What Makes a Fire Emblem Unit Good?
ฝัง
- เผยแพร่เมื่อ 25 มิ.ย. 2024
- There's been a lot of discourse on unit evaluation lately, so I wanted to talk about a few popular frameworks for Fire Emblem unit evaluations and what they value, as well as why I don't believe there is such a thing as an objectively good unit.
Patreon: / actuallizard
Community Discord: / discord
Twitter: / actuallizard
fe3 book 1 efficiency: • Fire Emblem: Mystery o...
Music:
Roy's Jawn - Pun
Memories of Augustria - Pun
You May Call me Marth - Lolo - เกม
I would really appreciate people being more openmided when tierlist discourse happens since the community "opinion" has been wrong multiple times before lmao
Yup! FE metas have changed multiple times, and people should be open to arguments that go against the common understanding of a strategy or unit, particularly when they are well supported.
*cough* Odin *cough*
There should just be tier lists for different scenarios. A tier list for an iron man run, a tier list for an efficiency run, a tier list for a first time playthrough ETC
@@actuallizard It takes time to fully understand a game and how using resources impacts certain units. The clearest example is Seliph being misunderstood for a long long time. I really hope we don't enter another "jagens bad, they can't be good under any circumstances" era
@@dakotadoyle7573 that too
Nice video. One thing that personally gets to me a lot is watching a tier list that tries to rank units under multiple different criteria. ex: "They don't do anything in LTCs, so they aren't that good." In a tier list that was never defined to have anything to do with LTCs to begin with. Personally, I rank units based on their investment/usefulness ratio, and how their respective tools effect their game. I find that doing an LTC character tier list is kind of redundant, as there is an objective way to get the lowest amount of turns possible (kind of like a giant mathematical puzzle). In contrast to Iron Mans, that have the interesting context of a single continuous playthrough, where anyone can be useful given the status of the run.
I honestly didn't even know what an LTC was until recently.
When I finally did run into the whole LTC discussion thing... I was confused as hell. I do get that efficiency is good in a unit. And I may not entirely get the difference between LTC and efficiency... As a more casual player. I do find these conversations interesting though.
Great vid! One point I'd like to make in response to the common criticism that unit comparisons are elitist and/or contrived is that evaluating things is simply a fun and useful task in and of itself. Learning how to order your priorities and avoid logical fallacies is a very useful skill that WILL translate to your life (Do you buy a new or used car? Who do you vote for? Is it worth paying extra for delivery over pick up?). Obviously the specific criteria we apply to FE units don't really translate, but some of the pitfalls and logical fallacies can appear in a variety of evaluation contexts. Learning that not everything needs to be weighed equally and that sometimes you need to tell yourself hard truths (that FE unit/car that I think looks coolest is going to do a much poorer job helping me reach my goals than other options) are other examples of translating skills. Fire Emblem unit evaluation strikes a good sweet spot of being basically consequence-free, but complicated enough that you can actually learn some of these skills in a fun and safe environment.
aye
I'd say scarcity also plays a role.
You could have a stat and growth wise good unit but if there are 5 other similar units who outclass the first one in their role(more or less slightly) then that would lessen the value of said unit.
That's fair, but I think it's somewhat balanced by permadeath. If the optimal character died due to bad luck or mismanagement then a substandard replacement that still fills the role is probably more important than someone whose role isn't as necessary in the first place.
The reverse is also true! For example, Resistance is all but nonexistent in a game like Echoes or New Mystery, so a unit like Malliesia is super valuable since she has high resistance.
I disagree, because for example in the mangs rebalance of fe6, one of the best early game strategies was to simply go for 6 paladins by getting the extra knight crests and promoting every cav. Variance isn't really that important, see 3H. Unique utility is something I'd say matters, if only one unit has the warp spell for example. Homogeneous armies aren't necessarily weaker.
@@aetherius6221 I hard agree with this, but you could've talked about any of the other myriad games where Cavalier spam is good than the rom hack of an abuser
@greenstatic180 I used the only game ive actually played like that, as I don't have a switch, and DSFE has infinite free reclass which means you can tailor every unit to the map. On the note of Mangs, 1st, I'm almost certain that nonsense is severely overstated but since he just got married, I suppose we'll see in time. Second, he makes the best FE content on TH-cam, and Andaron Saga and things like the Dungeon DnD style stream are some of the best content I've seen period. I'm not someone who really buys that we should judge people we've never met based on the words of others we have never met. Mangs has never said anything in any of his content that I found objectionable, and that's where my only stake in the matter lies. Plus his rebalances are good lmao
Reliability is one reason I liked Timerra so much. Until Timerra joined the team, my FE:E ironman run was plagued with units getting sniped off from unlucky crits. Her aura changed everything.
My general list for a unit goes
Can they take hits.
Are their growths worth giving them enemy kills.
Can they work support
Do they have a unique power to set them apart.
Are they a dancer unit
And so on. It's can make a bad unit valuable and a ok unit unneeded where I'd choose a dancer unit over a defensive one or give my healer extra chances to level up in the future expecting them to turn into a fighter
Though I'm glad they are trying to balance units out like dancer getting a weapon to defend themselves from olivia to the engage dancer but they are still squishy.
Same with early healers. Wish they would be early magic users not physical users as that would be perfect even if they have bad def stats
@@renren5660it doesn’t matter that you lose out on some Defense if the enemy has a 0% hit rate
tl;dr:
Are they Bartre?
If yes, unit good
Otherwise, unit bad
Bartre the Beast!
