Tiering Philosophy - It's that time again

[Hawkeye]

Average and mean are the same thing. And there really isn't any need for that, it's a problem that really only exists with Edward.

[Snowy_One]

Not really. Let's say you have ten houses. Nine of them have two people, but the last one has 20 people jam-packed inside. The average number of people per house is 3.8 as that's simply the number of people divided by the number of houses. The mean is the most common number in the set, in this case, 2.

Anyways, there is more to why I'm suggesting it. Namely fliers/rescue-droppers/warpers/the like. A lot of their value in turn-shaving comes from rescue-dropping other units. So the unit could be rather bad themselves but allow some other unit to do amazing things on specific chapters/with specific teams/both. So if we focus on the mean number of turns saved as opposed to the average/LTC, we can get a more accurate reading of unit quality as opposed to seeing units who may shave a lot of turns off in specific instances getting a higher than some other units.

[Baldrick]

Snowy, you're thinking of the mode. Mean is just another word for average.

I am of the opinion we should look at the circumstances of a chapter to judge how much value should be assigned to a given contribution.

Edited May 6, 2013 by Baldrick

[Snowy_One]

While I hold no problem with that (least in that wording), I would very much appreciate something more concrete to put down as a tiering guideline as that is simply a bit too open to interpretation for all ways.

[Hawkeye]

Not really. Let's say you have ten houses. Nine of them have two people, but the last one has 20 people jam-packed inside. The average number of people per house is 3.8 as that's simply the number of people divided by the number of houses. The mean is the most common number in the set, in this case, 2.

Lol

Anyways, there is more to why I'm suggesting it. Namely fliers/rescue-droppers/warpers/the like. A lot of their value in turn-shaving comes from rescue-dropping other units. So the unit could be rather bad themselves but allow some other unit to do amazing things on specific chapters/with specific teams/both. So if we focus on the mean number of turns saved as opposed to the average/LTC, we can get a more accurate reading of unit quality as opposed to seeing units who may shave a lot of turns off in specific instances getting a higher than some other units.

Rescue-dropping is generally pretty consistent. You're transporting a strong combat unit, a lord, or both. You're obviously using the lord throughout the whole game, and if you're using rescue-dropping you're probably being consistent with your combat unit as well, simply because they're going to be getting more experience. Also, the ***mode*** number of turns saved isn't that useful because honestly it's going to be 0 for most units, due to things like defend chapters, and the fact that most chapters can be completed in the same amount of turns with very different teams.

And Baldrick is suggesting a complexity rating similar to what Olwen is using in his FE9/10 tier lists. It's a good concept, it just needs to be worked on a lot more.

[CT075]

Not really. Let's say you have ten houses. Nine of them have two people, but the last one has 20 people jam-packed inside. The average number of people per house is 3.8 as that's simply the number of people divided by the number of houses. The mean is the most common number in the set, in this case, 2.

that's the mode

[Redwall]

Do concrete tiering criteria kill discussion? I personally feel that precise definitions are required to have any meaningful discussion at all.

As far as I'm aware tier lists have never had rigid definitions and many people have argued, in the past, that

a) rigid definitions are bad for discussion

b) rigid definitions create a tier list that applies to a few set of restrictions as opposed to a broader, all(well, most)-encompassing tier list

"brisk pace" is not explicitly defined so that I can say "don't be morons" while not having to explicitly ban things that kill discussion. It will never be rigidly defined because its wording is designed to be vague.

Consider the FE13 tier-list playstyle. In his playlog, bearclaw13 logged 98 turns by the end of Chapter 11. Meanwhile, Interceptor logged 75 turns by that point in his own playlog, and General Horace 65 turns. It seems that based on these playlogs, there exists some confusion as to what constitutes a "brisk pace."

This is a problem since, obviously, character rankings can change wildly depending on how quickly we play. If we complete Paralogue 4 in three or four turns, which I would consider standard since the fastest clear is a two-turn completion with very precise Rescue usage, then a growth character like Nowi will have her lunch eaten by a more mobile unit (Sumia, Cordelia, Wyv Panne, Wyv Sully, reclassed Avatar), and will subsequently be of no use in the later chapters. If we complete it in six turns, then Nowi has the opportunity to gain about five levels. Nowi is slow enough that even with a C-ranked Gregor support and a Speed Tonic, she won't be doubling the Paralogue 4 enemies (who have at worst 10 Spd). For this reason, she has at best a minor role in the most efficient clears of P4.

With well-defined criteria, one could argue Cherche over Nowi or vice-versa with some confidence. As it stands, it could go either way depending on how "briskly" we play.

[Chiki]

Yeah, the reason tier lists today are failures is because everyone has a different idea on how fast to go.

I logged 49 turns, if anyone cares.

[dondon151]

they are not any more failures today as they were 4 years ago, when players had even less of an idea of how fast to go.

[Chiki]

How fast should we go then? Where is this arbitrary limit which tells us how fast we should go?

I disagree simply because playing with ranks means that there is a tangible goal (5-stars, A rank, w/e) whereas not playing for ranks means there's the arbitrary, fuzzily defined goal called "efficiency." All of Vykan's arguments in that thread are true, though.

