Fire Emblem Tier List

[Fei Mao]

I agree with Pent moving above Harken since the level after Harken's is the dreaded Battle Before Dawn yes?

Combat The only threat in this level is the Heroes, maybe the mercs, Ursula, and Zephiel. Both Harken and Pent won't double the Heroes unless if Harken is using brave sword, both can damage units decently though Pent does more than Harken because he's targeting RES, Pent won't scratch the Bishop enemy much or some. Durability Harken might win there but that's pretty much what he has. Ursula, Pent will take less damage than Harken but Harken can damage her more than Pent would due to her high resistance.

Utility Zephiel, Pent wins here on this chapter because he can use the rescue staff on him so the players don't have to worry about him, Harken is a decent combat unit in this level considering if you reach to Zephiel first to repel his invaders, now I'm not sure how people level their Priscilla and Serra but at this level Pent might be the only one to use the Rescue staff and Psychic staff.

I'll assume that people went to the subchapter

28X

Combat Harken and Pent will do fine on this level.

Utility Pent will win here because due to the changing paths of this level and the Wyverns possibly One hit KO Priscilla and Serra (maybe) Pent can assist the offense with his staffs if he gets left behind Harken may be able to block some one tile spots from enemies but Pent is the one helping with his staffs.

29

Combat I may be wrong here Harken and Pent will not be able to double the Valkyrie and the Nomadic troopers, the annoyance in this level is the Generals and Knights spawn, Wyvern reinforcement, and maybe the Sniper Ballistas. Harken can damage the Knights and Generals tough he'll leave them with some hp left while pent can easily kill them without leaving them alive. Same deal with the Wyverns they have poor resistance so Pent will most likely one round them. Sniper Ballistas okay Harken has a better chance of reaching them due to his durability.

Utility So this is the long drawn chapters holding out till you drain or reach the forts, Pent offers staves and Magic while Harken offers only Swords and Axes.

30

Combat Harken will probably win here in durability against the intial sniper and poison archer rush against you in the beginning of the level, against the Druids with Nosferatu Pent will live longer than Harken would even though Harken will damage them more. However the map is later swarmed with more physical units than magic units so Pent will deal a great more than Harken will.

Utility If I recall this level needed heavy healing so Pent offers more than Harken would even though Harken would be a decent unit to block a tile.

32

Combat Okay so the dreaded level of moving inch by inch till you actually reach the boss Pent has better combat against the Wyvern Reinforcement and the Nomadic Trooper Reinforcement and since most the units that block your way is physical Pent will fare better against them. Okay so they both probably will not double the Nomadic Troopers or Wyvern Lord, Pent will deal a lot more damage to the Wyverns than Harken will, Harken only deals more damage to the Falco Knights and Bishops in the North west section of the map, not to mention the last forts near the boss consist of Generals and Knights so Pent will clear them easier.

Utility In this another long drawn chapter Pent will offer more help than Harken for his healing and such.

32X

Combat Okay so Harken will be dominant against Pent in this level due to the magic seal.

Utility Pent can offer some utility but only in the form of Psychics on a non magic seal tile or if you want the characters to be healed from Pent, While Harken basically just helps with killing the enemies on this map.

Final

Combat I may be wrong about this but both Harken and Pent will have a hard time against the bosses except for Jerme, the minions that accompany the boss, Harken will do well against the magic ones and the warrior ones, however Pent can easily deal with the physical units the comes in the map which is most of them anyways except for Kenneth's room and Nergal's room.

Utility While Pent and Harken can only handle the minions, Pent is there to help with staves while Harken can only do combat and just rescuing.

(Okay so this explanation might not be good but this is why I think Pent could go above Harken)

Edited August 31, 2009 by Jason W.

[CATS]

There's a difference between average stats and the boni from stat boosters: average stats always happen if you train that unit since you have no control over what stats a unit ups. Thus, its stats will likely always be at least near the average, even though they'll almost never exactly be the same.

No, average stats don't always happen. Not even close. 20/10 Bartre's average Str is 23.3, however, his combined chances to actually have 23 Str, or be off by only one point, are less than 50%. About 45% to be exact. Whereas his odds to be above or below the average by 2 or more points are obviously making up the rest of that. The chance that he actually has all of his average stats at any given level is in fact ridiculously small.

