DL Overview Topic #2: Status Rebalancing

[Dhyerwolf]

I’ve decided to enshrine my insanity with another DL Overview topic! This one is a bit more comprehensive than topic 1 (limit scaling). I’ve now come to the point where I think that it’s worth exploring reshaping how we look at status to better match its in game use. There are a few ways I would look at changing how we do status:

1. The most critical aspect is that instead of scaling status against enemies that can be hit, I will start taking status against all randoms (immune, resistant or not). This will better reflect statuses’ worth in game, since who cares about your 100% ID when it hits 20% of enemies (SH 3 Hilda is a great example here). We’ll actually get a measure of how often you can use the status in game instead of crowding it into a little box. I will discuss various other facets below.

2. Lower the threshold of status hit rate to 50% instead of 67%. Simple enough: We’ve scaled in more resistance, it's fitting to take a more weighed view of when status hits.

3. What to do about status blockers? I’m…still allowing these. I would be more much likely to enforce penalties for using them (which I don’t now), but there are aspects about the penalties I don’t like (Too uneven. It rewards a game having shit accessories by giving it a lower penalty. If a game only had status blocking accessories, there isn’t even a penalty (or worse, a game has other accessories but not ones that can be quantified in a statistical manner like good regen or counter accessories). I would say that if anyone wants to equip a status blocking accessory, there should just be standard penalties across all games).  I’d keep them regardless, because status is still more potent in the DL in general (it’s nice that Aika can force one person into Confusion in game, but it doesn’t allow her to tie things down like she gets to in the DL sometimes). Status in game is still often not as good as the numbers would imply (Ie: A Sleep spell that gets canceled by your other PCs attacking, a Death spell that functions to kill in a different manner when all your other PCs are attacking (and thereby detracting from damage and potentially leading to a non-optimal allocation of damage (unless the instant death is first or last).

4. What to do about bosses? Do we apply this to bosses? I personally will not be doing this for a few reasons. The first is that I now use Tal’s views on bosses (when I remember. I…often don’t! Tal, vote more so I can remember!), which is that I don’t need to scale bosses down on HP because they face clones of PCs equal to the number of PCs they face in game. This already is a pretty large change, and serves to make forms of ST status much less dangerous. The second reason is that PC equipment can vary from game to game. Should Wugui’s ID be thrown out because everyone could equip a Leonardo’s Bear (or…it is worth it to see multiple PCs die on turn 1 if you didn’t and have no way to protect them other than restarting?) It’s a tricky question, and I’m very open to hearing other opinions.

Now that we have the basics out of the way, we can start delving into scaling. There are a number of points that I believe are extremely relevant to scaling
1 . Time frame of enemies to consider: In games like LSSC, you go through a large chunk of the game with status being an excellent tool…until everything in the endgame starts nulling everything. There should be a point that reflects games like this. My proposal: For any spell/move/weapon:
--If gained in first half of game, it’s taken against the later half of relevant* enemies  it faces (we’ll talk about relevant enemies in another point).
--If gained between 50% and 75% of the game, it’s taken against the final 25% of enemies OR half the time you can use the spell/weapon with a minimum of the last 25% of final enemies.
--If gained after 75% of the game, it’s taken against all remaining enemies
--If gained “aftergame”, it’s taken against the final dungeon
2. What is a relevant enemy? Gut reaction: If you can OHKO an enemy with free damage, the enemy is not relevant (Example would be Hilda again. There are enemies that are ID-able but have 150 HP. Basically everyone can OHKO these enemies for free. I don’t think they should be counted)
3. Prorations/Additions- What, this already not enough work? Apparently, my answer was no because I realized that there was a category of relevant information that does not come up with just numbers. Much of this has to do with your ability to tell when status will actually have a point. Does the game give you tools to tell that you’re status is actually worth using in a certain situation, or is it just blind luck? Blind luck doesn’t make status numerically worse, but it will lead to a lot of useless turns because you’ll need to constantly assess (AND REMEMBER) when status is good or not. I’m going to list what I came up with:
--Weaknesses revealed by Scan
--Scanned data is easily accessible after the first time
--Proportion of enemies that can be hit (What’s better, 50% status on everything in the game or 100% status on 50% of the enemies
--Low hit rate (Simple enough. A 30% hit rate on a pure status spell is bad because you’ll need multiple attempts at first on average to even tell if an enemy is susceptible). I’m thinking that I would tier by either 20% demarcations (so 0-20, 21-40, 41-60, 61-80 and 81-100) or maybe 25%.
--Multi-target (quicker and more effective determination of susceptibility)
--Does the game differentiate between “Miss” and “Immune”
--Is the status a latent effect of something else? Can you contribute to your allies rushing with damage at the same time?
--Does your status get counteracted by things your allies will do? (IE: Sleep that is ruined by an ally hitting an enemy)
--Spammability- If you can only use status rarely, then it’s not worth so much in terms of being reliable (especially if you generally will get it in the DL)

