XeKr

I'll just note that people try to rank units based on efficiency. The very problem that's mentioned in this topic arises. It's subjective. Would you try to rank stuff by warmth? Which is why I personally like criteria such as expected turn count. Calling it "efficiency" for simplicity can represent the general idea, but when tested/questioned, we have some more rigorous criteria to refer to.

I require Matlab or R plots for verification @Red Fox of Fire: I'm getting the impression that we don't really disagree that much. Just that for me, stricter standards are more theorectically satisfying (there's a certain elegance to optimized strats) than lenient ones, when available. For example, if you could complete earlygame chapters 3-5 turns faster, why not 4-6 turns faster? Alternatively, why 3-5 turns faster and not 2-4 turns faster? etcetc I'll likely address your points in more detail later.

RE: philosophy: I'll let others more seriously debate this, but I'll just briefly say (to address a few points and to hopefully elucidate the position of some of us in this topic.): I don't think anyone wants to reiterate years of "efficiency" and turn count arguments. As far as I know, most people tend to agree that faster is better, more reliable is better. Expected turn counts combines both into a continuous metric, therefore it's one that resolves differences between characters using criteria we value. It's an arguably cleaner way to resolve turncount vs. reliability arguments which tend to crop up because they have somewhat a direct relationship. As turncount decreases, as does reliability, and vice versa. Statistics are useful (rather, that's their use), to summarize a wide range of potential occurrences (subject to constraints like fast and reliable play). There is already precedent for using statistics for average stats. The issue some have with "brisk" play has been stated ad nasauem already. Unproductive arguments arise like “for this chapter, we go this fast, but not that fast” or “the strat must be this reliable, not that reliable” or “noooo u, that’s not efficient/brisk because I don't think it is. This is" There is a tendency to use turncount arguments when convenient for their position but not otherwise. Redwall has already shown that Lunatic (including a few paralogues) can be completed in sub 120 turns with fairly reliable strategies (I think <40 chance of death in most cases, except C20). It's apparent that the character placement is heavily inspired by such a pace. He (and others) would probably consider that brisk. This is not an indictment of your playstyle/playlog (I apologize if it sounds aggressive/elitist), but you were at Chapter 10 then. If we consider multiple playstyles as equally important and both as "brisk", character placement can vary significantly (see: Sully). I realize some people like debating this kind of vaguer stuff, but generally I think more mathematically inclined people do not. There are numerous other objections to be made with these present criteria as well. Including but not limited to, the criteria currently favor 70% success or death strats over 70% strats that complete with 100% success in 1 more turn. In addition, the ranking can be roughly summarized as "optimal units" > "units that act as utility/support for optimal units" > "non optimal units". This would be more troublesome for intuition if the "non optimal units" were still good fighters or w/e, but in this case they are legitimately pretty bad (HM Donnel is an apt comparison). We have this issues which may be unintuitive but they are imo less arbitrary and still preferable than trying to subjectively place each character by alternate criteria. That's my 2 cents. Re: Lon'qu, He barely avoids the ohko, which is important since other Non-Avatar/Frederick characters are probably 2-3hko'd. Lon'qu has significantly better avoid (overall durability) due to how weapon triangle calculations work (the earlygame is primarily an axefest) and his skills/stats. For example, I believe Lon'qu faces around 20% hit or 8% true hit when he gets a C support. He has a ~3.4% chance of dying if faced with 4 enemies and is 2hko'd. That's far better than most other characters can claim (I believe so at least, I'm willing to accept other evidence regarding this). Lon'qu also has the base Spd/growth to double and C Swords, which not many characters can actually claim. The result is his offense is among the best of the team. EDIT: This was a lot longer than I intended. >_>

C2 is only the beginning too, as it can still be somewhat Frederick soloed. Enemy stats ramp up so suddenly when they promote that I imagine it would be quite tricky to deal with unexpectedly. Plus stuff like overusing Frederick, non-Veteran units, has pretty big consequences down the line. There's a lot of exp optimization to do in the earlygame which somewhat requires foreknowledge. And there seem to be way more bosskills when trying to do it efficiently and way less exp in general. Also, the final boss has overwhelmingly ridiculous stats; it's something you should plan for from the beginning. Like, it might be legitimately impossible depending on how exp was allocated. It's very impressive you're doing an efficient run first time through. I can't imagine trying that at all. Like, the first time I did an efficient run in this game was a HM draft and it was surprisingly hard (no Frederick, but still...). >_>

Doable in theory, but doesn't seem fun at all. Something silly like 3rd Gen Manakate Donnel!Nah!Morgan with Aptitude could probably do it somewhat reliably.