A status booster on the other hand? You have control over that, and only limited access to them.

Having control over it only matters if you take the stance that it's always going to the same one or two units. Otherwise, it's irrelevant.

Pretty much everyone can get one, and unlike with stats, you can't just assume some % value of the booster to be added to a unit's stats because the bonus will either be there or not.

No, not "pretty much everyone can get one." That's the point here. Only some units can get this Robe, while others can't.

"It will either be there or not?" Yeah, this has been said like a dozen times already. Got anything else to say except an obvious statement of the facts? Ofcourse you either get it or you don't. That's entirely beside the point. The point is about how you apply this fact. Again:

You need to realize that this is the same as assigning average stat values. Again, you can't split stat points any more than you can split a Robe. Again, by your definition, assigning Lyn 8.6 Str is every bit as "inaccurate," yet it is still done.

This is because average stats over the course of many playthroughs is what matters, not stats over the course of only a single playthrough ever; this is obvious simply by considering the fact that average stats do indeed include decimals. And when calculating average stats over the course of many playthroughs, the Robe affects units' overall averages in exactly the way I'm describing. If Lucius gets 33 average Hp over 10 playthroughs, then on one of those he gets an Angelic Robe, his average Hp across those 10 playthroughs goes up to 33.7. Just saying that he "gets the Robe or doesn't" gives us the same basic information as this, it's just that you're refusing to do anything with it for no reason, pretending that its value can't or shouldn't be quantified.

There is no in between.

Nor is there any in between with stats. Your units can only have one value in a stat at any given time.

Thus, while it may not be the most "accurate" approach in numbers to do both cases differently, it is the most logical approach in this situation and the most accurate logic. This also helps determine which units draw more use out of a stat booster than others.

Again, the "logical approach" you're taking here results in this:

And he'll either get 12 Str at L21, or 13, or 14, or etc.

Just listing the possibilities is worthless. The average sum of the possibilities is what matters. I mean, if you want to be "accurate" and "realistic" in the sense that you're describing here, then at least be consistent. Do away with average stats. Go into a debaet and do this:

L21 Erk

Possible Mag: 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20

Possible Spd: etc

List all the possibilities, talk about what happens with each possible value, and then point out which values are more likely and thus have more weight. Exactly as you are proposing to do here with the Robe. Then realize that this sucks and is illogical, and that it's exactly the same as refusing to assign an overall value to the Robe's effect on Erk's Hp. The only difference is that this is just more complicated because you have alot more possibilities, as opposed to only +7 or +0.

Armorslayers for offense, lancereavers for defense. Trade-switching.

You can't trade-switch during a round of combat, so it's not much help. One Armorslayer hit won't KO so he still has to take the counter with WTD while attacking with an Armorslayer. His problem is that he can't do everything with just one weapon. He either has similar offense and worse defense, or better defense but worse offense. Either way he's not clearly better.

This is actually bad news for basically everyone who is not a pegasus or a magic user, and it's still bad news for magic users not named Canas because they with enemy RES being actually existant here, they will likely not ORKO crap here.

Only the promoted enemies, which are also way more threatening to Guy. Lucius and Erk are perfectly capable of countering and one-rounding the crappy unpromoted magic enemies, while taking nothing from their attacks.

Armorslayers for offense, lancereavers for defense (unless there are some that carry axes, in which case he'd like to stick to that armorslayer). Trade-switching.

Generals have so much Hp and Def that he won't kill them even with two hits from the Armorslayer, or with a crit + hit from the KE. He could just stick with the Lancereaver, but then his offense sucks even more since he's now barely doing any damage at all. And yes, there's ones with axes, but they're mixed in with others that have lances. So Guy can't get a solid defensive advantage over all of them at one time.

Defense > offense in this case. Since wyverns like to come in groups, ORKOing them reliably isn't worth much if they have a good chance of killing your unit. Guy still has a decent chance to ORKO them (whenever he scores a critical) and can take them on basically infinitely as long as his lancereavers don't run out, so I'll prefer to have Guy answer their Enemy Phase attacks first and kill some of them, and then finish the rest of them off in Player Phase without danger since even the most pathetic playable character should be capable of knocking out their remaining single-digit HP in a single hit if Guy didn't crit them.