I have not yet mathed out a way to show these. I’m…not adverse to raising effective hit rates for scenarios when games make it damn easy to tell status worth.

With all that said and done, I’ve started putting together analysis of games. For some games, I am noting where I am approximating midgame and the 75% break point. Please note that most of these are based on FAQs and residual memories. I’m judging break points by level (So L20 is halfway on a game with an endgame of L40. I realize that EXP curves are rarely smooth, but it hurts much less mentally to standardize this when possible)

I’ve listed relevant status attacks and approximately what is appears their new hit rate might be. The first batch of games will be in the second post.

[Dhyerwolf]

GAMES THAT ARE DONE: Breath of Death 7, FF 6, FF 7, FF 9, Legend of Dragoon, Lost Odyssey, Shadow Hearts 2, Shadow Hearts 3 and Wild Arms 4

Breath of Death 7
Status is halved. Note that everything in the final dungeon immunes, and everything in the area before the final dungeon is susceptible (which is a pretty long walk). So I’m cutting everything right down the middle. Lita’s status comes earlier, but given that she’s relying on a money skill DLwise that is like L24 out of 32, I have little issue simplifying and taking everyone at the last 25%.
Lita- Had 80% Sleep, Blind and Silence and 40% ID originally->All halved (Sleep/Death combo becomes turn 2 instead of 1)
Sara- 100% Stun->50% Stun (No real loss)
Erik- 100% Poison->50% Poison (No real loss)

FF 6
If gained earlier in World of Balance, I’m taking all of WoR
If gained early WoR or late WoB, I’m taking Owzer, Phoenix Cave, Umaro’s Cave, Cyan’s Dream, Fanatic’s Tower, Ancient Castle and Ebot’s Rock (I’m calling all Mog late WoBish for sanity)
If gained late WoR, I’m taking the last 3 of the ones I listed above
For any move with a greater than 100% Hit Rate…randoms appear to be MEvade-less, for the most p. Correct me if I’m wrong, but seems like it’s almost completely irrelevant to have a hit rate of 100% versus 150% (against randoms). Similarly, I’m not sure that “ignoring resistance” is something of use.

CELES- Muddle- 94% Confuse-->53% (L32)
Imp- 100% Imp-->15% (L13)
Berserk- 150% Berserk-->55% (and resistant to going any lower) (L40)
CYAN’s Cleave- 125% ID-->57% ID
EDGAR’s Air Anchor- 100% ID-->53% ID (Cultists Tower)
Noiseblaster- 100% Confuse-->53% Confuse  (South Figaro. Near start)
-->Of all the DL statuses, Noiseblaster rightfully has the reputation for being one of the best lock-down statuses. It’s not unreasonable to see this as a form of ID (aka: it works once, the enemy loses barring weird circumstances).
GAU’s Roulette- 100%-->55% (Can’t be lowered by a statistic)
LOCKE’s Wing Edge- 20% ID-->11% ID (Odin’s Castle/Phoenix Cave)
MOG- Cockatrice- 70% Stone-->~20% Stone
Cave In- 100% chance of 75% Gravity and Seizure-->55% Chance of Gravity +70% Seizure
Snare- 75% ID-->41% ID
Earth Blues- 100% chance of 62.5% Gravity+Seizure-->55% Chance of Gravity+70% Seizure
Love Sonata- Spector- 120% Cfonuse-->55% Confuse
SETZER-Doomed Darts -20% ID-->11% ID (Odin’s Castle)
SHADOW’s Assassin- 20% ID-->11% ID (Cave to sealed gate)
Stunner- 25%  Stop-->17% chance(Fanatic’s Tower)
STRAGO- Sour Mouth- 100% Poison/Sleep/Imp/Mute/Muddle..VARIOUS (Daryl’s Tomb)->50%/37%/16%/42%/50
Condemned- 100% Condemned-->55% (Sealed Gate)
TERRA’s Break- 90% Stone-->24% (L68) (FINAL)

