Redwall

I found Lon'qu pretty good early on as a lead unit: great in P1, solid in C6-7 and P2-3 thanks to his Killing Edge. I didn't use him much as a Pair Up since Panne gives comparable Spd while also providing Str, though I think he does that well too. At the end of the day, this list is obviously very WIP, so suggestions are welcomed. Chiki: have you completed C2 w/o Fred? If so I'll probably change the name of his tier.

I reiterate: how much Renown should we assume? When Red Fox brought it up, I suggested 330 (about one playthrough) as a base, with 10 Renown being awarded for each completed level thereafter. Should we change the COD threshold to 5%, and should we assume that only Chrom and the Avatar are not allowed to die (thus permitting blood sacrifices)? This has the benefit of making more characters competitive (or so I would think): Nowi might go up, and I'd have a better reason for keeping Panne in a higher tier than "good Pair Up bonuses." You are free to provide evidence from your playthroughs to back this up. From my experience, I don't think Nosferatanks have much usefulness in cutting expected turns unless it's someone like the Avatar or Morgan, neither of whom need any babying. Sumia is S tier because without her Spd-botting and more importantly, her flying (Spd-bots are replaceable), our turncounts will go up by a lot. I stuck with the Sumia and Fred combo until C13 in my playthrough, and they also made an appearance in P17. Yes, I'm looking exclusively at Pair-botting for Sumia. Cordelia is useful for Rally Speed if you don't have Spectrum, and can use Rescue, but I don't think these alone put her above Sumia. Cordelia's offense can be a bit lacking depending on how quickly we go. Not necessarily the case if the person whom they ferry doesn't provide any unique contributions--if I have a million staff users (redundant) but only a single ferry for these staff users, the current set of criteria favors the ferry over each individual staff user.

How much Renown should we assume? When Red Fox brought it up, I suggested 330 (about one playthrough) as a base, with 10 Renown being awarded for each completed level thereafter. Should we change the COD threshold to 5%, and should we assume that only Chrom and the Avatar are not allowed to die (thus permitting blood sacrifices)? This has the benefit of making more characters competitive (or so I would think): Nowi might go up, and I'd have a better reason for keeping Panne in a higher tier than "good Pair Up bonuses."

OK, I'll just keep things as is. 100%-guaranteed character survival seems like the only thing that's theoretically satisfying, but I'm not sure if this is possible in every level. AFAIK, without a bit of turtling, no one will ever have a Grima clear with a 0% chance of death, not even after applying Rally Everything and weakening him with Hex and Anathema. Initially I chose a 60% survival criterion just because, well, I have more fun playing that way. However, increasing this to 100% may have empirical benefits like having more characters rise up from C tier, since under a 100% survival criterion, the player will be playing at a slow enough pace to make other characters usable. What do people think?

It sounds like Interceptor wasn't going to the lengths Chiki was to shave turns. The only thing we can conclusively say is that they were both playing to have fun. For this reason, I consider both playthroughs to be equally good, since it sounds like they both had fun.

OK, how about this: I separately include in Required tier the Premonition incarnations of the Avatar and Chrom. The "present" Avatar and Chrom can then have their post-Premonition contributions tiered as is. If Red Fox of Fire can tier the 3-13 Archer in RD, I think this should be OK too. Although having never played RD, I cannot tell if the Archer was a serious inclusion or a joke one. My idea is that they're being used primarily as Pair Ups by the Avatar (who benefits a lot from +Mag and +Def) and by the staff users (who can also benefit from increased Rescue range). I've never used Tharja and Henry as lead combat units, so feel free to make an argument in their favor. Under the criteria I've adopted, I don't see any need to penalize Sumia for being Frederick's ferry/Spd-bot. Combat in its own right does not necessarily have anything to do with tier lists.

I realize it's customary to ignore required things like the Lord seizing and such, but with the current criteria, I think making a separate tier, following SDS's example in one of the older tier lists, may be the cleanest thing to try. Chrom's worth in a given context is determined by removing him from that context and seeing how conditional expectation value of turncounts change; in the case of the Premonition, removing Chrom (e.g. sandbagging him) does nothing if the Avatar is used, but sandbagging both Chrom and the Avatar makes the Premonition "impossible." I would disagree with the claim that the Premonition has no impact on the actual game since it is necessary to get past it to even see the Prologue. Yes. I only consider her worth as a +Spd partner and ferry for Frederick. Intuitively we both prefer the latter, but the criteria as written favor the former. I don't know what to do about that; I'm open to suggestions.