Erk and Lucius can safely fight 2-3 of them at one time, and one-round all of their targets. Lancereaver Guy could probably fight huge numbers of them at a time, except that he doesn't one-round, so in reality he can only fight a few at a time, since he can only be attacked at melee range from 4 spaces at max, and those spaces stay occupied after an enemy has attacked from them.

Who's doing better? Just depends on how many there are. Guy might be better off in Ch 26 where there's a big swarm of them. Erk and Lucius are better against smaller groups, like what you see in Ch 24 (where there's large numbers of them overall, but they only come in a few at a time).

[Mekkah]

To solve the whole Guy/Erk/Lucius thing, imo just put all of them in High, with Guy at the bottom of it. I dunno about Guy vs Eliwood...Guy has a huge lead on Eliwood as long as Eliwood's not doubling. After that, iirc Eliwood's durability win is bigger, and then for mid/late he has Javelins.

Also, iirc CATS only took Erk, Lucius, Lyn, Florina and Serra into account, but don't forget about Kent and Sain. Their concrete defenses are better by a mile, so this means they also get much more mileage out of the robe. The former five will usually only be able to take one extra hit, but for Kent and Sain it might mean they are able to take two extra, which also compounds with their ability to move further.

At this point I feel it affects others too much to just entitle Erk/Lucius with a robe without giving Guy anything, like the Ch15 Dracoshield or something.

[CATS]

Well, again, you have to consider that the Robe is in reality only going to one of the LHM units which you are actually training. Will you be training all 7 or 8 of the upper mid-high LHM units at the same time? No, probably not. That's why I just went with 5 as an arbitrary benchmark.

At this point I feel it affects others too much to just entitle Erk/Lucius with a robe without giving Guy anything, like the Ch15 Dracoshield or something.

Oh, don't even start with some sort of favoritism argument.

Back to the Guy argument......while he doesn't like Knights or Wyverns, he's not having much trouble against anything that isn't Knights or Wyverns, and against promoted enemies / bosses (other than Generals) he's doing well because of the KE's crit. And I'll give him credit for CoD because, while he is stuck with 1 range in a chapter with 99% ranged enemies, he has the ability to one-round most of the Valks, which is quite valuable as even most of the high tiers are unable to claim this.

I still don't think it's clear whether he or Lucius / Erk win after promotion, but if Guy loses, it's not by much. So I agree with him sticking in high tier for now. And I wouldn't complain if Erk / Lucius moved up.

Edited August 31, 2009 by CATS

[Colonel M]

I'd say Pent > Harken too.

Also wondering why Osw1n is over Lowen but below the Cav duo Sain and Kent.

[Reikken]

_25_ _26_ _27_ _28_ _29_ _30_ _31_ _32_ _33_ _34_ _35_ _36_ _37_ _38_ _39_ _40_ _41_ _42_
0.01 0.02 0.05 0.10 0.15 0.19 0.18 0.14 0.09 0.04 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00

20/1 Lucius's normal stat spread for hp. Average it together, and you get 30.4, according to rpgdl.com. However, I guess because of rounding error, the decimals only add up to 0.98, so if one were to average the numbers given, the result would need to be divided by 0.98.

(0.01*25 + 0.02*26 + 0.05*27 + 0.10*28 + 0.15*29 + 0.19*30 + 0.18*31 + 0.14*32 + 0.09*33 + 0.04*34 + 0.01*35) / 0.98 = 30.3163

Not exactly 30.4 because of rounding error, but close enough.

Now, to add a 20% chance of getting an angelic robe, that means that 80% of the time he doesn't get it, and the other 20% of the time he does. So to make the averages reflect that, I could . . .

Take two sets of the numbers, and for one set, I add 7 to each stat number. Then for the set without the +7, I multiply all the probability numbers by 0.80 (or in this case 0.80/0.98), so that they add up to 0.80. For the set with the +7, I multiply all the probability numbers by 0.20 (0.20/0.98). And then I add the two sets together. So I have 80% without, and 20% with.