Final Fantasy 7
Remarkably consistent even going back pretty far. It almost doesn’t matter where you set it (I went back to Temple of the Ancients to confirm). Several dungeons just tend to spit on status in general (Temple of the Ancients, for example, has enemies that have at most 2 status susceptibilities)
Aeris’ Seal Evil->100% Silence and Stop->30% Stop and 10%! Silence
Barret’s Hammer Blow- 100%->30%
Cait Sith’s Toad->80%->33%
Cait Sith’s Mini->90%->37%
Cloud’s CrossSlash->100%->42%
Cloud’s Finish Touch- 100%->30%
Vincent’s Hellmasker’s Nightmare- 100% Poison, Frog, Sleep, Dark and Silence to 25%/42%/32%/20%/10% respectively (Still good as a comprehensive combo attack )

Final Fantasy 9- Basically doesn’t matter when you take it. Berserk and Confuse are very heavily immuned and the rest are generally not immuned.

Eiko Silence (Start asically): 46%->41%
Eiko Mini (D3): 46%->41%
Freya Luna- Trident (Desert Palace, Oielvert): 100%->41%
Garnet Odin- Dark Matter (Treno Auction House Disk 3): 40%->36%
Garnet Berserk – Magic Racket (Disk 3 beginning) 65%->27%!
Garnet Blind (Beginning Disk 2): 65%->60%
Garnet Silence (Beginning Disk 2): 40%->36%
Garnet Mini (Beginning Disk 3): 40%->36%
Garnet Confuse (Fossil Roo): 40%->10%
Quina Bad Breath- Mt Gulug: 28% to 7% on Confuse, 28% to 25% for Darkness, Poison, Slow, Mini
Quina Night- Fossil Roo: 91%->82%
Quina’s Frost: 18%->16%
Steiner Iai Strike- Rune Blade (Daguerro): 16%->14%
Steiner’s Breaks- I don’t even know if randoms resists these. I would need to test.
Vivi Slow- Ice Staff (Beginning Disk 2): 49%->44%
Vivi Sleep- Flame Staff (Early Disk 2): 39%->35%
Vivi Stop- Oak Staff (Mid Disk 2): 29%->26%
Vivi Break- Cypress Pile (Disk 3 Beginning): 29%->26%
Vivi Death- Black Hood (Esto Gaza): 34%->31%
Zidane’s Soul Blade (Pinnacle Rocks): 100%->90%

Legend of Dragoon
Everything taken from the endless Desert except Kongol’s Stun and ID. You could argue Rose’s spells going back further, but sanity/it doesn’t matter a lot given that LoD doesn’t ramp up resistance at the end (I think).
Rose’s ID (L3 Dragoon Spell)->85% Accuracy
Rose’s Fear (L2 Dragoon Spell)->75% Accuracy
Status Weapons (Dart’s Confusion (Kachua Glacier), Albert’s Fear (Deningrad), Kongol’s Stun (Zenebatos), Meru’s Stun (Vellwebb), Miranda’s Confusion (Fueno))->25%->19% for all minus Kongol’s Stun (15%)
ID Weapons (Kongol (Moon) and Haschel (Snowfield)->25% Acc->19% for Kongol and 21% for Haschel

LO
Taken disk 3 and on for the most part. Most statuses are cut by 70%, Blind by 50%
Jansen’s ID (Northern cape, LATE), Slow Petrify (L19), Curse (Crimson Forest), Seal (L19) , Sleep (10), Paral (White boa or shop) and Poison (10)->ID is 100% to 10%, the rest are 100% to 30%
Cooke’s Blind (43)->100% to 50% (Only 50% chance of Blind having an impact each turn. I wouldn’t let two 50% chances proc on the first chance)
Mack’s Fear (L25)->100% to 30%
Ming/Sarah’s= Jansen and Cooke spells

Shadow Hearts 2
Wine Cellar close enough to the beginning to give Grand Slam all of Asia. Neam Ruins is about 30% of the way in (Foreigner’s Cemetary and on). Idar Flamme is half (Mt Fuji and on)
Joachim Grand Slam (After Wine Cellar, L10ish?)- 19% (Turn 1->Turn 4)
Anastasia’s Euthansia (Neam Ruins)- Becomes 30% chance of Poison/Paral/Seal/Petrify. 51% Chance of something fatal (and Silence is pretty solid too)
Nightmare (Doll House) and Direct Press (Idar Flamme) are 60% (Turn  1->Turn 2)