Fiona and meg want to say hi, possibly mist but like staves are good
I am personally interested in how the community at large sees availability affecting units, if you want to broach the topic in a later video.
I personally believe in only judging units from their performance on maps they exist in, because obviously factoring growths into the earlier recruits does not mean you have to penalize the later ones for not being there, considering they don't actually have any opportunity cost in their non-existence, but I know that thinking tends to get bashed a lot in the community.
I think the reason efficiency is most common is because it's most similar to what you would expect from an average playthrough. A unit being good in efficiency is an extremely strong indication that most players can pick up that unit and have them perform well. Ironman or ltc or draft tier lists are cool, but ultimately only really pertain to those specific runs. Of course, they are still interesting to talk about and might also give us clues about efficiency or such, so they're absolutely still worth having.
Efficiency is the tiering system according to witch Mewtwo is worse than weedle, so it doesn't match average players at all.
@@noukan42 if you never get mewtwo at all until after you beat the game, it doesn't matter how good it is in the context of playing to beat the champion.
@@jerry3115 and i argue againist "playing to beat the champion" to be a terrible metric in any game with significant post game content, wich a number of Fire Emblems are. It is easiky my biggest pet peeve againist efficiency because most people want to do everything there is in a game, or at least everything that is reasonable(so not besting the game 40 times to unlock a character for example).
@@noukan42 if you wanna tier for postgame that is it's own thing, but often times postgame is way too easy, way too short, or just not interesting, if not all of the above, so it rarely creates interesting discussion. Especially in the case of the gen 1 pokemon games, by the time you get mewtwo there's nothing left to do anyway, so tiering it is ultimately meaningless unless you're talking about competitive play.
@@jerry3115 i'd argue half of the main games are easy and not interesting to play and they are tiered anyway so i see no reason for this discrimination
me when actual lizard upload :D
"Does this character have boobies"
Sonya > Deen real
One thing I've been thinking about is an objective standard that would better align with what beginner players would want in a tier list. In that regards, I think an "elegance" tier list could funny. In this format, instead of evaluation turn count, we'd be evaluating the number of actions required.
Nice in depth summary of evaluation logic. This is certainly an interesting topic. I've found that my personal framework is very different than most others that I am exposed to online on TH-cam/Reddit. My ultimate goal of a Fire Emblem playthrough is to make the best possible team for the Endgame (within reason.) When people talk about warp-skipping it is inconceivable to me due to all of the missed exp. And using pre-promotes that start with 12 STR at level 3 is baffling when I know other characters will have much higher potential. A character's maximum reasonable potential is what I value most, and that is what I try to maximize in my runs (again, within reason, I won't sit around spamming dance to get my dancer pumped up, for example)
I also recognize that most playthroughs don't require this (and my endgames end up being braindead easy) and I do appreciate other approaches to the game, though I admit it's hard for me to understand it at a personal level since it's so different to how I play. This video definitely gave me some insight into the other frameworks measuring character value - maybe I should try a LTC or ironman or max-efficient run some day.
I love this topic since it's such a widely debated topic.
For me, the general guidelines I use but not limited to
-Availability (I consider Gotoh to be good units due to their utility when they are available.)
-Utility (Useful mechanics outside of combat)
-Base and Growth (Overall package)
-Difficulty (reliant on the game design outside of the character.)
Some things I consider outside of my general guidelines would be how easy they are to train. I feel like Machua from Thracia is a high tier unit due to the ease of training her in the Manster arc on top of availability.
That makes a lot of sense.
I recently thought about that in regards to Awakening since it was my first game.
Great video and important to people new to the series (like me). I personally like the less objective experience of just what emotions did that character make you feel.
I have so much fondness for Mortal Savant Marianne because I just loved her character growth and stoopid crits in my playthrough. I love when others tried something because they wanted to and some underdog unit was their hero.
And likewise I enjoy getting frustrated at some highly rated unit that for (my run) just would always get doubled.
I guess in a way I like tier lists because I get to compare my experience to theirs. Putting too many criteria on it just diminishes the broad appeal I think. Not everyone plays for efficiency.
I gotta say man, you have a real knack for the titles of your videos. My TH-cam feed is pretty diverse as far as the franchises I'm invested in, and as far as just fire emblem I have at least a solid dozen channels I'm subscribed to. That being said, roughly once a week or so, I find myself clicking one of your videos because of a tantalizing title that references some meta concept that I've found myself thinking "someone should make a video about this" several times. You just seem to have a talent for asking high concept questions that matter to the overall community. I just wanted to take a moment to let you know I appreciate that you bring so much value to the community in this way. Thank you man.
Edit: omg bro, on top of everything I just said, you come thru at the end with the "it's something I want to be better about as well" self awareness and personal reflection. You are a legit paragon (lol) of what the FE community needs.
It's interesting to think about FE unit tier lists and ranking vs Pokemon tier lists and ranking. In some ways, there's similarities.
Pokemon with the same types often get compared. People say Garchomp is the better dragon/ground type over Flygon. If your team only needs 1 Pokemon of that dual typing, you'll either choose the Pokemon you like better or you'll choose the Pokemon that is better as a unit.
In FE it can often be the same but different too. There might not be enough reason to use two early game healers simultaneously unless you lack healing potions. If you have two infantry healers, you'll probably pick whichever has a better staff rank/access to better healing spells. But if one early game healer is an infantry and the other is a troubadour, you're more likely to keep using the troubadour.