Or, for a little bit less multiplying loads of numbers... 0.80 : 0.20 is the same ratio (4:1) as 1.00 : 0.25. So I could multiply the +7 set by 0.25 and multiply the other set by 1.0 (ie don't do anything to it), and then divide by 1.25 (1.00 + 0.25) after adding the two sets together. (Actually I have to divide by 1.25*0.98)

_stats: _25_ __26_ __27_ __28_ __29_ __30_ __31_ __32_ __33_ __34_ __35_ __36_ __37_ __38_ __39_ __40_ __41_ __42_ 
+0 Set: .01   .02   .05   .10   .15   .19   .18   .14   .09   .04   .01   .00   .00   .00   .00   .00   .00   .00 
+7 Set: .0000 .0000 .0000 .0000 .0000 .0000 .0000 .0025 .0050 .0125 .0250 .0375 .0475 .0450 .0350 .0225 .0100 .0025

SO his new stat spread is:

 . . . __25_ __26_ __27_ __28_ __29_ __30_ __31_ __32_ __33_ __34_ __35_ __36_ __37_ __38_ __39_ __40_ __41_ __42_
. . . .0100 .0200 .0500 .1000 .1500 .1900 .1800 .1425 .0950 .0525 .0350 .0375 .0475 .0450 .0350 .0225 .0100 .0025

The probabilities are now out of (1.25 * 0.98) instead of out of 1.00 because I don't want to go through and divide everything by (1.25 * 0.98). Yet. I'll do that at the end, since that's easier.

So now, average it all together, and I get...

(.0100*25 + .0200*26 + .0500*27 + .1000*28 + .1500*29 + .1900*30 + .1800*31 + .1425*32 + .0950*33 + .0525*34 + .0350*35 + .0375*36 + .0475*37 + .0450*38 + .0350*39 + .0225*40 + .0100*41 + .0025*42) / (1.25 * 0.98) = 31.7163

Or the easier way:

30.3163 + (7 * 0.20) = 31.7163

So I don't see anything wrong with just adding (7 * 0.20) to an average to take into account getting a robe 20% of the time. It's not like you have any kind of static numbers to work with. All you have are stat distributions like above. Do help me to understand what the issue is.

[Progenitus]

I don't agree with "Erk/Lucius/whatever" get the robe X amount of times. If they get the robe, no matter how often they get it, they're taking it at the expense of the other Lyn-mode units. By, say, not using Erk, and not giving that robe to Erk, I could have given it to someone else, like Lyn or Kent. Still, being actually capable of receiving that robe is better than never.

I see it being like supports. Is unit A going to be used 80% of the time, and we should then multiply the support bonuses by 80%?

[CATS]

I don't agree with "Erk/Lucius/whatever" get the robe X amount of times. If they get the robe, no matter how often they get it, they're taking it at the expense of the other Lyn-mode units. By, say, not using Erk, and not giving that robe to Erk, I could have given it to someone else, like Lyn or Kent. Still, being actually capable of receiving that robe is better than never.

Yeah, that's why I'm saying that it's only a small overall boost. The logic of "someone else could've gotten it, therefore it's worth nothing" equates to pretending that it doesn't exist, which is obviously wrong.

[Dat Nick]

Remind me when anybody EVER said the LHM Robe/Energy Ring was worth nothing.

You just can't assume they go to any particular unit, although I guess giving them to, say, Dorcas/Wallace would be pretty wasteful.

[Reikken]

So why is Guy a tier lower than Raven?

[Mekkah]

Hand Axes post promo seem like the biggest deal. Actually axes in general (WTA vs WTD on most common enemy type)

[Progenitus]

I don't agree with "Erk/Lucius/whatever" get the robe X amount of times. If they get the robe, no matter how often they get it, they're taking it at the expense of the other Lyn-mode units. By, say, not using Erk, and not giving that robe to Erk, I could have given it to someone else, like Lyn or Kent. Still, being actually capable of receiving that robe is better than never.

Yeah, that's why I'm saying that it's only a small overall boost. The logic of "someone else could've gotten it, therefore it's worth nothing" equates to pretending that it doesn't exist, which is obviously wrong.

I'm well aware that the robe exists. My statement even said this...

Still, being actually capable of receiving that robe is better than never.

I guess multiplying the +7 HP by some number helps quantify the value of the robe into a concrete number, but I still disagree with it =/

[CATS]

So why is Guy a tier lower than Raven?

Because the idea of a promoted Guy running around with constant A Matthew B Priscilla bonuses got debunked a long time ago.