[spoiler=Links to various chapters] Premonition: 2/2, Prologue: 5/7, Chapter 1: 4/11 Chapter 2: 7/18 Chapter 3: 5/23 Chapter 4: 5/28 Chapter 5: 6/34 Chapter 6: 4/38 Chapter 7: 3/41. Chapter 8: 5/46 Chapter 9: 4/50. Chapter 10: 2/52 Chapter 11: 3/55 Chapter 12: 3/58 Chapter 13: 1/59. Paralogue 12: 1/60. Chapter 14: 1/61. Not super duper optimized at all, particularly early on, I just wanted to see what turncounts roughly look like with a few conditions. - No Spotpass, DLC. - Anna shops, event tiles, barracks, etc are fine but I probably won't use them. - Renown starts at ~330 (maybe more so Celica's Gale is available for Walhart, etc.) - All Paralogues and Challenges are optional and count for turns - For the purpose of this topic, efficiency refers to fast and reliable clears (tending toward the lowest expected turn count). I think I generally favor more reliable clears, or strats that transition, compared to other LTCers. Decisions I'm making. - Not resetting for growths, etc - I do not like sacrificing units if possible. - +Def/-Luck. It's very possible +Spd or +Mag or something is better in the long run, but +Def snowballs quicker. - Not using Galeforce. It's probably faster in the end since the Renown Seal is there, but w/e. >_> - Testing a core of Griffon Avatar, Griffon Panne, Great Knight Lucina, and Rescuebot Morgan. Just for fun. [spoiler=Premonition: 2/2] Yep [spoiler=Prologue: 5/7] Essentially followed Redwall's strategy. Kill middle Myrm, then Mage while luring the top group, then Mage again while blocking some spaces, then boss. Chrom was unequipped the whole time so there were 4 nice and easy kills for Veterantar. I didnt really see a configuration where its more favorable to have a nonzero Dual Strike chance. I prefer taking advantage of Veteran over an unpaired strat. Also a minor modification at the end. If Frederick has 8 or more health, he attacks the boss with the Bronze Sword on turn 4 with Chrom unequipped. If Frederick has 7 or less health, Frederick heals and equips the Silver Lance with Chrom unequipped. Chrom finishes on player phase either way. The former occurred for my run. We eschew the 4 turn here (requires a Dual Strike or Frederick being at higher health and getting the bosskill) in favor of a more reliable 5 turn with Chrom consistently getting more exp. We also don't have Chrom face a 2hko at 65% true hit. Frederick's chance of death is probably ~5%, but there are always those rare outlier cases like Gamble crits or multiple Frederick misses on Barbs which result in death or lengthening the chapter. Imo, it's probably not worth the turn, but w/e I went with it. I think it's well understood 4 turns is easily possible with Chrom getting the kill. As well as an alternative and even safer 5-6 turn. [spoiler=Chapter 1: 4/11] Random enemy movements, Frederick's level up, random skills and various inaccurate attacks can all cause trouble here. There are usually ways to consistently 4 or 5 turn using a few key ideas. Pair Chrom and Frederick so they double Fighters. Pair Avatar and Lissa so Sully and Virion, unpaired, are available for chip. In my case, Avatar with Lissa with chip from Sully and Virion just kills the Archer which was convenient. It's pretty important to chip the Archer a little bit on enemy phase while still being relatively safe. I like to have Frederick with Chrom end the first turn 1 space left of where Chrom starts. Avatar with Lissa just out of range of stuff. If the Barb attacks Frederick from the South (Merc to the left), then Avatar finishes the Barb from the adjacent square on the right (allows him to counter the Archer on enemy phase) at high accuracy, Frederick kills the Merc from on the fort; he just kills with a Bronze Sword, which also gives him triangle against the coming Fighters. Virion trades the Elixer. Sully moves Up. If the Barb is to the left and Merc to the south, Virion moves and trades Avatar Thunder away so he doesn't steal the kill and chips the Merc. Avatar moves adjacent in the same square as the other case, trades for Thunder and the Elixer, and finishes the Merc. Frederick kills the Barb from on the fort. Each case appears to happen randomly 50/50, but at least that's half the time Avatar isn't attacking the dodgy Merc (against Patience, Avatar only has 78% true hit). Either way, we have nearby enemies dead, the Elixer to Avatar (not sure if this matters or not), and Frederick on the top fort by turn 2. At this point, Frederick has to dodge the Hammer (it's quite low true hit on a fort) but everything else is pretty variable for the next 2 turns. If Frederick got Str on level up, he can one-round Fighters that don't have Hp+5. If he gets Spd, he can double the boss, making it a little more lenient to 4 turn. If he doesn't kill Fighters, it's possible to get Avatar a little more exp but it can be dangerous. Should probably try conserve Silver Lance uses if possible as well. Just need to use Sully and Virion chip safely, and should be able to 4 or 5 turn with Avatar getting the bosskill. I ended up with a pretty common configuration. 1 Fighter was left alive, plus an damaged Archer and full health Merc over on the side. Frederick could weaken the Archer further for Avatar to kill with Bronze, while being in range of the Merc. Avatar finishes the Archer while in range of the Fighter but not the Merc. Turn 4, Frederick weakens the boss, Avatar finishes. Sully and Virion chip and finish the Merc. If Frederick didn't double the boss I think I would have finished the Archer with the Silver Lance so the Merc would be more heavily damaged and require less chip. Now reading through my strat, I just realized that maybe the initial Merc doesn't die (turn 2, in the case it's to the left) if Frederick doesn't proc Str. For some reason I never ran into that case. I suppose Frederick can use the Silver Lance first turn to solve that issue. Or maybe just keep Virion out of range and have Avatar to counterkill an extra enemy. I already hate Chapter 2 again. >_>