Sounds good; hopefully this thread does something to get the ball rolling for your own tier list. I disagree. We agree with Flavia and Basilio being ranked low because they appear in C23; I don't see any need to place Tiki that much higher for having another four chapters (C19-C22) on them. I can't see Tiki above mid- or high-B tier at best. Just as Frederick and the Avatar are supremely amazing in the opening act of the game, so too is Morgan supremely amazing upon joining; units like Chrom, meanwhile, are simply "pretty good" throughout the game. Perhaps third-best unit is a bit high for Morgan, but I cannot presently see any other unit filling that spot. I don't see the problem. There's no logically necessary reason to punish Olivia for being dependent on a combat unit to do the dirty work. Often a Lunatic team will benefit more from adding Olivia than from adding another combat unit since the EXP distribution is going to be concentrated onto fewer units with Olivia present. Morgan has 1-2 range over her, and can simultaneously have Galeforce (inheritance) and Veteran. Yeah, the Parallel Falchion rocks, but even with Veteran, I found Lucina really difficult to train since she was sword-locked. Promoting her to Great Lord didn't change anything since she would have needed 15 rounds of combat with a forged Log just to get Javelins. I'll move her into A tier now that you mention her perks, but I don't think Morgan vs. Lucina is a contest at all. Flight is pretty much it; she can also take a Master Seal to insta-promote for an extra Mov point, which is helpful for nabbing treasure and ferrying staffbots. She can also aid in cleaning up C12, though this is comparatively minor. I have in mind a pretty Rescue-heavy playstyle, making flying over obstacles pretty useful. I think all three of Nowi, Sully, and Stahl have situational uses, but Nowi is the only one who lays claim to being invincible after reaching critical mass. Stahl is way too slow, and gets doubled in his joining chapter. Sully takes a surprisingly long time to get going since she's Lv 2 and gets OHKOed in her join chapter by Barbarians unless she gets Stahl's Bronze Sword. I don't think Chrom is that bad since C5 and C7 nab him easy EXP and since Dual Attack+ nets him a lot of secondary EXP. I think I'll move all of Virion, Flavia, Basilio up slightly now that you mention it. In the playthrough I list in my sig, Virion served as chip damage in C1, C2, C5, and C6, though his post-C1 contributions could have been replicated by a Mage. In contrast, Flavia and Basilio were solid in C23 and pretty much required for C24; Basilio was also helpful in Final for me due to Rally Str. I agree. Not sure what I was thinking. I agree. How does one playthrough's worth of Renown (~330) to start sound, with 10 Renown being awarded for each completed level thereafter? Panne's a pretty good Pairbot even without the Second Seal: she gives +4 or so to each of Str and Spd and supports with Cordelia, who is pretty solid thanks to Rally Speed and being a flying Rescue user. With the Second Seal, she can just get Rescued by a staffbot if anti-air appears. I feel like we're talking past each other since we have different views on how quickly to play; for the time being, I'll stick with my LTC-oriented criteria, but you're obviously free to do things your way when you take over. Yeah, I'll move Lucina up a bit, though I think Morgan is still substantially better due to not needing Levin Swords to counter 2-range units. You should check out the playlog in my sig. The only levels in which I made huge gambles were C20 (not necessary since my team had enough bulk to power through at a slower pace) and Final (necessary because my Avatar was Mag-screwed and because I didn't spend much time training Chrom). Had I been playing more reliably, I would probably have needed no more than eight or so extra turns of Chrom-training and other miscellaneous things to complete the campaign. If this were Lunatic+, I'd agree, but Lunatic enemies are weak enough to permit reasonably fast play. Hmm, should there be a "Unique Utility" tier for guys like Fred, Avatar, and Chrom who have contributions that are required? Fred has the earlygame, in which he will be contributing either as a Pairbot for +Def Avatar or as a lead combatant. Avatar and Chrom have the Premonition.

Thanks, Vaike has been added. Some food for discussion: -is Panne S tier? Without the Second Seal (so far Renown isn't being assumed), she's a good Pair Up for Cordelia, and with the Second Seal, she can obviously be pretty dangerous. -C tier is kind of a mess. Maybe a D tier should be made, but I don't think there are very large differences among the C-tier characters: all have situational uses if we assume the best characters are fielded, while they can also become competent if we assume not all the best characters are fielded. What do you think?

