Register

Author Topic: Theorycrafting! (Because I like competitive metagames too much)  (Read 133675 times)

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #725 on: August 28, 2016, 07:35:47 PM »
5e

So...let's talk about the SCAG cantrips.

http://engl393-dnd5th.wikia.com/wiki/Booming_Blade

http://engl393-dnd5th.wikia.com/wiki/Green-Flame_Blade

Most of the analysis I do is stuff in the core three rulebooks only. PHB, DMG, MM.  But there is technically one other book, the Sword Coast Adventure Guide, which contains...two cantrips that people talk about constantly, and I dunno what else.  (Is bladesinger Wizard in there?  I think it might be.  OK, so that's pretty strong too).

I'm not sure what to make of these cantrips.  By level 17 if you get full value out of them you're looking at 7d8 damage.  This is 31.5 damage.  This is similar to sneak attack after 17 levels in Rogue (9d6 = 32.5 damage).  You can do both, of course, and have a Rogue hit really hard.  But on the other hand, Rogue could use the buff so....

In general, they encourage making single attacks instead of multiple attacks, which I guess is fine since typically the most optimized damage I found was via multiattacking.  They encourage melee, which is also fine since hey, Sharpshooter+Crossbow Master is pretty good.  They're a pile of damage added on if you hit, so you're not encouraged to use the +10 on great weapon master.

All of this is actually pretty reasonable, and offers a different vector to build characters.  Where things get a little strange is that Sorcerers become pretty good at melee.  These are cantrips, after all, and that means you can Quicken them, and do them twice in one round.  Draconic Sorcerers can also add their Cha to (one of the damage rolls of) Green Flame Blade.  I've actually seen someone on a forum ask "why play Paladin when you can just play Sorcerer instead?"  An interesting question.

So...making a Sorc front-line build here...I think we want Warcaster (lets us make attacks of opportunity with SCAG cantrips).  I think if we're really focusing on melee attacking, we want to dip another class for shields and a better weapon.  (Fighter 1 gets shields, and, say, Dueling fighting style).

Options other than Fighter are...Paladin (meh.  Like...ok, you could go level 2 for Fighting Style and smites.  Except smites aren't that great--if you want to deal more melee damage in a day, you'd do better converting some spell slots to sorcery points so that you can quicken more cantrips).  Ranger (meh).  Druid (you get a shield, but no weapon).  Rogue (you get a weapon, but no shield).  Cleric (yeah, this is a solid reasonable option, particularly because it keeps your spellcasting progression going; needs 13 wisdom though.  But I think taking Tempest Cleric to level 2 for the maximize channel divinity is a solid option; that build is less about being a melee master, though so I won't focus on it now).  Barbarian (you can't use this cantrip while raging).

Once you have Fighter 1, eventually you probably want to grab Fighter 2 for action surge.  And once you have fighter 2 on a character that focuses on melee attacking, I think Fighter 3 (Battle Master) is probably worth picking up; especially on a character that attacks once or twice per round (you can spam maneuvers and probably not run out).  And Fighter 3 is a fine place to stop cause hey, technically you'll still reach Sorcerer 17 for the level 9 spell slots.

I think being a dex focused character is fine for this--Draconic bloodline sorcerers can have 13+dex AC, which assuming you raise dex to 20, reaches 18 AC, 20 AC with a shield.  HP will be a little low for a front-line character, so having a shield is probably desireable.  Obviously, going fighter heavy armour and strength based weapons are an option.  But they're not even that great.  (Two handed swords get up to 8.333 average damage with the appropriate fighting style.  A one handed dex based rapier with duelist is 6.5 average damage.  When most of the damage comes fron the cantrip, I'll take +2 AC over an extra 1.866 damage).

Optimising hard for race...I think you'd want Variant Human for War Caster, just because your ASIs will be pretty delayed.  That said...Half Elf lets you start with 16 Dex/Cha/Con, while a variant human would have a 14 in one of these (Con or Cha).  Once you get to the first ASI, Half Elf gets warcaster, Variant Human gets 18 Dex, and still has that 14 in another stat.  But the 18 dex is pretty damn important (+1 AC, +1 hit, +1 damage).  So mmm...variant human sounds better TBH.  With all that being said, variant human does technically lose you 2 HP cause you can't take level 1 in Fighter (spellcasting is a pre-requisite for War Caster).

What stat would it sandbag between Cha and Con?  Not sure, actually.  With Cha...it does represent damage with Green Flame Blade (baseline, and you get to add more Cha at level 6) and obviously it makes your spells better (although you might be focusing on spells that don't rely on Cha, like polymorph and haste).  Still...you presumably still take Fireball?  And 14 Cha vs 16 Cha is a decent upgrade on a Draconic sorcerer's fireball.  Con represents HP (which you need more of, due to d8 hit dice from Sorcerers), and concentration saves.  Things that will come up frequently, although it's not a massive bonus to either.

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #726 on: August 28, 2016, 09:16:54 PM »
Quote
I've actually seen someone on a forum ask "why play Paladin when you can just play Sorcerer instead?"

I hope someone smacked them down verbally because this seems like one of those people who can't notice anything but damage.

There are many answers to this question, but the simplest answer is the paladin auras which do silly things like boost everyone's saving throws by Cha and half all magic damage.

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

Magetastica

  • Denizen
  • *
  • Posts: 174
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #727 on: August 28, 2016, 10:20:52 PM »
In a party with a Warlock, a Sorcerer, and a Paladin, I would definitely say that the Sorcerer was the least helpful of the group. Utility? Paladin does it better with Auras and Smiting magic. Damage? Warlock says Eldritch Blast. (I was the one playing the Sorcerer, too)

Also, Sorcerer only gets d6 HP, so if you are hell-bent on playing a melee Sorcerer, I really don't know why you aren't playing a Bladesinging Wizard. Sorcerer only gets so many times they can Quicken, and if you're splitting between 3 stats, you're going to have a hard time either hitting, dealing damage, or surviving. Bladesong Wizard doesn't need a dip (but benefits almost as much if they do) as Sorcerer, and has a better go of things at being a melee fighter. They get boosted AC, boosted Speed, a better list of spells to help enhance their ability to fight like that, the list goes on. So if we're talking a pure nova, then yes, Sorcerer can probably dish out more damage very quickly, but I'm not sure I see an advantage to doing a melee sorc instead of a ranged sorc for the exact same benefits.