Meanwhile there's very little downside to keeping two Pegasus units, or to keeping two cavaliers. With two of these same types, you can often play more efficiently than if you had only one. Some FE games turn into flier emblem or horse emblem because the cost of using a team of fliers and horses is so low that you might as well keep them and bench infantry and armor.
In Pokemon, a team of only dragons (or dual types dragons) isn't the most efficient nowadays with fairy type running rampant. Mono type teams aren't as efficient anymore, unlike with the flier and horse emblem teams in Fire Emblem.
As for objectively bad units vs objectively good units, I think it's easy to tell what is objectively bad. Perhaps it is more difficult to tell what is objectively good because so many things make units good.
With 3H as the example, objectively good qualities are:
Unit availability
Unit movement
Unit class typing
Unit weapon access/weapon level
Unit spell list
Unit combat art list
Potentially having a crest or not
Growth rates
Base stat total
All of these factors heavily influence units being objectively good or bad. But it also depends on the route too. An example:
Sometimes it's better to use a unit from the beginning (availability) but sometimes it's better to recruit out of house. But regardless of availability, it's always better to have a unit with a higher weapon rank because that means they can use more weapons (like silver weapons and brave weapons). It also means they'll have access to more combat arts unlocked, or spells unlocked.
In standard FE games, if you have the choice between using a unit with A rank swords vs a unit with D rank swords, the A rank is objectively better. The D rank unit needs to have more than enough worthwhile qualities to make up for that low D weapon rank or else it won't be useful to keep them in your limited slot team. This is why late game recruits can be useful too, despite late availability. And especially if they have high BST.
Nice video, and very well explained. One thing I would add (and I'm not sure if another comment mentions this, but @askeladd7271 touches on a related point) is that what you call "frameworks" (such as LTC, Ironman, ETC) only offer an objective metric when comparing two complete clears of a map or run, but there's still some level of subjectivity until you define how to "credit" each unit for their contributions to the metric in question (again, LTC, reliability, ETC).
A natural way to do this sorting units by "worst run without unit", that is, take a run without unit A versus one without unit B and see which performs worse with respect to LTC/reliability/ETC. But some method to account for the contribution or absence of the unit needs to be stated (and is harder to calculate) if you want to compare units strictly speaking.
Note also that while the suggestion above is natural, it fails to account for replaceability (which the comment by @askeladd7271 alludes to out), synergy between units (if unit A is only good as long as unit C is present, while unit B is always about as good as unit A, the suggestion ) and possibly other things that some people would like to account for when comparing two units.
I always thought efficiency and LTC were the same. Learned something new today.
That said, playing efficiently is what I usually aim to do. When I play the Tellius games, my metric is always reaching the Bonus Experience turn goals. Anything that helps me do that is good. Otherwise not so much. I think that is one big reason why Path of Radiance and Radiant Dawn are still my favorites. They give the player clearly defined goals and very good rewards for reaching them. Really wish they would bring back the Tellius BEXP system in a future FE game.
in practice they mostly are. You may see of efficiency as a "no-riggin" LTC that is willing to sacrifice a turn for realiability if necessary.
Unicorn overlord is interesting with unit evaluation. For most units you just take class into account because you can adjust personal growths and class is like 90% of the stats anyway. But the community gives the Jagen equivalent his own tier because even though he falls off hard, he is required on expert mode for early maps.
I wish I had this video when I was a rabid teenager on message boards
Nice vid!! I may be a tad bias but I think that an ironman run (compared to LTC or some previously calculated Turn Count) is much closer to what I would consider a standard experience because most games don't have a built in reset mechanic. On top of that permadeath has been a default feature through the whole series but can be turned off in the modern games.
It's been strange when talking about unit and someone instantly assumes that the default way to compare them is if the game is won in X number of turns and assumes resets are involved.
Move far
Win combat
Use staves
Yup that's about it
Use staves especially applies to Thracia 776.
move far can be a detriment though in the context of a new player and that's why context matter. For a new player combat stats are a lot more important than mobility because what mobility does is expand your options which means you got more things to consider which leads to information overload and more opportunities to make a wrong move. It's also a lot easier to send a mobile unit in a very dangerous position without backup.
There is a reason why new player really like generals and it's not irrational, it's because they are easy to use efficiently, you have limited options with them so it's easier to spot the best one, and they tend to be a lot more forgiving of mistakes due to their good combat stat and bulk in particular. They are vulnerable to magic but mage units are a lot easier to spot visually than things like a unit wiith a dangerous weapon which isn't always apparent if you don't check each unit carefully.
This is a similar reason why a new player will train their lord, because they will go blind in the game and it's a lot less scary if your lord is powerful as you don't risk losing the game to a reinforcement spawn you didn't expect or something like that.
Or if what you care about is the story then two units that work well together and have support can become a reason to use them even if they aren't that good because it's easy to play them in a way that build their support. If you want to build your paladin support with your lord suddenly your paladin movement loses most of it's value because you'll have to stay at your lord's pace (in earlier games at least).
Point is, context matters and there are a lot of ways to play the game where the common knowledge of what's "best" don't apply (and I'd argue it only applies to a very narrow subset of what fire emblem players actually do, like, all the run types presented in the video are actually very specific self imposed challenge, there are many more ways to play the game)
I'm guessing this video was inspired by the reddit discussion a few days back, and this was much more productive than the comment section there. I agree with everything you said, and if I would add anything to the discussion it would be weighing investments to outcomes when talking about how good a unit is.
In an LTC context, low investment units would be a premium, whereas in an Ironman getting units as reliable as possible would take more investment.