I admit I just threw out random numbers. All I actually remember is I babied her for a bunch of turns before she was usable. Here's a few real numbers: Base Sully has 20 hp/8 spd/7 def. Enemies by C2 have 22-26 atk. Also, she starts with 10 atk. Glancing at a few enemy stats, I see some 32/4, 33/3 in terms of Hp/Def. Sully in her best case scenario with Frederick 3hkos and is 2hko'd back. Except the team is outnumbered by very aggressive enemies and Frederick needs to actually tank stuff or help Avatar tank. I suppose she doesn't want to get doubled by Mercs and orko'd, so she Pairs with Stahl? 12 atk is a solid 4hko I guess. -_- What is Sully actually doing besides begging for kills? When is she actually going to be not a "weak" unit? Why would we take the time to give her those kills when we can perform faster and more reliable clears through other means? I'll ask again, how are you making her a competent combat unit in a reasonable timeframe? In what capacity is she contributing to efficient clears? (like, specific examples, not "you need her to kill stuff") Panne, Lonqu, I'd say are better combat characters. Lon'qu is almost competent at base due to Avoid and high Spd. Panne actually has competitive stats when reclassed. They also have utility purposes in terms of Pair Up. Tharja and Henry have Nosferatu. I said this a while ago when someone wanted to make a Lunatic list: Avatar and her children are overcentralizing (as you've noted). It's an inevitable consequence of Lunatic stats and Veteran being so broken. Moreover, Pair Up encourages a small combat team with lots of others for utility. The huge number of bosskills mid/lategame encourage a small combat team with lots of others for utility. I see no reason to rank a drastically inferior combat unit highly just because she's better than the other even worse combat units. Note that she nearly tops the list of the scrubs, as expected. Other B rank characters are there for primarily utliity/support. Which we value highly by the present criteria because they help cut expected turns. Combat is not the only way to contribute. Good at specific utility > Being a bad combat unit. If you don't like the current criteria, do you have alternate self-consistent criteria for ranking characters?