I personally think Funds should measure the total amount of assets obtained over the course of the game instead of the total amount in possession at a certain point in the game, because we want to use those stat boosters, promotion items, and awesome staves.

What happened to your views?

Edited June 3, 2013 by Chiki

[Scarlet]

Why not just set an arbitrary turn count goal per tier list? Maybe even per chapter? That way you could set the pace at which the player is assumed to go while going for a secondary goal (least effort, best reliability, most resources gained - could be pretty much anything).

For instance, consider a tier list that sets a turn count goal of 150 in an FE that has 30 chapters, while setting the secondary goals to be both least effort and best reliability. As soon as a unit is good enough overall to be able to keep up with the defined pace (roughly 5 turns per chapter in this example), better durability and the like would make a greater difference in how "useful" a unit is. As a bonus, it probably even comes closer to the average playstyle than total LTC tier lists.

Edited June 3, 2013 by Scarlet

[Chiki]

Because it's logically preferable to not create more arbitrary restrictions than necessary. It just violates Occam's razor.

I don't think it's necessary. I think my total LTC tier lists can be useful for the everyday player.

Edited June 3, 2013 by Chiki

[Scarlet]

You obviously need a couple of arbitrary restrictions if meaningful discussion cannot even happen in the first place simply because everyone has a completely different idea on how fast "fast" is.

Besides, having a tier list with a turn count goal doesn't conflict with having another tier list in which "lowest turn count possible" is the goal.

[Chiki]

As has been mentioned ad nauseam, tier list necessarily has subjective criteria, since a tier list ranks characters based on criteria we subjectively prefer; that doesn't mean we should add more criteria than we need. Although this set of criteria remains subjective, it at least makes some attempt to avoid such ill-defined guidelines as "you must play quickly, but not too quickly lest, gawds forbid, an actually-good unit eats Nowi's lunch."

This tier list considers complexity-weighted turns and reliability. If a character costs turns and/or hurts reliability, and we tier rationally based on these criteria, he or she is a bad character under these criteria. Without adding more criteria, ignoring recruitment costs is inconsistent with the current set of criteria.

[Scarlet]

And I'm saying that you do need more criteria. What's your problem?

[Chiki]

Why shouldn't we go as fast as possible?

[Scarlet]

Is that supposed to be a serious question or are you just trolling?

Going as fast as possible is not the only way to measure how useful a unit is. There are different kinds of usefulness depending on what goal you're aiming for. A LTC tier list is perfectly fine, but a lot of people may find more merit in discussing a more... ...pratical playstyle. Multiple tier lists can co-exist, can they not? You don't need to participate in ones that don't interest you, after all.

[Chiki]

LOL. Yes, I'm just trolling. What a brilliant counterargument. Not once did you address why we shouldn't go as fast as possible.

We agreed long ago that going fast is a good way to measure how useful a unit is, along with being reliable. Those two things are the criteria we take for granted in tier lists. None of us are really interested in tier lists that don't involve going quickly, period.

Edited June 3, 2013 by Chiki

[Redwall]

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.

[Scarlet]

@Chiki:

That was a question, not an argument. Wouldn't be a first to see you asking very questionable questions (which can be taken as trolling depending on the point of view and whether you're doing it on purpose), so I just felt like making sure!

There's a difference between going quickly and going as fast as possible. In an FE where the lowest possible turncount is, say, 100 turns, a playthrough that takes 125 would still be quick while allowing a lot more leeway in how to tackle things.

Besides, I think the others can speak for themselves.

@Redwall:

If that is what you were trying to do, yes. However, if you could, for instance, also achieve a 100 turn clear, albeit with a 50% chance of failure per chapter, then you could create two different tier lists - one LTC tier list that goes for the lowest turn count possible no matter the risks, and another that goes for a turncount that can be reached with 100% reliability (120 turns in this case).

Then again, it now occurs to me that this isn't much different from what has already been done - the problem may lie more in the fact that certain people just cannot respect certain tier lists' ranking criteria and try to measure them using their own. No change in criteria can solve that problem.

So disregard my idea, I suppose. I follow the tier lists around here only loosely, so it was just an idea that came to mind when I read this thread.

[Chiki]

There's a difference between going quickly and going as fast as possible.

I don't think you have any idea on what you're talking about. I'm asking you why we should go quickly rather than go as fast as possible, but you've been dodging the question for multiple posts now.

Edited June 3, 2013 by Chiki

[Da Bear]

I don't think you have any idea on what you're talking about. I'm asking you why we should go quickly rather than go as fast as possible, but you've been dodging the question for multiple posts now.

Going quickly gives us the opportunity to use several units who we can't realistically use if going as fast as possible.

[Espinosa]

It's quite simple I think; the more a unit slows us down, the lower they should be on the tier list. But that's just one of the criteria, amongst which you have bases, reasonably possible potential, availability, need for resources, etc.

[Redwall]

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."

Edited June 3, 2013 by Redwall

[Da Bear]

It's quite simple I think; the more a unit slows us down, the lower they should be on the tier list. But that's just one of the criteria, amongst which you have bases, reasonably possible potential, availability, need for resources, etc.

This is very true.

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."

Okay. That said going "as fast as possible" is often less reliable. Do you include reliability Olwen?