I'm well aware that the robe exists. My statement even said this...

Still, being actually capable of receiving that robe is better than never.

Indeed. I was just addressing the other part of your post, which contained that "doesn't matter cause someone else could always use it" logic which I've seen so often in the past.

I guess multiplying the +7 HP by some number helps quantify the value of the robe into a concrete number, but I still disagree with it =/

Well phooey on you, then.

Edited September 1, 2009 by CATS

[CATS]

As long as we're talking about Guy, I'd like to examine Guy vs Erk overall--for the whole game. Yes, I went there. I think Erk has a case.

Ch 14, offense. Let's say Guy is L5, part of 13 + 13x. Erk, L7 perhaps, 6 levels over 7 LHM chaps. Cavaliers have 6-8 Spd, and Erk has 10 Spd, so he's only doubling a few of them. However, Guy loses 5 AS from the Steel Sword, making him actually slower than Erk, and the Iron Sword does jack squat damage (we're talking like, double 6's if he's lucky, to their 24-25 Hp; his Iron Sword damage can easily go as low as a mere double 4's). Guy doubling with the Iron Sword is like 10 damage, while a single Steel Sword hit is a little less. Erk meanwhile does 14-15 damage with a Thunder hit, and if he manages to double, has the potential to one-round.

Iron Sword Guy doubling the Mercs is like 16-17 damage. Erk hitting once with Thunder is 14-15. Meh.

The one Myrmidon is so fast that Guy can't double it (12 Spd), though I forget what weapon it has, and if it's holding anything other than Iron, he should be able to take it.

Guy wins against the Pirates, which have like 8 Spd and (iirc) Iron Axes, although he is just barely missing the one-round on average (21-22 damage to their ~24 Hp).

However, Erk wins against the Knights, and by a larger margin. Guy does like single digit total damage even when using Steel, while Erk very comfortably one-rounds.

Then their offense against the other enemies looks to be pretty similar.

Guy has the Killing Edge to go super saiyan, but it has very limited uses and isn't Prf to him. Matthew has base D in Swords, so he easily can have a C by now, and Eliwood starts with a C. Yeah, Eliwood has the Rapier, but the KE still helps him against enemies that aren't weak to Rapier; for example, it can be the difference between one-rounding and not one-rounding the Soldiers, not to mention the +30% chance to one-shot things in general. Matthew basically wants to use the KE as much as he can, and we'll say Eliwood would like it maybe half as often as Guy/Matthew. So 40% Matt, 40% Guy, 20% Eliwood; splitting 20 uses that way results in just 8 uses of it for Guy. And there's more people who can make use of it coming up (Low1n should hit C Swords pretty soon here, if he hasn't already with his base D); this is also not considering that you almost definitely already used some of it in Ch 13 and 13x.

So Guy maybe has enough KE super saiyan juice to win this one chapter. If he has to use it all up winning here, then he's definitely going to be losing in the long run, since the general rule is that the further into the game we get, the better Erk is doing since he becomes able to double more and more often. Until we finally get to Dragon's Gate and are able to buy Killing Edges, atleast.

On defense, Guy has about 4 Hp and 3 Def over Erk. Pretty nice, but remember, Guy always takes counters (except vs archers, which don't attack Erk anyways, but might attack Guy), while Erk never does, and this chapter is not nice to Guy WT-wise. An Iron Lance Cavalier with 7 Str and 5 Skl does 7.83 average damage to Guy (assuming that he's using Iron and has maxed out Avo); after taking that counter, Guy and Erk both die in 3 hits, and Guy's defense lead has disappeared. Granted that Guy still wins overall because of times where he doesn't take a counter (i.e. finishing something off after someone else has weakened it), and because this chapter does have a few axe enemies that won't hit him, but it's a small win.

And again, the general rule is that Erk's case gets stronger the further into the game we go, since Guy's biggest lead (Spd) becomes more and more superfluous. Guy develops C Matthew somewhere along the way, but this is definitely going to be much less significant than Erk's gains in Spd. In terms of defensive growth, Guy has 10% more Hp, while Erk actually has 5% higher Def, although admittedly Guy is 2 levels lower here and his Avo growth is considerably higher. However, at the same time, Guy loses Avo when using anything not Iron, and Erk has that Res lead that matters in some situations.