Tier lists and unit evaluation are always going to be a big chunk of disscussions in the community, but the most important criteria is the most subjective between every player (and thus not likely to be discussed directly much since the framework is so varied): having fun. This is a game, after all! Some people enjoy the challenge of LTCs, some people enjoy the risks of ironman runs, some people enjoy unlocking supports, some people just want to dump all their bexp on fiona probably. We play these games and have these disscussions because we enjoy them, and that's more valuable than any unit's stats.
It depends on the game in question and what kind of run I’m doing, that’s how I tend to judge units
New lizard ‼️‼️
The only thing that makes a character good is how attractive I find them as a character. I will go to the ends of the earth to make a mechanically unsound character good if she has a large personality, for example.
Each game the quality of a unit can vary which is fun to analyze
The gba games your early healer can become a great magic user if they gwt to level 10 but i feel i use the early units a lot more with range and canter.
Though i always find the great base stats of a hero unit useful in the kong run especially in shadow dragon and sacred stones.
Same with chapter quality and release order as lissa gets a whole 5 chapters of healing time where Maribelle comes right after making her harder to level while fates gives elise in conquest the starting maps making her very valuable.
Fates funny enough the capture system can give you a balanced roster with haitaka rally def as well as being a good unit making for extra support and amiboo units are great bases with unique skills and powers to fill out the early roster.
3 houses i just focus on the healers and warp units. Byleth solos my games so they can level up faster hust by using their magic skills
Engage dodge is king so yunaka and any theif characters i get are ny front line. But being good with magic and atk is more useful for combat with the magic weapons going through armor like butter.
Basically each game has it's own values to them
I will note that on Maddening specifically, Engage is not very friendly to overly tanky or overly dodgy builds, because a unit that's too dodgy or too tanky will simply get ignored by enemies and not actually see combat.
Magic does beat out physical though, especially since on Maddening lategame if you're not critting or using effective weapons then you just don't get to kill anything.
@@red5t653 yeah I wish engage balanced itself out a little more or gave out skills instead of giving it to the lords.
@@ivanbluecool Tbf most of the lord skills aren't super important, and I'd honestly argue that some other units have significantly more useful skills that actively affect their performance.
Panette's flat +10 crit with such a simple condition as "don't have full HP" or the various damage buffing adjacency skills (like Alear's personal) can massively swing the thresholds you need to deal with, and the various big hit buffing skills with simple conditions like "only fight enemies as they come at you individually" (Amber) or "have a bunch of dudes in the area" (Pandreo) can also make damage output that much more reliable with little fuss.
Good stats and playability. Mobility is nice too.
I remember in a play through of radiant dawn I capped all of Edward's stats save for luck of course without using any bonus exp in first job.
I was like "I guess I'm gonna end up taking Edward into the end game."
Honestly, my major qualification for character usage is how early I get them.
I always level my units to 20 and arena grind so bases and growth rates don’t effect them as much. Weaponry doesn’t matter either.
But it doesn’t matter if you got the best stats in the game if you show up 3 maps before the end.
Like in Sacred Stones, Ross is better than Amelia and Ewan because he shows up in map 2. Thus has time to grow.
But I still use my favorites over objectively better units.
Importance tier list is:
#1: Are they hot?
#2: Are they hotter than the other hot one?
#3: Are they that fucking hot?
#10: Statgrowth
A good unit is whichever one I personally have the most fun with, so the most fun is the most good. if that means using the worst units because it makes the game harder when i'm in the mood to hurt myself, then the worst units are the good units.
I feel like on þe efficiency spectrum, I lean more on þe reliability or "Ironman" side þan þe LTC one, þough I do of course like to minimize grinding wherever I can. In þat sense, intuitively I find I trend towards a metric of "who most effectively uses a given map's exp" (which sometimes can be þe Jagen! "2000 free exp" is noþing to sneeze at, after all, and of course any unit in mortal peril bears a risk of wasting time if I reset, or exp if I let þem go, neiðer of which I am fond of wasting).
One problem with defining efficiency as "the most quickly and reliably" is that you can't optimize for two variables at once without also choosing a way to trade-off between them. If you're optimizing for a single variable, it's very easy to know what's better, since you just pick the bigger (or smaller, depending on the nature of the metric) number, and that's your answer. But with two variables, unless they're perfectly correlated (in which case they might as well just be one variable anyway), you need some way to decide how much x is okay to lose to get more y. The efficiency formula does accomplish this goal, but the definition of "the most quickly and reliably" does not. In any given situation, it's unclear from the definition alone how unreliable a turn save is allowed to be in order to be worthwhile.
I know you mentioned the definition is vague, so you're probably aware of all this. But I think in fairness to the other definition that you prefer not to use, by itself your definition isn't necessarily much better. It's only when you bring in the formula that defines the relative value of each of the variables that you get something concrete. But, as intuitive as that formula may seem when you break it down, it is ultimately an arbitrary choice for how to pin down the trade-off between speed and reliability.
I was recently thinking about this question. Units which are at the extreme end of a statistic lack the balance needed to be considered a strong character. Every character has an objectively best first class they can choose. The Avatar in Awakening is less set in stone due to the slightly different statistics and growths they can start with. Cleric will always be the best first class to Seal into due to this being the most optimal option due to resistance growths which are better than other classes. On the other hand, Vaike has a weak speed statistic, so Hero is his best first class since this will result in a better character in the long term. General is a weak class due to how unbalanced its statistics are. On the other hand, Great Knight is superior to General as both a class and pair-up due the more broad distribution of statistics. In Fates the Spear Fighter and Spear Master is a prime example of what a perfectly balanced class looks like. Greater balance leading to better unit feel. Which unit feel directly relates to how consistently reliable the unit is. So basically, I believe reliable clear over the course of an entire run is the single most significant metric for a run.