Shadow Hearts 3
Hilda joins before Edna, and has this attack for about 70% of the game. The remaining 35% is basically Uyuni and on.
Hilda’s Grand Slam (Start)->100% to 20% (Turn 1 to turn 4)

WA 4
Arnaud’s Sleep- 50% to 24% (L47. Go back to Valley of Oblivion to reflect that he could get it notably earlier if he so choose)
Arnaud’s ID-33%  (12.5%x3)->14% (5%x3) (Gained final dungeon)
Raquel’s Poison

[Dhyerwolf]

Extra space for expansion of games.

[Dhyerwolf]

Extra #2!

[Dhyerwolf]

More

[Dhyerwolf]

So, the first set of games is finally up. I wanted to note for basic purposes, who the big winners and losers are so far. Theoretically, losing more than 25% of the accuracy on your status generally makes you a loser and less than generally makes you a winner.

BIG Winners- None yet!
FF 9: Technically most of the cast, minus that upon reviewing more thorough notes, I decided on a different enemy group that made the status hit rates worse anyways. They didn’t gain from what I had previous to this endeavor).

BIG Losses
(Numerical Summaries: 8 1->2, 1 1->3, 1 1->4, 2 2->3, 1 2->4, 1 3->5) (Means that 8 PCs had turn 1 status and fell to turn two)
Numerical Summary of Which Divisions are Effected (Borderliners increase each division by 0.5)
Heavy- 4.5 Duellers
Middle- 6 Duellers
Light- 2.5 Duellers
BoD 7: Lita- Sleep-ID combo Turn 1 to Turn 2 (Middle)
FF 6: Edgar- ARGUABLE: Some may see his Confuse as not guaranteeing lock down (I’m okay with it) (Heavy)
FF 6: Mog- ID goes from turn 2 to turn 3 (Middle/Heavy)
FF 7: Aeris- Stop/Silence goes to Attempt 2 for Stop and Attempt LOL for Silence (Light)
FF 7: Barret- ID limit goes from turn 1 to Turn 2 (Middle)
FF 7: Cait Sith- Mini and Toad go from Turn 1 to Turn 2 (Heavy)
FF 7: Cloud- Paralysis and Instant Death Limits goes from Turn 1 to Turn 2, (Heavy)
FF 9: Freya- Luna Turn 1 to Turn 2 (Middle)
LO: Cooke- Blind goes from 100% to 50% (And only has a 50% proc rate when it hits. I would say this pushes the effective status out a bit) (Light)
LO: Jansen- All status from 100% to 30% (Turn 1 fatal Stone combo->Turn 2) (Middle)
LO: Mack- Fear Turn 1 to Turn 2 (Light/Middle)
SH 2: Joachim- Grand Slam Turn 2 to Turn 4 (Middle)
SH 3: Hilda- Grand Slam Turn 1 to Turn 4 (Heavy)
WA 4: Arnaud- Sleep Turn 2 to Turn 3, ID Turn  3 to turn 5 (Heavy)

So, what games are next:
ATL 2, BoF 2, BoF 3, BoF 4, FF 4, FF X (Potentially), FF 13, PS 4 (Potentially), SO 3, Suikoden 2, WA:ACF, Xenosaga

[Dhyerwolf]

Oh yes, any help would certainly be very appreciated (for example, you know a FaQ or something similar to certain games that would have the information I want). This goes doubly true on helping determine when a status spell is gained.

Speaking of, I'm nearly done with SMRPG, but I had some logistical questions in determining "relevant enemies." It seems to me that based on pure HP count, a lot of enemies in the final dungeon are OHKOable by near anyone. Am I correct in this? And is this often true of low HP enemies in the dungeons preceding the final dungeon.

[Dark Holy Elf]

For SMRPG, I don't trust my memories enough to comment on enemy durability (with my most recent playthrough being SLRPG). But... my boss stat topic includes PC damage against average at many points in the game, so you should be able to compare enemy HP in previous dungeons to those figures.

With regards to, say, Edgar, the fact that his Confuse is 100% against vulnerable enemies is pretty important, and does mean you can lock them down. While I'm fine with baking in an "immune" rate, I sorta feel that once he has passed that check, every subsequent Confuse should hit too. Really, what your interp is doing is given a certain % of the DL immunity to a skill (to represent the struggles the move faced in-game) so that check should probably only be made once.