Have you used her in Lunatic? She's orko'd and like 6hkos or something silly for quite a while unless you take 20+ turns babying her. Err, and to clarify. I have used her in Lunatic for my No Avatar run and No Second Seal run.

If Chrom only got to 12/1, and a bunch of those kills were stuff he had effective damage on (Knights, Wyverns), how is Sully going to become a competent fighter within a reasonable timeframe? Why would we slow down to train weak units when we can steamroll with strong ones? This is something that happens across all FE, most apparent with FE7 Marcus, FE8 Seth, etc. btw I think Chrom is too high though. >_>

Seems like we will eventually get the last 3 that Japan has but they're every 2 weeks now. And Europe even later.

Bonus box, where you get the paralogues and summon the teams, has a set of free legendaries. We don't have the full set yet (of note, missing Naga, Forseti), but will eventually. They're not thaat op, I was mainly asking clarification.

Re: Spotpass. If yes, does this also imply we are allowing the Bonus Box legendaries too? The ability to hire Spotpass units if we wish to spend the turns/gold? How much is Camus anyway? >_> Introducing easy access to Celica's Gale/Braves/Mire/Legendaries (Book of Naga. :O) could certainly shake things up though... I don't really mind (the items at least), but the consequences are hard to gauge.

Finishing the main story up. This was written quite a while ago, I just finally got around to posting it. >_> [spoiler=Chapter 25 in 1 turn:] General strategy and map overview: Enemy density isn’t that high but enemy stats, skills, and weapon variety are very dangerous. Plus there’s Mire raining down on your team. Most enemies can be dealt with by Rallied and Paired units. The Dark Pegasus enemies are particularly dangerous because of their huge mobility and Tomes. Manakates or units with Nosferatu/Sol are probably the only units that can realistically take multiple hits from the variety of enemy and weapon types. Avoid stacking from Wyvern or similar units isn’t that reliable here because of Magic and effective damage hitting so hard. If not 1 turning, it is still necessary to move somewhat quickly or flying reinforcements will arrive and ambush the team. Also moving relatively fast still helps to not get swarmed and overwhelmed in general. Because of the lowish enemy density, a balanced team (but well buffed with Rallies and Pair Up) can still exploit favorable combat matchups for a reliable clear. But this chapter is extremely simple to complete in 1 turn with Rescue staffs, a flier and Acrobat Anna, so I would recommend that strategy most. Essentially just use Rescue Staffs and Dancing to leapfrog units forward and take advantage of Anna’s and fliers’ ability to ignore terrain costs to traverse the mountains. Kill Aversa quickly with effective weaponry. No Second Seal strategy: This strategy is for a reliable 1 turn clear. Mag tonic on Morgan. Str Tonic on Sevara. Lucina and Gerome get 2 kills with Olivia’s help. Anna Rescues Morgan who flies to kill a Berserker then Galeforces to Aversa. Sevara buffed by Rally Spectrum and Rally Strength onerounds easily. Stat benchmarks/Other RNG/misc comments: Essentially 100% reliable. I forgot to check Sevara’s hit rate. I think it should be 100%. I do know the Mag tonic was for Bolganone to cleanly one-round the Berserker. Also Lucina and Gerome got their S Support from the 2 battles in this chapter. ^_^ [spoiler=Endgame in 1 turn:] General strategy and map overview: Reinforcements never stop, so rush the boss. Take the first turn to move everyone closer. On the second turn, use all the Rallies you have and kill the boss. Rally Spectrum and Rally Strength (remember from Basilio) are probably the most important. Rally Speed if necessary to meet the 50 Spd benchmark to double. Might as well use any others if you have them. Have a strong unit with A or S support with Chrom or Lucina attack with Brave weapons. Chrom and Lucina will Dual Strike often and do heavy damage. They don’t work as well as the primary unit because the boss has 80% Pavise activation. If necessary use Olivia to attack again or have adjacent units Transfer around the Falchion users and attack with them. Do everything possible to get as many Dual Strikes by Chrom and Lucina in and you should be able to reliably clear this chapter. Using Magic can work as well, but will probably require Ignis or Luna to do significant damage. No Second Seal strategy: This strategy is for a reliable 1 turn clear. Str+Skl+Luck tonics on Gerome and Lucina. Avatar Rally Spectrums. Basilio Rally Strengths. Libra Rally Lucks. Cherche with Anna flies way up to Rescue Olivia. Sully with Lissa goes way up to Rescue Gerome with Lucina. Cordelia Rescues Morgan with Sevara who kills the Berserker in the way. Gerome attacks with a Brave Axe and Olivia dances if necessary. GG. Stat benchmarks/Other RNG/misc comments: Gerome does 12 (13 with Special Dance) x 4 with a Brave Axe. Lucina does 16 damage at 93% Dual Strike chances. Not perfect hit, but still very high. Olivia