I mean, if you're a melee Sorc using Dex, then your best weapon is a Rapier, which is d8, going off of Dex, and then you use your Cha to add a little more damage to it. Or, you can use Firebolt, d10 and same scaling progression as GFB, Cha based, and still get your Cha damage boost to it. The biggest benefit to GFB is the splash, which if you're fighting smart enemies they can pretty easily work around that anyway.

I dunno, as someone who's playing a Bladesong Wizard and has played a DPS-based Sorc, I feel waaaaaay more useful as the Wizard, including helping deal damage in fights.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #728 on: August 29, 2016, 06:31:50 AM »
The biggest benefit to GFB is the splash, which if you're fighting smart enemies they can pretty easily work around that anyway.

If they're splitting up for splash, then you use Booming Blade.  They can choose not to move, but then you've effectively lowered their movement to 0, which is pretty cool.

EDIT:

Eldritch Blast, yes, is good.  (Although Sorc can pair that with twin spell too, so once again Sorc can do it better).

Draconic Bloodline Sorcerers get extra HP (so more like a 1d8 HP class).

Bladesinger Wizards are just strong in general.  Like...highest AC at level 1 (when singing assuming you use mage armour).




For sorcerer in general...you should be able to output more damage than a Wizard.  You get all the core damage spells, but you have metamagic, and draconic bloodline can give you a small damage boost as well (to one element).  The big value in general comes from what you can do with metamagic--like...twin spell Polymorph, now you have one person using their concentration to polymorph two people.  Similarly, twin spell haste.

Sorcerers also get Con save as one of their two starting saves,  which is quite nice.

Screwing around with cantrips isn't necessarily the focus of the class, just something they do fairly well.  (See also Warlock 2, Sorcerer X for Quicken + Eldritch Blast).
« Last Edit: August 29, 2016, 07:25:58 AM by metroid composite »

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #729 on: August 29, 2016, 07:13:29 AM »
Booming Blade isn't a splash move; both parts of the attack must target the same monster. As you note, it's a soft-control move being either quite damaging or applying a don't-move effect (sadly, the enemy chooses which). Greenflame Blade always gets the extra damage, but it's unfocused... still great for finishing off an enemy who has just been revived by Healing Word, as my party discovered. Granted, since this is being compared to Extra Attack, it's not that special, I suppose.

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #730 on: August 29, 2016, 07:42:22 AM »
Granted, since this is being compared to Extra Attack, it's not that special, I suppose.

Well...they are cantrips, so they're supposed to sort-of scale for that (with damage increases at 5, 11, 17).  They're also on the high end of cantrips, in that if you get both damage procs off, then they're close to Eldritch Blast.  (If you don't get both damage procs off, then they're a more standard cantrip in terms of damage).

Granted, this isn't enough to keep up with the extra attack class feature on its own, but if you get to do these twice in one round (or three times with an opportunity attack) suddenly they're quite good.

Booming Blade in particular also has the advantage of not really needing any magic stats.  A purely physical character can get booming blade through a feat or a 1 level dip or whatever.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #731 on: September 11, 2016, 01:38:42 AM »
D&D 5e

A cool build that I saw (did not come up with myself) -- Tempest Cleric, variant human with Spell Sniper as the feat.  Why spell sniper?  When you get that feat, you get one cantrip with it.  Pick Booming Blade.  BB normally has a range of 5 ft, Spell Sniper makes it 10ft.  Booming Blade people with a reach weapon (such as a whip).


And...on to much sillier things--Bladesong.  It's not a spell, and it doesn't require concentration.  This means that you can use it while in Barbarian rage!  Now, it has some restrictions.  No wearing medium armor, no shields, and no attacking with a two handed weapon.  This still opens up two weapon fighting, and grappling as options.  (Grappling in particular benefits from the +10 movement provided by bladesong, as well as from the longstrider spell--more movement means dragging grappled people further).

Is this even worth it?  Well...maybe.  To be clear, it's not the sole reason you go Wizard.  (You pick up Longstrider, you get a familiar, you get some utility).  I mean, if you're really maximizing AC, you can go 16 Dex, 14 int/con/str (you're an elf so 16 dex is easy to hit).  And then you use Mage Armor, and get 18 AC at level 1 while having both hands free.  A barb typically maxes out at 17 AC if both hands are free.  So...that's higher.  Realistically, you probably don't want to be a dex barb, though, especially if you're going for grapples.  You'd start with like 15 str 15 con, 14 dex, 14 int, and then use your first ASI to get 16 STR/CON.  This...will get you 17 AC (matches medium armour).  Of course, the cap is much higher, but it requires raising CON or DEX when you probably want to raise STR.

So mmm...it certainly doesn't break the game.  It has some uses (50 move barbarian; 60 with longstrider).  But also has some downsides I haven't outlined yet (starting the bladesing is a bonus action, but so is starting a rage).

So...crunching the numbers...it's a build that exists.  There aren't hugely compelling reasons to play it, but you could use it wouldn't be terrible.

(Where I think there's a lot more potential for a multiclass is with a Monk.  Monks don't want shields or armour.  Wood Elf is already a good Monk, which takes care of the Elf requirement.  Monks can actually cast spells during combat, such as Shield.  16/14/14/14 is pretty doable.  Mage Armor can compensate your AC for not starting with 16 Wis.  So...it's a 2 level dip that gets you +2 AC and +10 movement, and some utility).