For me, considering the way my friend introduced me to Fire Emblem was by telling me "yeah it's basically chess but anime, but the characters have really cute interactions with each other and really cool designs". I evaluate based on the following:
1. Do I like their design? Is it too busy? Were the colors chosen well? Do they have interesting parts to them or tell a story with the outfit or look?
2. Do I like their personality? Generally characters who are womanizers, flirts, or otherwise heavily focused on dating put me off (not always, Sylvain managed to still be one of my favorite characters).
3. Is the way they interact with others entertaining? This can be in any form, funny, thought provoking, unique. Maybe they have an existing tie to another character(s) and I want to see how that plays out.
I don't really mind having to do loads of training or drop stat boosting items on "less useful" characters, because the returns on seeing characters I like turning enemies to corpses and one shoting loads of bosses fills me with the joy of "oh, they do that because of me. I made that happen. They weren't going to do that without my effort" and it's a really nice feeling. So my judgement of a good unit or not is whether I like them as a character, stats and abilities are just numbers and words most times.
Interesting video, have a like and comment to boost you in the algorithm!
Experience black hole is a good term, I'm stealing it!
This post was approved by the Orson Sucks Gang.
Final note: can't believe I'm not allowed to self-promote, how am I supposed to clear the next map without my promotion gains?! Hope I can find a lord to promote me before I die to the brigands :(
Units are good based on whether theyre my scrunkly.
what
Great video but did you consider that Surtr from Fire Emblem Heroes is even taller
Therefore he's objectively the best
I don't play FEH, but I just looked him up and is quite lorg. I would say S tier, but if we're looking at FEH heights, I think Fomortiis is in a tier of his own!
I'm a very casual FE player in that for the most part I don't replay the games (FE 7 is a rare comfort game for me that I just go back to and play through every other year pretty much) so Evaluating units doesn't really have a place for me personally, any unit can put in work, so it more comes down to if the unit is fun to use to me. (Example: I love using Lyn, her animations are chef's kiss)
Simply put movement is more powerful than stats, availability is more powerful than movement, and stats only matter to hit certain benchmarks. If a unit is barely capable of handling generic unpromoted enemies but has 8 flying movement and is available from chapter 3 they will generally be more useful than a unit that can easily handle the final boss who's only available for 5 chapters and has 6 infantry movement. However both will be situationally more useful than the other. So basically just use whoever unless that whoever is meg or Fiona, don't use those 2, they're not worth it.
I always thought that based on the community it was that if a unit is mounted then it's good. If a unit is sword locked then bad. Flying units in particular then to do better than average overall which leads me to believe maybe Fire Emblem in the future should just have mounted units, no swordlocked units and no lords so you won't get game over screens.
dopamine production efficiency is my metric of choice
so basically, the lower the bases and the higher the growths and number of possible levelups, the better?
Based thumbnail.
ROSHE SHOWN FIRST IN VIDEO!!
To me a unit is good if I like their character design, their dialogue and they are not unusable on the map.
I think a general tier list should aggregate most of the specific tier lists.
If a unit is excellent for ltc and bad Ironman then it's a good unit, same goes for the opposite.
But if a unit ranks very well on ltc and Ironman then this unit is exceptionally good.
Glad to see it's somewhat normal to panic when the enemy even has a 1% chance to crit you. XD
IMO the more usable and more reliable the unit is, the better they are. That is why almost all healers are good IMO.
I'm a bit late to this video, so I'll probably get buried, but I have some pretty strong feelings on efficiency and how ETC=/= efficiency, nor is it a good substitute for it's vagueness.
The "broad" definition of efficiency is a better definition. The fact that it's harder to define is because it's made up of so many moving pieces, but I think it's a mistake to try and create an "objective" formula, because you'll end up missing out on part of what makes units good.
Don't get me wrong, I really enjoy watching and hearing about LTC and ETC discussion, but extrapolating ETC back as a replacement for efficiency just doesn't work for a number of reasons
Just for the sake of clarity, when I say "effciency", I'm going to be talking about the broad definition and I'll use ETC to mean ETC.
1) Difficulty.
This is a really important point that I think still goes commonly overlooked. Fire Emblem is not the hardest series ever, but it does have game modes that can get harder and take more thought. Generally, in an efficient context, we can accept a loss of 2 or 3 turns if it makes something dramatically easier to complete.
To go back to your Orson/Ephraim example, both of them could have 100% hit and never have a chance of being crit and Orson would still be a better unit because he allows the player a considerable amount of room in their strategy both to make mistakes, and for any adaptations based on a number of previous factors, whether it be earlier RNG screwage (ok it cant happen in this map but you get the idea) or a change in the strategy or meta.
Using Orson just makes the map dramatically easier and his presence on the map constantly gives the player more positive options. Orson is a very stable unit in the sense that regardless of how fe8s meta changes, or how badly you mess up your strategy, he's a consistent rock that bodies the entire map.
I think people often try to counter this with "Oh, but the easiest strategies afk in a corner and take a million turns of unequipping and heal grinding like training amelia", but this doesn't work in my opinion. In a broadly efficient setting, we still accept that low turns is better than high turns, but we don't assign a specific value to it. We can look at a situation like that, or grinding and say "ok, that is clearly too far". We might have differing opinions on where we draw the line on what "too far" is, but I would argue that this in and of itself is an interesting discussion that varies chapter by chapter.