gives another round of attacks. Morgan and Sevara are also there for a little more damage. Easy enough. [spoiler=Some misc comments:] Really, a no Second Seals (+other restrictions) run in Lunatic just tests one’s babying skills. Once the children get a few levels, they can really easily take on the rest of the game. Stat inheritance from Chrom+Sully’s combined 55ish levels resulted in units that dominated up until Chapter 24. And even then, Lucina and Kjelle were on par with Flavia+Basilio and only had their issues exposed because I was doing those chapters so quickly. They would be fine in a more medium paced clear with Rally spam. Inheritance from parents with 65ish combined levels resulted in super Gerome and Sevara. Though I think it’s very likely Sol played a large role in that and their classes of Wyvern and Hero are toptier. Morgan clearly wasn’t as strong b/c of Dark Flier bases, but she was mostly a Galeforce ferry lategame anyways. I don’t think anything was particularly RNG blessed except Gerome’s Str and some stats actually look below average. Regardless, children with no reclassing are about on par with early reclassed parents in a more typical run and it’s a similar mid/lategame steamroll. It would be similar even with no self-healing like Sol. The stats alone are still enough for reliable, if slower, clears. The slowest chapters were probably the early ones because I was training Chrom/Sully (and I didn’t care at all to go quickly). The children were trained fairly fast even with a conservative pace until Chapter 21. Optimized clears were surprisingly fun later on once the children’s growth maxed out and I had a variety of units to use. Anyways, I really should have passed Ignis instead of Rally Spectrum to Avatar+Cordelia’s children and maybe Speed+2 to Sevara instead of Rally Speed. I had anticipated more turtling with Rally spam but instead Morgan and Sevara were awesome and destroyed things. Ignis would add tons of reliability to their attacks. We’ll see if I find uses for a forward, more durable, Rally user in postgame maps. I also shouldn’t have done Luck flaw for Avatar which slightly hurts Armshrift Helswath, but it ended up okay I think. Kjelle probably should have gone Great Knight for the move and Luna. Rally Defense was pretty useless (again, much less turtling than I expected). Pavise is okay, but Luna would be so much better to increase her offense. It’s actually very debatable if Sully should have gone Great Knight as well so Lucina got Luna. Kjelle and Lucina have pretty lackluster 1-2 range offense lategame/postgame without it. Gerome is awesome. Dat Strength. Plus huge Hp, Def, Avoid and Sol. And 10 move. Comes late and in a hard paralogue though. Resources allocation was okay. I didn’t pay much attention to Gold at all, forged when I wanted, bought Rescue like crazy, and still ended with 15k. As for statboosters the Spirit Dust on Morgan was probably useless. Even though her magic is bad, she doesn’t need that much for her purpose. Statbooster spam on Sevara was good, since she’s by far the strongest fighter with no weaknesses and amazing 1-2 range with Armshrift Helswath. Probably should have saved more misc. statboosters for Aversa though. Regarding earlier chapters: they shall remain low priority for now, unless I get requests or something. Midgame Lunatic isn’t that interesting anyways imo, because you can outstat the enemy by so much. Maybe I’ll writeup some of the harder paralogues since they were pretty challenging considering how early I did them. Regarding postgame, my plan is this: I’ll play through every DLC once and use most of the early ones as an excuse to train Flavia/Basilio and some of the parents a little more (maybe also marry off Olivia and Libra and build up misc. other supports). The characters and scrolls may be useful as well though obviously no reclassing to Bride/Dread Fighter. I’ll then eventually do optimized clears of all the Spotpass paralogues and harder DLCs like Rogues 3, the Challenge Pack, and the Future Pack. Likely no Apotheosis though easy path (and maybe even secret path) is probably “technically” doable if planned for from the start with children from 2 20/20 parents + level up rigging + statbooster buying and etc. >_>. Though maybe I’ll allow Second Sealing into the same class for stats alone to emulate those stat cheese methods. But most likely not. No Apotheosis unless it’s really interesting for some reason. All the Spotpass paralogues/DLCs are beatable without extra training, but will likely be slower and less reliable. Besides, the main fighters are all at 20/20 already. The real reason is I just want to use Basilio and Flavia for this but they need to build up support for better Dual Strike chances. Also DLC nostalgia is always fun. From just leisurely going through the postgame stuff, they seem quite challenging and fun in No Second Seals. I think they’ll be interesting logs once I get around to doing/writing them. However, my priority for now is probably a Lunatic efficiency run or Lunatic+ PMU, leaning toward the former.