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #732 on: September 13, 2016, 05:34:52 AM »
Archiving a link to this post here.  (Basically, I rolled 2d12, looked up two classes on the 5e class table, figured out plausible reasons to multiclass aaaand repeat).

https://www.reddit.com/r/dndnext/comments/52azuk/ive_been_rolling_2d12_and_then_looking_up_the/

Really, this was mostly a brainstorming tool for interesting class combos I hadn't thought of, and class mechanic interactions.  There's some high-power ones I hadn't noticed, although I think my favourite one just for the pure flavour is Druid/Warlock.  (Have Devil's Sight, Cast Darkness, Wild Shape as a bonus action.  Your enemies expect a scrawny caster, to be hiding in the darkness and then get mauled by a bear with demonic eyes).  I mean, that's probably not mechanically any more powerful than shooting Eldritch Blasts from Darkness, but it's still totally sweet.

Probably the highest power one was Rogue with a 2 level dip into Wizard (Bladesinger).  Rogues only need Dex and Con, so 16 Int with starting stats is pretty reachable--pretty nice with Bladesong.  Wizard gives Booming Blade/GFB which Rogues want, and frees them up to pick up whichever Rogue archetype.  Wizard gives Find Familiar, which among other things is a pretty reliable source of advantage.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #733 on: May 13, 2017, 11:24:28 PM »

Captain K

  • Ugly Old Man
  • Denizen
  • *
  • Posts: 1210
  • Saving the world with curry and coffee
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #734 on: May 14, 2017, 03:40:38 AM »
Your terminology is confusing.  Triangle Adept Raven makes me think of the character Raven.  And Assassin makes me think of the Assassin characters like Jaffar.

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #735 on: May 14, 2017, 03:56:20 PM »
In fairness, Nino is an assassin by that definition too, and Jaffar is built to be an assassin by the definition mc uses. (Although you probably shouldn't use a colourless in your core 3 unless your options are limited.)

The Raven tome's name is unfortunate, yes, given the character of the same name. But Henry having a raven tome is just so fitting. (A shame it's the least-used one...)

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #736 on: June 15, 2017, 06:02:55 AM »
So...a while back I had a tier list for vanilla FFT, and there were problems with it (it tried to both give weight to the ridiculous power level of stuff that helps builds like Gained JP Up and Move+1, and tried to compare that to stuff like...Samurai and Calculator).

I think my mistake is that I shouldn't have one tier list, I should have two tier lists.


Play-through Context: Co-op the context of this will be the way in which I have mostly experienced FFT recently, which is co-op play, where each player controls one character, which means a lot of the more obnoxious combos probably aren't happening.  Quickening?  Nah.  CMQ no.  Sunken Stated Dance?  Yeah, good luck convincing your friends that they should petrify each other rather than playing the game.

Two tier lists: Dip and Focus:

Focus So...when people play co-op, they just make things.  Like...someone will be like "I'm making a bard" and it doesn't really matter that there's no really optimised build in FFT that  Bard, they're making a bard.  This would be their focus.  High tier Focuses would be like Calculator, and Summoner.  Low tier in this category would be like Squire, Bard.

Dip Their dips would be everything they get along the way.  Gained JP Up, Move+1, Bolt and Bolt2 from black magic, maybe a bunch of skills from Item.  Maybe once they're done mastering Sing, they take it to a high HP class like Geomancer.  High tier in this category would be stuff that lots and lots of builds should always use like Squire, Wizard, Chemist.  Low tier would be like Lancer, Samurai, Calculator--classes that you are likely to ignore if they are not your focus.

Sometimes classes will be high on one list, and low on another (Calculator is an overpowered focus, but a rare dip.  Squire is a nearly mandatory dip, but a terrible focus).  A decent amount of the time the classes will be similarly positioned, however.  (Mime is dead last on both lists.  Summoner is near the top of both lists).

It's also worth noting that because of the way spillover JP works, the highest value dips will nudge up the value of focuses.  (Ex: being a Chemist focus is generally quite good for the party because everyone has uses for Chemist JP, but not many builds want to delay their build to spend time in Chemist).

I'm in the early stages on this, haven't really cemented an order for either list yet.  Going to work on a preliminary ordering, but at this point it's a work in progress, and I'm quite open to suggestions.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #737 on: June 15, 2017, 07:01:26 AM »
OK, preliminary pass on the dip list.


Dip List (draft)
1   sq
2   wi
3   su
4   ch
5   tm
6   or
7   pr
8   ni
9   me
10   th
11   ar
12   kn
13   ge
14   mo
15   ca
16   sa
17   la
18   ba
19   da
20   mi

Squire vs Wizard I think there's some debate on.  Everybody, literally everybody dips Squire, but all of the stuff gets replaced.  Nothing from Wizard gets replaced; MAU is best in game.  Highest MA in game.  Black Magic falls off pretty hard, but if you're using a wizard for the MA, then you get it for free, so it's rarely a wasted investment.

Summoner for one 200 JP summon is kind-of a disgustingly powerful dip, especially for the JP cost.  It's kind of more egregious than anything in Wizard, but Wizard just has a lot of things you want to dip for (including Black Magic, which is a solid dip for the first two chapters or so).  While there's a whole lot of focuses that use item over summon, it's like...most of the low tier focuses.  Chemist is also pretty egregious though, it just...can get replaced by better stuff in mage builds.

Time Mage is just...clearly ahead of everything below it, can't touch the top 4.  Teleport.  Demi for bosses (with higher hit rate and lower JP cost than Life Drain).  Haste.

Oracle...well, getting a bit of JP in it for Silence Song or Life Drain is pretty legit.

Preist I'm starting to be unsure if they should be so high.  Fantastic dip for a calc, but that's one focus.  Definitely a dip I do use occasionally when I want 110 speed.  Acceptable healing dip, even if I respect Chemist more.

Ninja...go to physical carrier for all those physical skillsets, like Punch Art, and Jump (which...doesn't always want Ninja due to Ninjas often being 1 speed faster than the enemy) and umm...max PA Wiznaibus from Dance.

Mediator I might be overrating dip for Invitation, but it's there.  Dip 200 JP for occasional faith raising when doing easy fights also a thing.