Which leads me to my next point
2) Not all turns are created equal.
Old-Style efficiency is broadly applicable to most players. ETC is not, because players don't care whether or not your lord spends 3 turns walking up to the throne on an empty map and seizing, vs making a complicated rescue->drop chain to get him there. You can even get into some totally wacky scenarios where you could feasibly argue it's worth using the rescue staff in awakening to transport your units around maps where all enemies but the boss are dead and you just want to open the chests.
No sane player is going to agree to that because while it saves turns without impacting reliability, the turns it saves aren't meaningful. A turn where your jagen flies in and crushes the hardest map of the game and a turn where your incredibly terrible unit kills the last enemy on a rout map due to their spawn position are treated exactly the same in an ETC format. This is why simply pumping a number out of a formula doesn't work.
This doesn't mean that warp/rescue skipping is bad. I've long been a proponent of the idea that the warp staff could be renamed the "nuclear bomb" staff, because that's essentially what it is- it removes all of the enemies on a map instantly because you beat the map in a single turn. The difference here, though, is that we're saying that the turn saves is useful, because of what it provides for us in terms of making the map easier. We're not just treating turns as an intrinsic good, because that's what leads to the flawed definition we have here.
The reason this is accepted is broadly because in most games, units that save large numbers of turns tend to be units that are also really really good in an efficient context. However, that isn't always true, and we should be open to the proposition that a unit that makes the entire game way easier, with exactly the same reliability, that takes less real time to execute and has a way bigger margin for error, but loses 1 turn across the entire game because they had to walk the lord closer to the throne, is a better unit.
3) RNG.
ETC attempts to address RNG by accounting for Chance of Success, but there are a number of issues with this.
Probably the biggest one is that it is often nearly impossible to actually calculate a chance of success. I'll go on record and the person that says a lunatic+ ETC of awakening is never ever going to be calculated, because each run of the game is so incredibly variable. There are so many different approaches to a map, which can roll different amounts of dualstrikes and dualguards. Some enemies have random AI and will move differently just to give the game variance. The game is longer and allows much more levelling, so the random nature of growths can be taken more into account. There's the normal FE RNG of hit rates, crit rates, etc. And then, on top of that, every single enemy in the game can spawn with 2 of 7 possible random lunatic+ skills, along with a chance to spawn with other skills from their own class. AND on top of that, the game gives even more RNG with things like the barracks and sparkly tiles, should you choose to use them.
This is such an unbelievable amount of RNG that trying to ram it all into a formula would take you way past the heat death of the universe to actually calculate. This is even dumber when we consider that Frederick is just so clearly good and we don't need a formula to tell us that. Frederick completely dominates 75% of the game on lunatic and a good 50% of it on lunatic+. The guy takes the hardest section of the game and completely walks over it.
This is because his high stats and move make him a unit that offers an absolutely colossal "margin for error" and opens up literally thousands of new strategies by existing. Chances are, whatever RNG seed you get, Frederick has an answer to it. The same can't be said for supposedly "efficient" units like Sumia.
Ignoring the massive lead time in even creating a CoS, ETC is still flawed when it comes to RNG, because it ignores how resilient a unit is to RNG "going bad"
Frederick can deal with your other units getting stat screwed earlier the best. He can deal with you not getting any dualstrikes the best. He benefits the most from dualguards as he sees the most combat. His 7 move makes him good at dealing with multiple AI formations the best, and his excellent combat makes him good at dealing with a wide variety of different lunatic+ skill setups. He even benefits more from sparkly tiles and the barracks as he is more likely to be able to use the weapons there, and the boosts will push his good stats even further.
Like I said earlier, I do respect and enjoy ETC content and discussion, I just really don't feel like it should be "replacing" efficiency. It's a fun way to play and discuss the game, but I think it's a mistake to try and "objectively" rank FE units.
Just as an aside, wanted to drop a note that I really respect you as a creator. FE TH-cam needs more content and yours is consistently high quality and thought provoking, this video included, even if I disagree with a part of it.
That's all. Adios
Hi Lizard!
But if we're rational and explain our goals, how can we have nasty flamewars in the forums? Where's the *drama?*
Imo, a unit that doesn't need to by babysitted is already good enough for me
A snatched waist
Personally, I prefer to evaluate a unit based on how well they fulfill their role/purpose in the game on a mechanical level. I also ignore the usage of Infinite XP Grinds (Monster-Hunting or Arena-Abuse) and stat boosting items, and the ability for units to return after being defeated in battle (Assumption of an Ironman Mode not required but implied); these things can all circumvent the core design of the game and, while fun, make evaluating a unit's ability vary far too greatly.
An example of this would be "Does my starting Horse Unit perform well as a starting Horse Unit?". So, for example, Kent and Sain both perform well within their designated role as an early-game mounted unit, whereas Forde in FE8 does not perform as well in that same capacity. This does not mean that Forde is BAD at his role, but that other units in the same series fulfill that role better.
The core problem with this is that the evolution of the series, and its elevation of things such as growth rates and weapon properties, fundamentally make it difficult to evaluate a unit between games (EX. Times Two weapon effectiveness in FE7 versus Times Three weapon effectiveness in FE6). This way of evaluating also does not really work if one fails to either articulate that the role is not universal (EX. Evaluating that all thieves are worse than Seth because Seth out Seth's the thieves at being Seth) or that the object is not to evaluate the roles themselves (EX. Saying that thieves are worse than starting pre-promotes.)