Re: criteria. I agree this should be resolved first. I prefer a low chance of death because it rewards more thoughtful strategies and potentially makes more characters viable (note I wouldn't necessarily reject strategies that don't meet it, they're just not as preferred. Much current evidence doesn't meet that lofty and ideal optimum). Recruiting all and not sacing them seems to be more in the spirit of FE (there's no unnecessarily convoluted cases in this game), so if you prefer that then it's fine with me. I can already see potential silliness like C20/C21's reliable 2 turn clear be boosted by mass sacrifices to preserve various utility unit lives. But yes, it's inevitably arbitrary, and the alternatives may be faster/more reliable. Make a poll or just decide or something. Do people like arguing these particular details? >_> It would be useful to have the Renown rewards we do already have after C3, when it opens, listed in the OP.

Vaike gives a decent early Pair Up and maybe Hero can help in routs, even underleveled, due to Sol. Hero is also one of the best Pair Ups, giving Spd and Def. Does Virion do anything at all? Chip Wyverns? Say’ri gives 7-8 Spd even without Support. That’s probably worth something given how high Spd benchmarks get. She’s also probably the one supporting Tiki since Anna has Rescue duties. Tiki with Say’ri and Rally Speed at base is a competent fighter and she snowballs insanely fast. I know she has uses in 23 and 24 at least. Though if we want to get technical, I think Tiki costs turns overall (there’s also no way Anna is that high). To that end, I was thinking we can consider either optional paralogues as stated or mandate the auto-unlocked paralogues (1-4, 17, spotpass. Turns still count for those. Child ones are optional). More characters, more content, etc. Or is the ruling that they are free? The latter also gives reason to train more units beyond Avatar and Morgan (maybe?). I want to comment on Morgan vs. Lucina but maybe later. >_> Maybe split C and give Donnel his own tier.

Eh, it makes sense. I was wondering why 330 in particular since people probably play a random number of child paralogues and skirmishes. @Chiki: Is this seriously a semantics thing? >_>

We can reduce most chapters in FE history to trivial in complexity if we're taking 40 turns. Oh look, I can take 40 turns on 4-4 in FE10, luring 1 enemy at a time with Nailah. Or I can turtle in a corner in 4-3 and let the BK rout everything. That requires no thought at all. Obviously we should do it that way, no? This is the whole point of trying to have fast, reliable clears. -_- I see the main difference is, 1-P is pretty trivial with or without Edward. The 4-6 turn clear in Prologue with Frederick is not trivial. It requires precise positioning, along with contingencies depending on how the enemy randomly moves, plus random misses due to Avoid+10 and weapon triangle. Not to mention Frederick still saves numerous expected turns in other earlygame chapters as well (Chapter 1, 2, 3 clears are all significantly more efficient with Frederick, and other chapters as well. Probably everything up to 12 or so). A side note, I don't actually like the term complexity, as defined, at all (see: dondon arguments in Tiering Philosophy). I just mentioned it to address the intuitive notion that not all turns are equal (who cares about Premonition?). Still, I don't see why the hundreds of turns Frederick saves early on is worth that much less than the few turns other characters save later on. If you feel that strongly about it, any further meta-complexity arguments can go in Tiering philosophy so I can ignore them. >_> I agree with you about Lissa. In that if they don't actually do anything later wrt to expected turns, we can sac them. However, Lissa probably increases reliability and therefore improves expected turns, by healing. And has Rescue, Magic Cry, Speed Cry. Re: Renown. Why is 330 the baseline? 0 makes more sense. Doesn't matter too much to me, however. Re: Lon'qu. I don't know if I agree with that specific position but he has an invaluable speed Pair Up. He's one of the better earlygame fighters with high avoid against the early axefest. Can go Wyvern for Panne-lite later on.