Thief Archer are the Move+2 and Concentrate that make Ninja focuses work.  Archer is dipped by more classes (Charge is an acceptable secondary for a Chemist) but Concentrate can often be subbed for Martial arts or Attack Up or Equip Sword, and move+2 is a pretty reasonable pickup for physical builds that don't want to go near Time Mage (pretty reasonable pickup for, say, Agrias).

Knight is dipped often for Weapon Guard, occasionally for Equip Sword, or Equip Shield.  Occasionally used if nobody else has called dibs on a Knight Sword.  Best physical class in Chapter 1, if you have a physical build and need a bit more power in Chapter 1.

Geomancer is a carrier, but not an impressive carrier.  Wizard carries most of the magic focuses best, Ninja carries most of the physical focuses best.  Attack Up exists and kinda matches Martial Arts, so ahead of Monk certainly.  Good carrier for Agrias.  Can be preferable to Ninja for Jump users.

Monk is mostly an all or nothing class but does have a few dips.  Martial Arts for a Ninja.  Occasionally Counter.  Sometimes HP Restore but it costs more JP than Auto Potion, so only really if you need Monk for a prerequisite.

15-19 are all the advanced classes with hefty unlock requirements.  You don't usually dip any of them.  Tentatively put Calc near the top cause if you have a calc in your party, you might just get Damage Split from spillover--it's 300 JP, and a very good ability.  Samurai...yeah, some people do literally get blade grasp and then use it on a non-samurai build.  Lancer...you don't really dip usually, but I've heard of people using Jump secondary while training a Samurai.  Bard...Ramza could conceivably grab Bard and like Angel Song...and then take one or two songs to Calculator which works well with sing/dance because they don't care about speed.  Dancer is more expensive to unlock than Bard, and in the wrong part of the tree for Calc.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #738 on: June 15, 2017, 08:08:41 AM »
preliminary pass on the focus list

focus list (draft)
1   ca
2   tm
3   su
4   ch
5   ni
6   or
7   sa
8   la
9   da
10   pr
11   mo
12   wi
13   me
14   ge
15   ar
16   kn
17   ba
18   th
19   sq
20   mi

Calculator...you can expect an uncooperative party that won't equip Chameleon Robes.  Calc is kind of famous for being busted, though, and that stays true even with uncooperative teammates.

Time Mage as #2...it's something like the second easiest SCC, and a lot of its weaknesses as an SCC are easily fixed with dips.  Dip summoner for Ramuh, and Squire for Gained JP Up.  Alternatively, you can dip MP restoring.  If you focus Time Mage, you get all of Teleport, Short Chage, and Meteor.  The spillover JP effect is real, too (going Time Mage might get your whole party Teleport).

Which makes Summoner #3.  It's a disgusting dip.  It's not a bad focus, either, but it's not that much of an increase in power to go from having just Ramuh, to having Ramuh, Shiva, Ifrit, Titan, Leviathan, Moogle, Fairy.

I'm not sure Chemist should be this high (4th).  They're the easiest SCC, but they don't need or even gain that much from dips (Charge for guns...).  That said...I'm literally never unhappy to have a Chemist on the team.  The spillover JP effect is real; it doesn't necessarily get everyone Auto Potion, but it might get people Hi-Potion.  The big thing I like about Chemist compared to a lot of the options I currently have below it is that you never suck; it never feels like you're holding your party back.  Go to Ninja, have fun being a Thief for a bazilion fights.  Go to Samurai, have fun being underwhelming for ages.

Ninjas kill things fairly efficiently.  Just being a Ninja, with your JP in other classes spent on skills designed on boosting Ninja baseline abilities (Move+2 makes throw better, Concentrate makes Throw and Autos not miss) is quite good.  But yeah, need to be a Thief to get there.

Oracle also takes a long time to get rolling, but is a bit of a swiss army knife when the skillset is near maxed.

Samurai...Draw Out is arguably a stronger skillset than Yin Yang Magic, but you need to be bad for a really long time.  First you get to be a female monk/knight/thief to unlock the class.  Then you get this class with terrible stats/equips that demands about 3000 JP.  And once you do that you're still not super impressive until you get Teleport.

Lancer...Jump kills things, not as good as Ninja or Samurai kill things.  But it kills things.

Dancer...done this focus before in co-op builds.  Nameless Dance is legit, and Wiznaibus is really not bad if you pump PA and brave enough.  I might be biased cause it's a focus I've done, though.  That said, Dancer has really strong moves, and struggles in assassination missions, and there's lots of ways dips make assassination missions a joke.

Priest...I've been in quite a few parties with focus priests, and I'm usually not that impressed.  But like...it's still a mage class, still one of the easier SCCs.

Monk...I've seen a lot of people start going Monk, and then say "man, I need to decide what I'm doing with this character because they feel bad" and then leave Monk for greener pastures.

Wizard...why would you make a focus Wizard?  That's just weird.  Don't think I've ever really seen this, but it sounds...playable.  Dip for Ramuh to get around evasion when needed, then sure, pick up MAU, Countermagic, Flare.... 

Mediator...raising/lowering faith/brave as a party service.  Mimic Daravoning stuff.  Have gun.  Sounds...okish.

Geomancer...Another carrier class like wizard where it doesn't make a whole lot of sense to focus the class.  (You can just hit things hard with a sword and have a shield, and have lots of elemental tile coverage to get around evade).

Archer...mmm...possibly this should be above Geomancer as a focus?  (I actually don't remember the consensus on which SCC was easier, although a lot of SCC ease comes down to stuff like Counter Flood vs Arrow Guard as reactions, which...whatever, you probably get a better reaction).  Charge, at least, can be taken to gun classes.

Bard: Angel Song exists and is legit.  You can focus Bard to get Move+3; it's not better than Teleport, but it's useable if you happen to learn it.

Knight: I'm not sure about the ordering between Knight and Thief, but at least if you focus Knight you can do stuff like use Weapon Break on a gun.  (Don't focus Knight/Thief).