To use the FE8 Orson v Ephraim as an example, I would say that Ephraim fills his role in the game as a Lord surprisingly well; he is a capable combat unit with great potential who is also critical to keep protected with support from other units in spite of his ability to protect himself. Orson, meanwhile, is a worse combat unit than Seth in that he is not available outside of the singular mission his is contained within...however, Orson still fulfills his role incredibly well, as he can be used in the same fashion as most Pre-Promoted Starter Units: He has remarkable utility and combat potential for this portion of the game, and can be used to teach the player how to play, protect them if they mess up...and serves himself as a means of punishment for overuse. I would argue that Orson is better than Seth as a Starting Pre-Promote in that he falls off in usefulness quicker (With Seth typically staying game-relevant to the end, and not really punishing the player for his repeat usage).
“The least valuable part of a tier list is the image itself” is a bar and a half.
Not being 3H Anna definitely helps
clearly what makes a unit good is how much you like them, why else would people run nino over pent
If PoR had wind edges Stefan would propably be the best unit ever, but alas.
In my opinion, I think the definition of efficiency where you're aiming for both turn count and reliability is very contradictory. Being reliable requires taking more turns, so you're not really hitting a low turn count. But if you try to hit a low turn count, it becomes less reliable. There's an equilibrium there, sure, but that equilibrium will always end up taking more turns and being less reliable than the definition implies it is.
An example that comes to mind is FE6 chapter 5. It is always going to be fastest to rush the gate and the boss, sending Shanna to get the village. But it's extremely unreliable to do so. Your other option of taking the north route is much more reliable, but is likely about 10 turns slower, at least. How is this map tackled under your preferred definition of efficient play? You can't have both reliability and low turn count.
They are often (not always) opposed, but not in a non-measurable way. You care about both, so you just apply a penalty to your turn count based on the reliability of the clear. So if you want to compare two strategies and it's too close to eyeball the efficiency difference, that's what the expected turn count formula is for. If you can take a turn to massively increase reliability, that will improve your ETC. If you can shave a turn without a major reliability loss, that will improve your ETC too. The optimized equilibrium is what we seek to achieve.
Not familiar enough with fe6 to engage with that specific example, here's a simplified example of how the formula works.
In a hypothetical 1 turn clear of a map with 33% reliability would have an ETC would be 3. A 2 turn clear of the same map with 100% reliability would have an ETC of 2. The second clear would be the more efficient as I define it.
The math gets a little more complicated on clears that take several turns with multiple points of failure, but that's the premise. One could apply it to any two strategies to compare them.
This is a classic problem of multi-objective optimization: en.wikipedia.org/wiki/Multi-objective_optimization. To give a simple example, you may care about a car's top speed as well as its price, but these goals will typically be in conflict (speedier cars are more expensive in general). The fact that the two goals are in conflict doesn't mean it doesn't make sense to say you care about both.
One way around this is to either establish a way to combine them into a single metric (for example, you may say that you're willing to pay an extra $1k per extra mph). Another is to look for the fastest car at each price point (or cheaper car at each speed benchmark). The first option is what the ETC formula does (it takes "speed" and "reliability" and turns them into a single number), the second option is akin to trying to find the Pareto front (en.wikipedia.org/wiki/Pareto_front) among all candidates.
What makes a Fire Emblem unit good, at least to me, is their characterization.
The only objective answer is which unit is the best waifu.
They have green hair and are a mage
There's one critical component to unit evaluation that I pretty much always see get ignored, and, I do think it's a valuable metric to a lot of people, even if they don't realize it, or, probably scoff at it.
FUN.
I think how fun a unit is to train, use, play around with is valuable! Especially to newer players! Telling a new player essentially "just use Seth lol he's the best" in FE8 and then they go and have Seth solo the whole game- do you think they're gonna have fun doing that? Do you think they won't get bored and drop the game or the series? And, Fire Emblem is a game, most games are meant to be enjoyed- if the absolutely only metrics that should be used are all quantitative, then you might as well build a supercomputer to play Fire Emblem in whatever method you think is most valuable and never touch the game yourself.
Vantage + Wrath enemy phase tanking with Nephenee in Fe9 is absolutely not efficient or ironman friendly, and, it's almost definitely not LTC viable, but goddammit it's FUN to do, and it doesn't really require that much comparative sweat since you have BEXP, you just have to give her your Vantage scroll. it doesn't make Neph good from a numbers standpoint, not at all, but people talk about it and other hairbrained strategies like this in more casual places because it's fun and makes for great visual content!
I fully expect to be called a bad, inefficient player for this, but, I stand by this, people ignore enjoyment too much when talking about Fire Emblem, and treat it more like studying thousands of chess plays to find the statistical best plays- and maybe there's a reason the bongcloud opening gets memed and joked about so much over there, people want to have fun playing their game, and sometimes fun, and statistical efficiency don't run down the same rivers.
Fun isn't a bad word, but so many FE places treat it like one.
To be honest I don't think doing unit evaluation in any of these frameworks precludes you from considering fun. Nephenee is a great example, she's frequently rated poorly in unit evaluation, but she's popular and people use her a lot anyway.
This leads me to believe people can do statistical analysis, and still use the units they find fun. I use Fiona in every Radiant Dawn playthrough, even though I would rate her poorly in every context I can think of.
I think conversations about fun builds and units can coexist fine with other forms of unit evaluation, as long as people are clear about what they are talking about.