Then promotion and blah blah blah, but we already spent a while talking about post-promotion performance, and I certainly didn't get the impression that Guy is doing considerably better.

Overall, I say that a tier gap between Erk and Guy is not justified. If I had choose who goes where, I would initially vote Erk to high tier, as he's usually been there along with Guy in the past.

[Dat Nick]

IIRC we've been trying to move Erk up for a while anyway, so this is just another ticket to.

[Raven]

I've made some changes to the list... The largest change is the dropping of Guy to below Erk. I don't believe Erk is as good as those above him on the list, and a good argument has been given to prove Erk is efficiently better than Guy, so... Guy drops. For now, at least.

[_____]

Are those enemy stats finished yet? I'd really like to argue Lowen > Kent.

[CATS]

Stats for 11-20 are already posted. I can provide for up through 24.

[Fruitloop Multipuck]

Hector vs. Harken:

Hector is second best unit early on and has basically top tier performance until other people start promoting, at which point his position he declines quite a bit. Harken probably joins too late to do all that much in Pale Flower of Darkness, but then he wins for three chapters until Hector promotes. The Berserker only has 1 empty slot, Harken's likely not coming to that. 3-2, Harken's favor. Then we've got 4 "real" chapters left in the game if we go to 32x, though spots are a bit limited there so Harken might not get to go. But even if Harken wins all the remaining chapters, which is arguble, Hector by that point is a more than competent combat unit himself and he had all that early game epic win and mid game lesser utility, so I'd see him winning overall fairly easily. Hector to Bottom of High, Harken to top of Upper-Middle?

[Raven]

Hector vs. Harken:

Hector is second best unit early on and has basically top tier performance until other people start promoting, at which point his position he declines quite a bit. Harken probably joins too late to do all that much in Pale Flower of Darkness, but then he wins for three chapters until Hector promotes. The Berserker only has 1 empty slot, Harken's likely not coming to that. 3-2, Harken's favor. Then we've got 4 "real" chapters left in the game if we go to 32x, though spots are a bit limited there so Harken might not get to go. But even if Harken wins all the remaining chapters, which is arguble, Hector by that point is a more than competent combat unit himself and he had all that early game epic win and mid game lesser utility, so I'd see him winning overall fairly easily. Hector to Bottom of High, Harken to top of Upper-Middle?

Seems valid enough to me. What do others think?

[Fei Mao]

Enemy Stats 11-20 (Just for repost I'll try to get the rest from 28 to endgame SO MANY PICTURES TO TAKE :( )

[Dat Nick]

What? How did Lucius get above Erk? Seriously, what?

First off: Erk's level>Lucius's level. Erk comes earlier on both LM AND HM. Then there's the more obvious Anima>Light, which lessens Lucius's offense lead on top of Erk's level lead already closing the gap. And that's not the only gap, the durability gap also increases, and Lucius's shit for luck hurts that. And Defense>Offense on HHM for obvious reasons I've stated several times already. Especially on mages as their offense is massive anyway.

[CATS]

Yeah I don't know how that happened. Erk > Lucius no doubt.

Also agreed with Hector > Harken, aswell as Hector going into High. I think Eliwood vs Hector is also clearly in Hector's favor. Hector wins until Eliwood promotes, at worst, and Eliwood promotes at Ch 26. This gives us 8 chaps max of Eliwood winning. 10 if you assume Eliwood goes to 30 and 32x. And this is also assuming that he's still winning after Hector promotes. Whereas 11 up through 25 is......unless I'm miscounting, 20 chaps if you go to 19xx, 19 otherwise. Twice as many. Sort of a blowout, really, I think.

What about Pent? He has like 1.5 chapters over Harken, and he gets points for high staff level, but I believe Harken is better at combat after he shows up. IMO wherever Harken goes, Pent should join him, maybe staying in the space right above him.

Edited September 2, 2009 by CATS

[Colonel M]

To ask again: why is Lowen < Oswin but Oswin < Kent and Sain. Seems a bit odd IMO...

[CATS]

No idea. I could agree with Low1n > Osw1n; Osw1n wins for a good portion of the earlygame just due to invincibility, but Low1n also gets really tanky really fast, and sucky move is sucky, not to mention Low1n is faster.