Squire: how exactly do you focus Squire (if you're not Ramza)?  Like...don't get me wrong, I've done co-op with people who insisted on learning the whole Basic Skill skillset before moving on, but that's like...600 JP, which is still dip range.  To spend any significant amount of time in Squire you'd have to stay in the class long after you have nothing to spend JP on, and...why?  It's like...Geomancer with 20 subtracted from most of its mults.

Mime: I'm not sure if you'll have Mime unlocked before the end of the game.  If you do...enjoy having no equipment with the pretty close to average HP mult, 5% front evade, inability to set any RSM.  This might be more impressive than someone who has stuck in vanilla squire the whole game but...I'm not actually sure about that.  Squire can be an imitation Geomancer and use Twist Headband and Power Sleeve.  Squire can hit things with Swords, which are really good weapons.  I guess if you were really trying to make Squire work, maybe you'd get Two Hands or Two Swords and bring it back to Squire; maybe you'd get Teleport and bring it back to Squire.  Maybe you'd get Autopotion and bring it back to Squire.  Mimes can't set any of these things.  They can Mime, but I'm not sure that's better than having Two Swords, Auto Potion, Teleport, and equipment.
« Last Edit: June 15, 2017, 08:22:28 AM by metroid composite »

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #739 on: June 15, 2017, 09:22:00 PM »
So...thinking about it, I should probably lower Ninja on the dip list.  Yes, highest PA of classes that have real gear.  Yes, highest speed.  As a carrier these are theoretically good.

But it feels really optional for every focus that might go there.  After mastering Punch Art, do you go to Ninja?  I mean...you can, but you can also just stick in Monk and set Attack Up.  After mastering Lancer do you go to Ninja?  I mean, in many cases not if it makes you faster than your enemies and unable to jump on them.  After mastering Agrias, do you go Ninja?  You can, but in order for it to really keep up with knight swords you need scorpion tail.  I've also gone to Ninja with Dancer (max wiznaibus damage) and math skill (equip H bag, use Equip Shield, and be a speed status machine).  These are cute, but not entirely serious builds.

By contrast, Move+2, there are meta builds that just do not skip dipping this (like...focus Ninjas, not to be confused with dip Ninjas--two tier list confusionnnnn).  For the dip tier list, probably move Ninja below Archer and Thief.  Probably still above Knight and Geo.  (Knight Geo and Ninja are all sort-of in the same boat of "well you can use these to suplement other focuses; not a big deal if you don't, though!"  Ninja just the strongest of the three).

Speaking of things you can dip this but not a big deal if you don't...Mediator.  Yep, probably below Ninja somewhere.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #740 on: June 16, 2017, 04:32:35 PM »
Was talking to Laggy and he was a little confused on this, so I will try to define better what the focus tier list is.

To be clear, I have an efficiency tier list, where Samurai is a bottom 5 class because they take a lot of training and are not Calculator, and Knight is moderately high on the list cause Weapon Guard.  I know how to make that teir list, I've made it before, not the kind of tier list I'm making today.

In D&D especially 3e and 3.5e there was a tier list measuring the relative power of classes.  DMs would use this to figure out how to balance encounters, or whether there would be power imbalances in the group leaving some members feeling inept.  Cause people will just pop in and say "I'm playing a fighter".  In 3.5 the balance was so extreme the DM might need to say "no, you won't be able to keep up with the party".

Co-op in vanilla FFT is kind-of similar.  It's vanilla FFT, it's not exactly hard, so people build whatever they want.  Obviously there's no DM that's scaling difficulty, but if you've only got 3 players doing co-op you probably won't want a Bard, for example, and might try to talk that player into doing a different build.

It's not an SCC, you still set RSM and secondaries, and you usually pick decent budget options.  Within reason, though--Laggy was saying "well, if I was forced to play Priest for some reason, I would first get Magic Attack Up and Ramuh, and then stick those on Priest and largely ignore white magic."  And well...both of these are low enough JP cost to grab quickly sure.  But...if I was doing co-op with, say, 3 PCs and trying to figure out how to balance the party, it's not what I'd expect if someone said "I'm making a Priest".

That said, while I'm not sure it "counts" for being a priest focus for this tier list, and I guess I need to define what counts better, I'm also kind-of curious how good it is, cause Priest has issues like "awkward speed for casting Ramuh", so I've started a quick vanilla playthrough of a bunch of middle-of-the-road classes that I'm not sure how to sequence ("MAU Ramuh 'what is white magic' Priest", Monk, Dancer, Lancer).  Still in Chapter 1 unlocking classes (and getting enough JP for MAU).  Just finished sand rat cellar.

The Dancer and Lancer have been in Thief unlocking classes for most of the time, dealing 20 melee damage.  Monk has been dealing 30-36 melee damage, and still doesn't have the JP for a relevant Monk skill (maybe by Chapter 2...).  Priest (been in Wizard the whole time) with Bolt1 saving for MAU has been dealing usually 50-60 AoE range and completely carrying the party, but that's to be expected of running a Wizard in Chapter 1--at some point she'll be done grinding for MAU and forced to actually be priest.

In theory the Knight into Monk has been my second strongest character due to more damage (although...1 less move, probably slightly less HP than Thieves).  But in practice, right now everyone's getting carried by a Chapter 1 Wizard so the distinction of best no-skillset melee character doesn't feel too important.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #741 on: June 17, 2017, 05:28:11 PM »
Finished Chapter 1--one reset on Fort Zeakden.

Priest is now a Priest with MAU and Ramuh--Silk Robe gives enough MP for two Ramuhs.  While certainly still my best PC right now this setup is...really not impressive on a Priest yet.  She has 5 MA.  MAU raises this to 6 MA.  Ramuh dealt 67 damage to Algus.  Wizard with Bolt1 and no MAU can match that.

Monk got Wave Fist in time for Wiegraf to get around counters...Wave Fist made its debut by missing a lot.  The Monk is genuinely feeling like the second best character he's supposed to be at this point, though.  Dancer feels equally irrelevant in all classes (20 damage in Theif, 21 damage in Monk, 24 damage in Knight...).  Lancer is a Lancer and...can't deal damage yet, but did get to bait one spell and then Jump to avoid it.  Having Potion and Phoenix Down makes no character feel like dead weight, though.