NOTE: No need to read most of this topic (just vote, and perhaps chime in with suggestions on the direction of a long-term list), but perhaps you should start at page 7 to get a sense of how a long-term Lunatic tier list might work. This is an experimental tier list for Lunatic, with criteria inspired by those of other tier lists, in which the player is assumed to move as quickly as possible subject to reliability constraints, with varying combinations of characters which may include anything from Wyvern Sully to Villager Donnel. Although I'm not really interested in maintaining a tier list in the long term (too much work), I'd like to at least get a little bit of discussion going. If, for some strange reason, people continue to comment on this thread after a week or so, I'll probably ask a mod to close the thread, and anyone who's interested in managing a tier list of his or her own can pick up wherever we leave off. [spoiler=Tiering criteria]Chance of death: The chance that Chrom or the Avatar die and force a Game Over. We subjectively adopt a threshold of 30%; if a strategy has greater than a 30% chance of a Game Over, then we neglect that strategy. Note that sacrificing other characters is permissible. Conditional expectation value of turns saved, given no Game Overs: A character saves n turns in a given context if and only if the lowest possible turncount is n turns higher in his or her absence. The conditional expectation value of the turncount, given no Game Overs, is the average value of the turncount assuming that a character death does not occur. Consider a strategy for the Prologue that has a 30% chance of clearing in four turns, a 34% chance of clearing in five turns, and a 36% chance of getting Chrom killed. The conditional expectation value of the turncount, given no Game Overs, is simply 4*30/(30+34) + 5*34/(30+34) = 4.53 turns. Now assume the optimal strategy in Lissa's absence has a 25% chance of clearing in four turns, 39% of clearing in five turns, and a 36% chance of getting Chrom killed. The conditional expectation value of the turncount, given no Game Overs, is 4.61 turns. Hence, we say that Lissa can be expected to save 4.61 - 4.53 = 0.08 turns if a death does not occur. There does not exist a self-consistent method of accounting for the chance of death via a turncount penalty, necessitating a separate treatment of the chance of death. [spoiler=Other rules] 1. No Spotpass. 2. A base Renown of 330, which roughly corresponds to one playthrough, is assumed at the start of the game, with 10 Renown being awarded for each completed level thereafter. 3. No DLC. 4. Save for Paralogue characters, all characters are assumed to be recruited. 5. All Paralogues are optional and may be visited at any time. I can change these rules if people wish, as they are obviously arbitrary. Renown Awards The Glass Sword, Second Seal, Orsin's Hatchet, Seed of Trust, Levin Sword, and Energy Drop are free as soon as Renown becomes accessible. Beast Killer (after completion of 7 post-Prologue maps) Spirit Dust (after completion of 14 post-Prologue maps) Celica's Gale (after completion of 22 post-Prologue maps) Secret Book (after completion of 30 post-Prologue maps) Longbow (after completion of 39 post-Prologue maps) Top Frederick Avatar (M/F) S Tier Morgan (M/F) Olivia Libra Anna Sumia A Tier Chrom Cordelia Lissa Maribelle Lucina (Avatar) B Tier Lon'qu Panne Tharja Henry Cherche Tiki Gaius C Tier Nowi Sully Stahl Kellam Miriel Ricken Say'ri Gregor Basilio Flavia Vaike Virion D Tier Donnel

Saw the first episode on Crunchyroll. Pretty Good, though just a tad bit overdone.

Basically everyone you listed is usable, but many can be a drag depending how quickly you want to play. Stahl and Vaike in particular require a bit of babying to keep up. It is also preferable to train a smaller team of superpairs over a full team. You might want to check out the efficiency playlog in my sig.

The strikethrough suggests that it was only previously that Pair Up bonuses were not to be taken into consideration, and that there was subsequently a rule change

OK, here is another example I feel is handled more cleanly by my way of doing things. Suppose team A has 60%, 20%, and 20% odds at clearing a level in one, two, and three turns, respectively; suppose team B has 60%, 30%, and 10% odds at clearing that same level in one, two, and three turns respectively. Under your tier list's criteria, team A's clear is just as reliable as team B's clear, precisely because you consider perfect play to involve resetting for the single lowest clear within your reliability constraints. Most people would consider their definition of perfect play to be pretty much the same as yours, but without the resets; they would play optimally, and should the RNG push back the clear by a turn or so, they would continue to play optimally under that constraint. Looking at things this way, they would consider team B's clear to be better because while the minimum turncount occurs with equal probability in both cases, the expected turncount is lower for team B. When people play reliable LTC, I don't think they care too much about RNG screwage as long as they make the optimal gameplay decisions subject to any RNG screwage or blessings.

My post was indeed written under the assumption of perfect play. What I was trying to say was that even with perfect play, you can, for example, have Morgan fail to kill Grima on turn 1 due to the RNG but succeed in killing him on turn 2; you can also have him successfully kill Grima on turn 1 due to the RNG. In both cases, perfect play is assumed, but going by your current definition of reliability, at most one of these two possibilities (a one-turn clear or a two-turn clear) is going to count, when I think both (and possibly higher-order possibilities resulting from turn 2 RNG-screwage, and not from imperfect play) should count.