On a similar note, you asked this question, so I'll answer it: 50% status that works on all randoms is significantly worse than 100% status that only works on half the randoms in a game. If you use both blindly, sure, they're equal, but as soon as you learn which enemies are which the 100%-but-runs-into-immunity status is way better. (Also, if status in a game tends to be at or near 100%, it becomes easy to learn immunities that way!) I'm not sure how to address this.

Also on a related note to, say, the scan which shows enemy status vulnerabilities, there are cases like PS4 where 95+% of the time you can guess an enemy's immunities based on their species. I would tend to have a more favourable view of statuses like that, so Spark is reliable against mechs and Elim is reliable against organic enemies, even though if we took a strict "% of enemies they hit" they'd both look much worse.

I may add more comments later.

For now, some help for your documentation diving:

DQ8: randoms' status properties are found in the mechanics guide.

-FF1: there are mechanics guides which list the elemental (i.e. status) properties of every random.
-FF5: status immunites can be found in the BMG, as can MEvade and Level, which also influence status odds.
-FFX: Split Infinity's guide includes status susceptibility for all randoms
-FF13: randoms' status susceptibility is documented on the FF wiki. I think Tal already accounted for it to some extent, but your methodology may yield different results.

-PS4: the mechanics guide includes enemy elemental properties, which against includes status susceptibility. FreshFeeling's site includes enemy Strength/Mental/Agi stats, which additionally determine hit rate. Note that currently the PS4 status is taken against enemies with above average stats (although no status resistance).

-Saga Frontier status immunities are documented... but it'll take quite a bit of diving to figure them out! You'll need a guide which lists enemy equipment (one of the monster lists has this IIRC, but make sure it's one for the monsters themselves, and not for PC monster forms...), and then the mechanics guide lists the immunities found on said equipment.

-The VP bestiary which the DL created lists which enemies are immune to status. (All bosses are, as well, except where noted.)

[Dhyerwolf]

On a similar note, you asked this question, so I'll answer it: 50% status that works on all randoms is significantly worse than 100% status that only works on half the randoms in a game. If you use both blindly, sure, they're equal, but as soon as you learn which enemies are which the 100%-but-runs-into-immunity status is way better. (Also, if status in a game tends to be at or near 100%, it becomes easy to learn immunities that way!) I'm not sure how to address this.

Also on a related note to, say, the scan which shows enemy status vulnerabilities, there are cases like PS4 where 95+% of the time you can guess an enemy's immunities based on their species. I would tend to have a more favourable view of statuses like that, so Spark is reliable against mechs and Elim is reliable against organic enemies, even though if we took a strict "% of enemies they hit" they'd both look much worse.

I guess the flip is that the 50% move can be applied to everything (so needs more time, but usable on more enemies). So...it's a tricky question and may often depend on the game (which is true of every status). It could be that either one could more more useful depending on the game, so I'm not even thinking too much about that. Of course, based on one of my addition/proration notes, 100% would get an extra potential boost over the 50%.

In terms of PS 4, it's so clear that I agree that I wouldn't be giving any penalty for that type (I'll just be looking at normal immunity types). On the other hand, for Xenosaga, I would be much more likely to factor it in (but I think that has more to do with all the ways to inflict statuses quickly generally being limited since they all need to be equipped and equipment/tech slots are very limited).

I think agree on Edgar's Confuse, although I'm not 100%. Since the whole theory is based around trying to better reflect use/value, given someone a bonus for having that great type of strategy feels very valid to me (it really helps that he gets it super early, so it packs more of a punch in use).

Edited to add: It looks like only 7 VP enemies are noted as status immune. Is this correct? Would sure make it easy.

[Tide]

I don't really vote anymore unless there's a match I'm interested in personally, but I'll just drop my 2 cents on this. I'll also note that this isn't even necessarily criticism just in case it comes off sounding as too negative.

To me, I don't feel much value there is in this entire exercise. The main issue stands from how intuitive it is (or lack there of) and how much additional work would be needed to document future games/numbers. It is much easier to understand that say Celes has an 150% chance of inflicting Berserk versus going, "well she has a 150% chance in the first 2/3 of the game but 0% chance in the last 1/3 so she only has 50% accuracy with it". In the former case, I can easily say "Celes will probably hit with Berserk", but in the latter case, I am assuming there is a good chance of it missing, which isn't reflective of how the spells works in-game anyway. Also, I feel like I'm asking myself where I would see her fighting a certain character. Is it at the end of the game, where I know the spell will guarantee miss? Or is it near the beginning where it is a guaranteed hit? I know that's not necessarily how this should be interpreted but the point I wanted to address is that I don't think it is easy to use at all.