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #742 on: June 17, 2017, 07:17:04 PM »
Quote
Archer...mmm...possibly this should be above Geomancer as a focus?  (I actually don't remember the consensus on which SCC was easier, although a lot of SCC ease comes down to stuff like Counter Flood vs Arrow Guard as reactions, which...whatever, you probably get a better reaction).  Charge, at least, can be taken to gun classes.

Geomancer is considered quite a lot easier since they deal way more damage and still have the option to kite at range like archers do (their range isn't as good as archers, but... not as much worse as you might expect). There was some debate as to whether Geomancer or Monk was the second easiest physical-job SCC (ninja being the easiest); Samurai would be in that debate too if you considered them physical. Not counting stuff like Chemist obviously.

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

DragonKnight Zero

  • Denizen
  • *
  • Posts: 518
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #743 on: June 17, 2017, 08:10:49 PM »
I've yet to write anything down, anywhere, but I have thought about how much JP do I need to invest in a skillset before it no longer feels like dead weight/waste of a skillset slot.  Which is highly subjective and probably beyond the scope of this topic.

Anyways, scattered thoughts and questions;

- How much JP will one earn over the course of a game?  Not counting spillover JP here; there's no guarantee it will be in the classes you want. 

- What about JP investments to unlock classes?  Angel Song and Nameless Dance are both useful tools and only cost 100 JP but unlocking their classes takes investing over 1000 JP (after factoring in the 100-199 JP every unit starts with in every class).  1100-1200 JP is a more likely minimum commitment since the investment is unlikely to be perfectly distributed.

- Crystals: is there in impact?  Uncooperative teammates are a thing, I guess.

- How many classes and/or setups can use a skill well?  Ramuh is awesome and 200 JP isn't too far out of the way though needs good MA and enough MP to make use of it so mainly the five main mage classes and maybe Geomancer will get mileage from it.  Without high PA, Punch Art come out like a cherry tap and some will argue Martial Arts is needed as well.  At the other end, Item with Potion and Phoenix Down can contribute to sandbagging at worst and Time Magic for Haste and Slow can be viably wielded by many classes.  These are just examples.

- Golem seems like a handy Summoner skill to have access to.  Or is it?

- Shield using setups like Abondon sometimes, even if it only shines in chapter 4.

- Magic DefendUp is handy at Velius and a few other select places.  I guess offensive supports are more favored?

- Everyone benefits from Maintenance in the three fights with Mighty Sword using opponents.  Though Math Skillers can escape without it at Rofel 2 due to AI manipulation.

- Pray Faith is one of those skills I never seem to want to spend JP on yet it has its uses with mage allies.

- Mustadio really likes Equip Gun though anyone else will likely have more desirable training options.  I feel Mimic Daravon is the real draw of the Talk Skill set.  Threaten is fun but seems to be only situationally useful and Invitation's low hit rate is something I only put up with if I'm looking to grab gear.

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #744 on: June 17, 2017, 08:39:54 PM »
Mimic Daravon is one of those skills I've gotten mileage out of on the SCC but I have a hard time respecting normally. 40+MA just isn't good enough. Yin Yang delivers the same effect with the same horizontal radius (less vertical) at far greater accuracy. Additionally, inaccurate skills like Mimic Daravon and Steal Heart have the problem where you lose out on considerable amounts of JP by going for them with any frequency.

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #745 on: June 17, 2017, 11:49:26 PM »
End of Chapter 2

Monk now has Earth Slash, Wave Fist, and Chakra.  Earth Slash is nice cause it has actual range and doesn't miss, but still feels trashy at 48 damage.  Been using Power Wrist just to hit 8PA, but the 3 move sucks (he doesn't quite have enough JP for Move+1).  Notably his being in Monk did give the Dancer HP Restore from spillover, so that's nice.

Lancers are good in Chapter 2--200 HP, and deal about 100 damage (starts the chapter at 72).  I screwed up though, and got Vertical Jump 8 instead of Level Jump 8...so...tainted experiment I guess?  Lancer carried a decent amount of the party weight this Chapter; Priest would assassinate key targets at 6 range with MAU Ramuh and then run out of MP, Lancer would clean up.

Priest has been solid.  100-130 damage with Ramuh usually.  Only enough MP for two Ramuhs (and maybe one Cure)--I only got Chakra at the very end of the chapter, so three Ramuhs was largely out of the question until then.  Since she's faster than the enemies, I just embraced it and went Green Beret, usually wait on spot first turn; does mean that Archers can get big charges on her.  Got Holy just in time for Queklain which was funny.  In order to hit 9 MA (so that MAU would take it to 12) had to use Wizard Mantle, which...meant 3 move; not a big deal thanks to her range, but makes it harder to set up big AoEs.  She usually hit 1-2 people with a summon.

Dancer was okish, since most of the chapter was Geomancer (at least it has range, more than the Monk could say for most of the Chapter) and Lancer (decent in Chapter 2).  Damage wasn't hardcore tied to PA like monk, so she felt free to use battle boots.  Damage in Geo generally lower than Monk--like 15 with Elemental and 40-50 with her auto (but she makes up for it in mobility/range).  Damage in Lancer was higher than Monk if she jumped (81 in theory; Monk was typically 60 with punches and wave fist).  Got actually to Dancer at Gate of Lionel, and Nameless Dance is a good dance, like really good.  (This power spike probably makes her the best PC for non-assassinations now; MAU Holy being the best for assassinations).
« Last Edit: June 17, 2017, 11:52:54 PM by metroid composite »

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #746 on: June 18, 2017, 05:13:51 AM »
End of Chapter 3

Dancer was dominant for a brief period in the first part of the chapter when the party didn't have much damage.  And then...well it's not that the Dancer got any worse, but other characters got much more damage, lowering the value of status, and there were a number of assassination missions where Dancer was not valuable.  The Dancer was fine in the second half of the chapter (Nameless Dance is good) but just fine, instead of standout.