Well, OK, but there can in general exist stages that have a 60% chance or so of being cleared in (say) two or fewer turns, with both one-turn clears and two-turn clears being very realistic possibilities. My question is, in those cases, whether you consider such clears to be two-turn clears or one-turn clears for tiering purposes if neither the two-turn clear or the one-turn clear occurs in >= 60% of your attempts. I think it is preferable to instead compute the expected turncount in order to account for both possibilities, and possibly accounting for things like three-turn clears.

^You're addressing me, right? My example was only limited to that context; I agree that we should consider many contexts (combinations of characters) in tiering. The challenge is figuring out how we weight the different possibilities. edit: can someone link me to a source describing what N/P calculations are? My only exposure to probability/statistics in the last few years has been in the context of quantum/stat mech, so I'm not familiar with the parlance of statisticians/economists.

Suppose we adopt the criteria in your tier list: if we fail more than 40% of the time, then we don't consider the strategy. In my example, a one-turn clear occurs 40% of the time, and a two-turn clear occurs 24% of the time (so there is a 64% chance of clearing before turn 3). Higher-order possibilities are not accounted for with your way of doing things, so we will restrict discussion to these two possibilities. How would your tier list criteria treat this: as a one-turn clear, or as a two-turn clear? While the assumption of a 0% COD is unrealistic and imposed solely for the sake of simplicity, I don't see any problem with separately treating the chance of death should it be nonzero. I don't see any self-consistent way of "converting" COD into a turncount penalty.

OK, let's say Morgan has a 40% chance of procuring two crits (thereby defeating Grima) on any given turn. Suppose for simplicity that Morgan cannot die, and that Grima gets healed back to full health during the enemy phase. Also assume that Morgan can't attack on the enemy phase. The probability of a one-turn is 0.4 The probability of a two-turn is 0.6*0.4 The probability of a three-turn is 0.6*0.6*0.4 ...and so on for higher-order possibilities. The expected turncount would wind up being a series <TC> = 0.4*1 + 0.4*0.6*2 + 0.4*0.6*0.6*3 + ... I think this particular series simply converges to 1/0.4 = 2.5. So I would say that the clear has a 0% chance of death, and can be expected to clear in 2.5 turns on average.

My proposal doesn't neglect those possibilities. Looking at your example, we can compute things like a) Morgan getting the requisite number of +Str level-ups, b) the possibility of Morgan procuring the two required crits, and c) the possibility that, if Morgan doesn't get two crits, he will get the two crits on the next turn.

I prefer a different definition of reliability: one that is based on chance of death. If we have 0% chance of death for a strategy that is 40% likely to clear in four turns and 60% likely to clear in five, then we simply compute the expected turncount and separately note the chance of death, which in this example is 4.6 turns with 0% chance of death.

I don't disagree. I don't think we're on the same page. We can, for example, play as quickly as we can, subject to reliability constraints, with Donnel, Virion, Kellam, and so on. The argument is that this is less arbitrary than playing "pretty quickly," subject to the same reliability constraints, with the same set of characters. I've never played the Tellius games so I can't comment on that part of your post. But your claim that versatility matters is one that is already accounted for in tier lists. If Michael Jackson indeed required a very specific team to get a lower turncount than team B, then MJ would be penalized. I don't know if there are any cases quite so extreme in the games, though.

bearclaw: You don't understand what we mean by "going as quickly as possible." I can, for example, go as fast as possible (subject to reliability constraints) using trash like Virion, Donnel, Kellam, and so on. I can also go as fast as possible using top-tier units like Sumia, Cordelia, and Frederick. It is less arbitrary to say "subject to reliability constraints, faster is better" than it is to say "subject to reliability constraints, faster is better, except when I don't like it."

Scarlet: suppose the rule in a tier list is to clear in 150 turns. If with a given team, I achieve 120 turns with 0% chance of death, and 120 turns is the fastest clear possible with that team under the 0%-death constraint, it is arbitrary to somehow equate that clear to a 150-turn clear with the same team that has a 0%-death chance. The 120-turn clear remains better.

Maribelle is definitely an above-average unit because of Rescue. Pair Ups and Tonics are useful for fixing her low Rescue range. With a Master Seal, she'll gain more Mag and Mov, as well as tomes, though this may not be practical depending on how quickly you play. 6/10 unbiased, 7/10 with bias since she has hilarious support conversations.