TLDR: I can say a unit is bad in a specific context while still finding them fun. In fact, I often do talk about units being fun in the evaluation content I make. Just like we can evaluate chess openings and still meme about the bongcloud
That's totally fair, and, never meant to point any fingers, just bring up what I felt was missing. @@actuallizard
No worries, and sorry if I came off as standoffish, I do think it can be good to talk about what makes units fun or not fun when doing analysis
How do I evaluate units? Simple.
• How do they fight? I value units that can defend themselves when, suddenly, enemies!. Even if I like someone, if they'll just die the moment I try to use them, it's a no-go.
• How do they sound? I value good line deliveries and good lines, usually in that order. The accompanying sound design of the game they're in tends to skew this one. (Not applicable to earlier titles, so it's not a factor in those) Frederick lives rent-free in my head.
• How do they look? Good looks and good vibes are provably linked. Lady L'Arachel stays winning.
• How do they think? Interesting people with interesting perspectives - I don't need to sell you on that. I'm in no rush to press A on a dialogue box thanks to these. This one is nearly as valuable as the others, but it only loses out because of the frequent experience of using the unit compared to the relative infrequency of catching a glimpse of their inner monologue through supports and such. Ricken and Gaius get shelved quickly even though I get to sneak a peek at their characterization before they do any fighting, BECAUSE their fighting is lack-luster. Ricken's still got that dawg in him though, let's be clear about that.
I intrinsically value units who fight well, feel good to listen to, look cool or pretty or whatever, and have perspectives that catch my interest. If they don't have some of those things and I'm using them anyway, it means no one stands out and the game's easy enough for all of them to survive unless I make numerous blunders, OR it means the unit is one I've manually ascribed my attention to because I want to give them a chance to impress me. To most Fire Emblem games' credit, they usually do.
Simple, right?
Do I like them?
If so, good.
Do I hate them?
If so, objectively trash.
This is the way
Can this unit become a General? Then yes it's among the best in the series, no cap, I win
All these forms of play sound like an utter nightmare to me, like doing a formula for a run disgusting but aiming for lowest turns and stuff also sounds miserable and 0% I will never understand. I think the reality and the juice of fire emblem is that unit goodness is subjective, that is what makes Fire Emblem good. You can turtle and spend hours playing a map if you want and you can rush in, you can play just aiming to do ships, your target goal for play is what is valued and outside of the hyper focused FE community online that plays these challenge runs I think typically these aims are: "Beating the game" "Seeing the most content" "Shipping" "Narrative Value" "Narrative fun" These are general aims for like a general player who is only ever going to beat the game one or maybe two times. The average FE player was resetting the game there is a reason we got turn wheel so even a unit dying wasn't a big factor or not. Especially as most causal players experiences with FE move to emulators and save states now mean a reset is maybe a few minutes of progress rather then 20 minutes of progress.
For me I value special abilities, like if I am playing Sacred Stones I always like to get Knoll because no other FE game will I get the utility of a summoner. I typically play all girl runs or as many girls as possible runs which means automatically a bunch of units are not going to have high value for my personal play style. Like Seth will be used at the utter minimum in an SS run because he is stealing Erika XP and ultimately I want her and Vanessa from early game to eat as much XP as possible since they are key pieces to the run. I do value supports fairly highly too so not having a support with anyone deployed lowers a units value compared to having a support I can unlock and watch.
Yeah, I agree that for new players or players that just want to play once or twice, there is no reason to care about an efficiency tier list or an LTC tier list. TBH I don't think these players need a tier list or performance based analysis at all, they should just use units they like.
These frameworks are only useful for people who are either interested in what we can learn from evaluating a unit's performance in these contexts, or for people who enjoy playing that way. That's why I think their outputs frequently can't be generalized, and why I think the discussion is a lot more valuable than the tier list itself.
Also agree that the vast majority of people will not be interested in doing ETC math. Fortunately, you only really have to if you think it's fun or if you're trying to prove something really specific. Most people in efficiency discussions will probably never use the formula.
I honestly find it a really interesting form of gameplay and love the Fire Emblem community. This is honestly what makes me such a Fanatic of the game. Definitely Valid opinion with your description of Gameplay. What matters is we all love the game and I'm looking forward to more content like this by ActualLizard 🤍
I care way more about an iron man context than a LTC context. LTC's are fucking stupid and lame for how much they seem to GRIP fire emblem discourse. An iron man context makes so much more sense for comparing unit strengths and weaknesses, I don't care if this pegasus can win the map 2 turns earlier if it means my entire army gets less XP overall and they have a 35% chance of dying on the way to the objective. That is a bad unit, it actively hurts my army and leaves success to RNG. Units are good if they can be expected to reliably one-round an enemy in the maps they are available on with minimal RNG needed for levelups and little item investment needed (dracoshields etc). Units are also good if they can be relied on to survive combats without needing to soak up staff support, which is why I personally believe armor knights are highly underrated. Movement units are good if it means I can use that movement to make game-altering decisions, like stuffing reinforcements, saving villages, clearing maps before a lose condition happens, or in games like Path of Radiance where lower turn counts means you get more BEXP to spend.
their supports. Next question
There might be some merit to evaluating how good a unit would be in a blind play through. Take a look at Anna from Fire Emblem Engage. While it is generally agreed that her insane magic growth makes her a high tier unit, a first time player would see her base magic of 2 and see that she starts as an Ax Fighter and evaluate her on her physical prowess.
There are also plenty of units in 3 Houses that can learn a particular spell or combat art that makes them good, but the game gives you no reason to believe that they might be good in that class. An example is Annette. She learns some awesome Ax skills, but her base stats and growths would tell anyone that she should be a spellcaster so why would you ever level her axes?
All units are bad.
Except for SETH