Lancer...by the end, using an Equip Spear Monk with a Bracer had 14 PA and dealt 231 damage.  Generally two-shots most enemies from 8 range, sometimes one-shots.

Priest...between a Monk that now had Chakra, and just...better gear with more MP, and shorter fights cause everyone's damage was going up could generally just spam Ramuh every turn instead of saving MP.  Holy mattered in the assassination missions, which is like a third of Chapter 3.  Damage...generally 160ish.  Range, AoE, much faster charging than Jump.

Monk...was barely relevant for most of the Chapter (Earth Slash was aroud 75-80 damage for most of the chapter) and then spiked up hard after Yardow, when I picked up Attack Up and a Bracer simultaneously.  Suddenly with 13 PA and attack up...Earth Slash hits for 150, basic physicals hit for 223.  In my brief period outside of Monk, I found out that my Monk has bad compatibility with most of my team, and Revive had a 65% hitrate, so I went back to Attack Up Monk with phoenix down, rather than heading to Ninja.  Note: one less PA than the Lancer because Monk has bad PA growth.
« Last Edit: June 18, 2017, 05:37:04 AM by metroid composite »

Dark Holy Elf

  • DL
  • Denizen
  • *
  • Posts: 8161
  • Well-behaved women seldom make history
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #747 on: June 18, 2017, 05:49:19 AM »
I think Dancer is pretty overrated on the focus list at the moment honestly. Like... you spend an awful lot of time being a sub-optimum physical fighter in order to get a solid randomslayer who falls flat against bosses. That's an awfully large percentage of fights to be mediocre in just to be good (but not Calculator- or Summoner-level amazing) in the rest.

I mean I've done multiplayer fights with both focus dancer and focus priest allies and there's not really mich question in my mind the priest is a better contributor to the team overall. (They also have a significantly easier SCC, etc.)

Erwin Schrödinger will kill you like a cat in a box.
Maybe.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #748 on: June 18, 2017, 08:01:17 PM »
Well...my thinking with Dancer is that you master everything you want in Dance, and then you take it...wherever you feel like really, but probably something with damage.  But this is pretty vague, so I just went with a really JP expensive build on her so that I could mostly watch the Dance aspect, and watch the other characters. (So I took her to Samurai...and then in early Chapter 4 when I realized I basically wasn't using Draw Out over Dance, Dancing Time Mage for Teleport, then back to Samurai).  Dance itself actually picked up value in Chapter 4, with a lot of non-assassination fights (Sluice, Balk, walls, etc).  No assassination missions till Limberry, and Nameless Dance won a bunch of fights till then.    Sequence of assassinations after Limberry where I just left her in Samurai, and she was fine, functional, could deal about 200 damage or so.  And she basically trained in Samurai the rest of the chapter (got all the relevant skills out of Samurai by like...UBS5, and still switched back to Nameless Dance for Balk).

Monk was the real loser in Chapter 4.  Eventually his PA went up to 14 (189 damage Earth Slashes with Attack Up).  But it didn't matter cause maybe 20% of the Chapter was Earth Slashable.  Just...Earth Clothes on enemies through Germinas Peak, a whole bunch of floating enemies including zodiacs.  He Earth Slashed some Knights, wave fisted some charging enemies, but mostly he was the phoenix down water boy, used Chakra too.  It is worth noting, though, he was the one unit in my party that could really take advantage of hitting charging enemies.  Also, credit where credit is due, my whole team ended up with HP restore, and it was solid, and he ended up with Hammedo, which straight up won some fights.

I wasn't sure where I was taking the Lancer Long-term.  I was thinking Monk with Equip Spear, since I already had some Monk skills from a crystal, and this would give access to some (bad) revival.  Laggy convinced me to go Ninja, though.  Which...turned out fine.  My concerns of being too fast didn't really materialize--by the time I actually unlocked Ninja, enemies were speed 8, and the Ninja was speed 8.  When the Ninja hit speed 9 a few fights later, enemies were mostly speed 9 (Flash Hat enemies at the church, zodiacs).  So...it was mostly all upside.  PA was 16 with Power Sleeve Twist, Bracer, so 288 damage with Jump.  Final dungeon also had me switching Twist Headband for Thief Hat, so having that option was nice.  (Worth mentioning that sigificant weapon upgrades exist through Poaching.  Scorpion Tail rare poach from Hyudras is 368, allows support abilities that aren't equip lance, and improves your auto by letting you two swords.  Holy Lance common poach from Sacreds is 336.  Dragon Whisker rare poach from red dragons is 408.  Javellin II is 720).  I will note that Jump's max vert actually kinda was an issue; I even had vert8, which I never planned to get, and it just...annoying.  I also definitely felt the lack of revival, although I dunno if Monk with Jump would really be better.

Priest...Ramuh damage largely didn't move (still maybe 170 or so on average) but for much of the chapter all my long range damage was charge based (or dancer who has better things to do like Nameless Dance, or Monk who is crying in the corner because everyone is wearing Earth Clothes).  That said, the harder fights for this party were honestly probably the assassination fights, and Holy was party best damage...when useable (the 6ctr charge time was a significant limiter).  Also, the healing and revival from white magic was relevant.  A lot of these "final form" builds didn't have revival (X class with Dance or Wizard with Draw Out, Ninja with Jump) there were enough slow fights (Balk, which I took slow, for instance, Altima) where just healing was relevant.


Resets from Chapter 2-4 (note, I opted against doing cheeze equipment, so no Chameleon Robe vs Wiegraf, no Rubber shoes in Lionel, no Flame Shields or Ice Shields vs Balk).