Apparently no chapter is complex at all if we turtle, who knew? -_-

Aren’t you using Frederick then? Also, if you do mean Avatar as the main fighter, please show us your reliable 5 turn strat (have fun getting doubled by Myrms, etc). Prologue is not trivial if people have trouble just beating it. Not to mention actually trying to get a reliable 4-6 turn. Just see how long and detailed those write-ups are from Redwall, Colonel M, Kngt of Titania. Just how many lategame maps can be summarized by lolrescue, anyways? >_>

Earlygame chapters are not so trivial as to outweigh Frederick saving literally like 999 (probably underestimating) expected turns. >_>

Re: Premonition. We can actually just consider it trivial by complexity, so turns saved there mean little. That's a fairly clean resolution. Re: Prologue: Expected turns without Frederick is stupidly high. Part of the reason I suggested a higher threshold is to have a slower pace and make more characters viable (in theory). We can consider it somewhat in terms of diminishing returns. The difference between 90-100% survival is less important than the difference between 55-65% survival. A threshold is fuzzy at best (i.e. strats that are 59% vs. 61% are not so different) so it also helps if we minimize that dropoff effect of that arbitrary cutoff. Basically I'd prefer tending toward higher success if possible. 60% just seems low. Maybe that's just me. On the other hand, ~100% is indeed very strict, particularly when accounting for growths, so.... Probably just keep the current criteria until we observe a case that seriously tests it.

I understand your self-consistency concern, but we could try something like computing separate "best-case" and "worst-case" expected turns for a range. The former is derived as written, the latter is derived by something like the reasonable max of turns/%chance of not failure or death (obviously 6 turns with 1% chance or something demolishes this metric so we'll just handwave and say it's 5/0.64 = 7.81). For prologue, it's 4.53-7.81 with Lissa and 4.61-7.81 without her. Else, once obtaining the expected turns with no death, simply divide by probability of not failure or death. 7.08 vs. 7.20. Note that failure in this context means not achieving a goal such as visiting a village in time or not killing a certain enemy in the way. Something that changes the strat completely that's like death, such that reseting is preferred. Not as clean methods, but I think they can give reasonable values. The latter is particularly simple and I can't immediately think of an example that greatly strains intuition. Else, if we do wish to be more rigorous, I would personally prefer a much higher cutoff, like 95% or even ~100% (as possible). Essentially, to the best of our ability, favoring higher order strats with contingency plans over binary success or failure. If we're ever focusing on minute differences between character roles in ~95% and ~100% strats, that's a good thing. 60% vs. 100% is a more significant matter. In the end, I'm fine with tiering under these criteria as written, but the implications can theoretically lead to quite dissatisfying conclusions, which is never a good feeling. >_>. If there's no quick and easy compromise, I'll concede to the established criteria for this topic. There may not be cases in this game where this hypothetical significantly shifts rankings. I like the Premonition!Chrom/Avatar idea, xd. Or we just consider the expected turns consistent with the criteria, it's probably just a turn or so which shouldn't shake up things too much.

Suppose we have 60% chance to clear in 1 turn, or death. Would you consider it superior to a strategy that has a 60% to clear in 1 turn and 40% chance to clear 2 turns?

Because 0/10 no Galeforce.