3-6 with 3 full time tanks and 2 backup tanks for when the laguz on the east overflow once or twice, and 3-12 can be done with 3 units, and 3-13 can be done with 2 units plus ledgeholders, there really isn't a need of giving levels to anyone who isn't making it to endgame.

I focus here mostly on endgame, because 20/15 isn't exactly a stellar part 4, not when the gms are --/20/1 or something. It really takes until 4-3/4-4/4-5 with paragon usage before they can start doing the doubling they are capable of (those dbs that have good spd growth and a 34+ cap), or until 4-E itself if they don't get paragon. Plus, you can raise plenty of gms to tier 3 or near tier 3 with little long term cost, but training a couple of extra db units could mean 20/10 instead of 20/15 for most. As such, most of the support units in part 4 when the team is split in 3 is going to be gms that aren't going to endgame.

As a result, I think it is silly to rank someone highly for a good db part 3 but a bad part 4. Unless of course you treat the db as if they don't do anything in part 4. In which case, having a good part 3 doesn't hurt the dbers which have a good part 4, since we aren't letting them go to part 4. But then that results in possibly dropping a different set of dbers on the tier list because they no longer get their part 4 credit on the list.

Oh, one more gripe with your theory, though, and this is only because so many people are trying to bring Sothe down in the tier list. I think this setup where you don't care about the tier 1 dbers likely boosts Sothe (well, in the case of this tier list it merely helps to justify his position). He is now more important, since in a lot of chapters Nolan and the like aren't being trained so Sothe is basically carrying 1-2, 1-3, 1-4, Volug does most of the work in 1-5, but needs some support to not die while still attacking things going after Jill. Sothe supports with some kills to allow Laura to get close enough to heal. Sothe is still big, and in 1-6-1 he is needed because Volug Taur and Zihark alone can't likely speed through the chapter. 1-6-2 he is probably not needed since Jill can hide in river tiles while ferrying taur, but again that means in 1-8 he is even more important, since Nailah is needed to clear the left, Muarim and Tormod in the south, Vika can help against the enemies on the east, but Sothe and Zihark are likely the only units that can be deployed in the middle that can actually do stuff by this point since the other dbs are not as good. So Sothe and Zihark are carrying the way over to the boss. 1-E likely doesn't need Sothe, though, since two units can obliterate.

Again in part 3, though, you are looking at Sothe, Volug, and Zihark holding the line until BK shows up, since no one else is capable of surviving at tier 1 level 12 or 13 (Jill level 14 or 15), even with a seal. Then he still is helpful with Volug as well for clearing out the laguz. Plus Volug is mostly looking at 1-5 and 1-6-1 to build strike, so he is unlikely to have S strike anyway. Further inflating Sothe's usefulness, because the BK can solo the thing, but it would take extra turns. I suppose you can just deploy Sothe and Micaiah and give Sothe saviour and just find a spot where he can survive, but even then he is your sole unit for a few turns. In 3-12 you are looking at Volug/Zihark combo, which might not actually have A support because 1-6-2 is 3 turned, 1-8 has Volug far away from Zihark, and 3-6 might not even see them deployed simply because it is the easiest way to keep Laguz from attacking anything but BK. (Sothe needs to stand where he faces one Laguz unequipped and burn vulneraries, I guess.)

In 1-7 and 1-E, Volug and Zihark can hide in a corner pulling shove or rescue hijinx while the LEA clears 1-7 and Nailah BK clear 1-E, but I don't know if it is enough to get to A.

Anyway, if they can Sothe might not be necessary in 3-12, but if they can't then you are likely looking at Sothe and Volug teaming up on the right. Of course, in 3-13 unequipped Taur and unequipped Zihark can likely handle it just as well as unequipped Sothe and unequipped Taur/Zihark, since you just need two positions, so 3-13 doesn't help him. I don't know if this pads Sothe enough to justify his current position, but considering the chapters where a lower leveled DB team means Sothe is pulling even more weight than before, I think it almost might.

[charmander6000]

Not this Meg bullshit again.

What about her strikes you as being anywhere near the people in Low? Low tier is full of shit units that come pre-levelled, but most of them can do SOMETHING when they are around. Meg is a shit unit that you have to grow like a Chia Pet, except when she's done growing she's still shit.

She's not that bad if you train her. This is more about how badly Astrid/Lyre/Fiona are than how good Meg is. At least she's trainable while the other three requires a lot of things just so they can damage enemies enough to kill them without trying to add everything up.

Personally I'd rather leave Meg there and drop the three down another tier, but only one unit in a tier seems like a waste of space.

[Progenitus]

Meg vs Astrid (which would imply Meg > unholy trinity since astrid's at the top of the bunch) was covered a long time ago

http://serenesforest.net/forums/index.php?...trid&st=880

the arguments start there

[grandjackal]

So, since no one's gonna argue Danved vs Bastian...

Danved's surrounded by bastards with horses his entire CRK existence, only being semi-reliable 3-9, then he is hot garbage come the mercs.

Bastian on the other hand comes with a full arsenal of effective weapons (all laguz), and magic damage, since Danved's never gonna double shit. Sniper weapons, staffs, chance to actually put that effective power to good use (joining chapter, dragons)

How does Bastian not shit-stomp Danved?

[-Cynthia-]

So, since no one's gonna argue Danved vs Bastian...

Danved's surrounded by bastards with horses his entire CRK existence, only being semi-reliable 3-9, then he is hot garbage come the mercs.

Bastian on the other hand comes with a full arsenal of effective weapons (all laguz), and magic damage, since Danved's never gonna double shit. Sniper weapons, staffs, chance to actually put that effective power to good use (joining chapter, dragons)

How does Bastian not shit-stomp Danved?

Not like Danved's a great unit come 2-3 and 3-9, but he has some minor positive utility, he's about on par with Marcia/Makalov statistically, he's just unmounted, which isn't bad in 3-9 at all.

Danved's growths are meh, but he grows into a semi-usable unit. I would say having usefulness on 2-3 and 3-9> having usefulness on 4-5.