Gate of Lionel: 1 (didn't set Nameless Dance)
Wiegraf3: 2 (one where I went in with Lancer--which I realized wouldn't work.  Switched Monk with Jump, and then had at least one reset where Lightning Stab crit and one-shot).
Wiegraf3 Resets that don't count: 3 (forgetting to set jump after job change >_>)
Velius: 1 (Needed to retreat and Jump with Ramza rather than jumping in spot.  Did not help that my Dancer was basically irrelevant--100 damage Koutetsu that she struggled to get in range to even use)
South Wall of Bethla: 1 (Doesn't count--forgot to buy Phoenix Downs)
Adramelk: 1 (Bahamut'd.  Was not in a good position from Dycedarg cause Ramza kept getting one-shotted before he could move during the Dycedarg part)
Zalbag: 1 (locked on Holy without checking, got hit while charging)
Altima: 2 (Ow demons hurt.  Probably my fault for setting up too much for form 2, and having two 10-11 speed people who can't do much to 9 speed Altima.  Altima2 just wasn't a problem anyway; two people with status curing and healing who can keep Alma alive, means that Alma basically solos the fight).

OK, I'll do a pass on updating the tiers in the next post.

metroid composite

  • m_ACac
  • Administrator
  • Denizen
  • *
  • Posts: 4380
    • View Profile
Re: Theorycrafting! (Because I like competitive metagames too much)
« Reply #749 on: June 18, 2017, 10:00:43 PM »
So...I think I'm going to be upgrading mages in general, and here's the thinking.  What's the purpose of picking things near the top of the tier list?  Well...let's say you're doing a co-op with 3 people, and then the first person says "I'm mastering battle skill and taking it to Archer" and the second person says "I'm going Bard", and you're just thinking "really y'all?  Really?"  You pick something towards the top of the tier list to carry their asses.

So...yes, Priest moves up.  Oracle moves above Ninja.  (Oracle is perfectly functional with a Summon dip, and you definitely use the Oracle skillset--Spell Absorb in particular deals with MP problems that come with summon spam.  Silence Song wins some fights.  Life Drain is good if you've been hit, and kills zodiacs.  Also getting your whole team Life Drain from spillover.  Sleep locks down fights).  Wizard should move up as well (Wizard is really a weird choice to focus since there's not much you want to spend JP on, but you can be a god for the first two chapters with just Black Magic and Item, and just dip for something that ignores evade by the time enemies start wearing Ageis Shields).  Geomancer....probably not gaining the same kind of bump as Wizard--at least with Wizard, Wizard with Item is a god tier setup for nearly half the game.  With Geo, sure, you could dip for Ramuh and take it to Geo, but Geo has less MA and less MP than Priest, and obviously Geo like Wizard doesn't have much you want to spend JP on.  I get the appeal of "I want things to do after I run out of MP", but I'm not sure Geo stands that much above, say, Mediator for that (guns outdamage elemental for most of the game, and Talk Skill, while worse than Yin Yang, does indeed not cost MP).

Monk doesn't move much; was...sometimes the second best character, but only when other characters were just failing; never felt like it was carrying the party.

Samurai moves down.  Certainly below Lancer, Lancer impressed me more at most points in the game.  Below Dancer?  Hmm...possibly.  I took my Dancer to Samurai and used that to grind out Samurai JP, but I could have made my Dancer a much more helpful unit much earlier in the game by going to like...Ninja with Dance to be relevant in assassination missions, or Priest with Dance to be relevant in assassination missions and also have revival.  That said, Samurai, when all the pieces finally come together (3000 JP in Samurai, 600 JP for Teleport, 400 JP for MAU, switch to Wizard), is very strong.  Samurai will be less of an early drag than Dancer cause you can do it on Ramza.  You unlock  it earlier, and it's pretty quickly playable when you unlock it (84 damage Koutetsu or something like that).  But it's melee, with 3 move, no HP, so it's...kinda on the bad end of playable.
I guess...there is the option of dip Teleport, go to Samurai (kind-of what I did).  Or Dip Ramuh, go to Samurai (gives you a range Koutetsu--probably only enough MP for one using Wizard Robe and helmet.  Black Robe...isn't guaranteed to get you two Ramuhs, either; needs like level 20-25 depending on MP growths.  Also less damage than Wizard Robe even with Ramuh, and obviously with Draw Outs cause Samurai MA is low).  Mmm...ok, dipping Ramuh doesn't sound that great; dipping Teleport before bothering with the physical job tree sounds like an ok plan, though.  That makes Samurai not terrible when you unlock them.  Teleport + ~100 damage Koutetsu is playable, not terrible in Chapter 3.  Teleport+162 damage Muramasa is playable, not terrible in Chapter 4.  Both kinda sub-par, though.  Although given that you already have Teleport, you have the option to switch out of Samurai for a hard fight and carry the team, and switch back to finish your training later.  Yeah, ok, leaving Samurai above Dancer.

Monk Mediator Geo?  Probably still Monk at the top of these; neither Geo nor Mediator really make that much sense as a focus.

New tentative focus list:

1   ca
2   tm
3   su
4   or
5   ch
6   wi
7   ni
8   pr
9   la
10   sa
11   da
12   mo
13   me
14   ge
15   ar
16   ba
17   kn
18   th
19   sq
20   mi

I'm not entirely sure how to order Chemist, Wizard, Ninja, Priest.  Wizard makes no sense as a focus (don't you want real spells what?) but probably just kills everything until enemies start using Ageis Shields, then picks up any summon and goes back to killing everything.

Ninja makes fine sense as a focus, but I'm not sure how much they can carry a bunch of party members who are screwing around in bad classes.  (Ninja has to suck for a while when in Thief/Archer, and even once they're done sucking, they're still a glass cannon that needs to get close to deal OHKO damage, and still entirely singletarget...a quick calculation, at level 30 they deal like...286 damage with Concentrate, Thief Hat, Power Sleeve, Angel Ring, Spell Edgex2, which...is not guaranteed OHKO territory.  Dropping Concentrate for Martial Arts is also not guaranteed OHKO territory).  Ninjas are really really good at pulling their own weight, but not sure how much they can compensate for a weak team.

Chemist is where it is there partially for the "giving everyone spillover Chemist JP is really strong."  Priest is like...a bad mage, but still a mage and never has to be weak at any point, unlike Ninja, and can access multitarget, unlike Ninja.