Then there are new games. While there are barely any new stat topics nowadays and the main DL is dead, how would you propose this data be tracked by people gathering the data? It's a lot of work getting the accurate status rate as is (especially if its not documented anywhere since the person has to do the tests themselves) and what you're proposing is adding even more. I think it will be more stifling to the chances of new topics coming up.

There are other questions too. Like, we have people commenting about how much they 'respect' move X and Y. How would you handle that? I feel that this method is supposed to address this (as the thing with Hilda shows) but I'm sure people will still have their personal biases and double dip. In some senses, it's already factored in. Sure, Hilda's 100% ID might be the most blatant of cases, but Gate for example doesn't give much respect to Arnaud's Sleep and already holds this against him if he has to rely on that to win.

Also, assuming you do penalize people for using Status Blockers. How would this work exactly? If Beowulf only has a 45% chance of landing Confusion, would I penalize the opponent for using a Status blocking accessory just to guarantee his victory? Or would I pro-rate the penalty as well? Speaking of Beowulf, how would you take characters like him? You can theoretically get him early and never use him, which stifles his JP anyway and thus he gets late access to his abilities if at all. Or you can get him late, grind a bit so the very first story battle you do, it's all already available. Further, I also saw you didn't take into account that Raquel's poison bite does run into resistance. Against bosses, it tends to only do 2500 damage instead of 25% like normal, which by the way is really crap. 2500 damage at the end of Wild ARMS 4 is basically garbage tier damage. However, are you proposing that because it is not resisted, we just treat this at 25% and give the bosses special treatment?

Altogether, I think that while this is a neat exercise, I don't think it is really a workable method for me. At least for me personally. The DL was never perfect at reflecting everything as is in-game (sup Ramza). There will always be some people who get hosed and some who get more credit than they should. I'm perfectly okay with this because how people interpret certain things and what standards they use is part of the fun of imagining fantasy style typed matches. All this really seems to do is redefine how people get screwed

Apologies if this came out sounding really negative. But since you addressed "we" in your first post, I thought I would just drop some thoughts down. I doubt I will be using this method at all, even if everyone else adopts this convention.

PS: I'm going to second NEB regarding the 50% but works all the time versus 100% but only works for the half the game thing. The latter case is superior because you can quickly tell whether or not it works, and anytime you do use it, it basically isn't a gamble. Consider if you used a Scan ability and saw the information. In the former case, even if you know it's vulnerable, you still only have a 50% chance of having the status hit so you're still gambling. In the other case, you can quickly re-adjust your strategy. Without a scan type move, the 50% move gets infinitely worse because sometimes you can't even differentiate if the move simply missed or it is immuned. Some games don't give you this distinction. The only case where the former is a better option is during speedruns.

PPS: You laugh at the enemies with piddly HP in Shadow Hearts 3, but those enemies are often the worst because they appear in packs . And packs of enemies are what kills you more so than the HP tank enemies. Also, not everyone kills them in one hit the way I play so Hilda's ID could be relevant (no, not really)!

[Dark Holy Elf]

Re Dhyer's question about VP:

That sounds right. Should be noted that in general this is based on observed behaviour to stun and freeze, the only statuses I've used extensively. But from what I remember of a playthrough where I used Janus a lot, poison lines up that way too (petrify is the one I've used only rarely so it's possible it could be different. And silence, but lol VP silence). VP just... didn't seem to make enemies status immune, outside the boss palette-swaps.

I guess the flip is that the 50% move can be applied to everything

It's true, but this is inherently less valuable. The reason is because, against any enemy who gets hit by a status 100% of the time, the PC with this status has a great option. Against everything else, they just try damage, which might well be more useful than 50% status odds anyway (depends on enemy HP).

To illustrate this, try taking the numbers further. Celes' Imp is listed as 15%. However, against those 15% of enemies, if you know who they are, it is really useful. A hypothetical version of Imp which no random was immune to but only hit 15% of the time would be a worthless spell; why try for a status which is unlikely to hit in the first four tries when you could just pile on damage? You never would, at least in randoms.

[Dhyerwolf]

A lot of this is where prorations/additions start kicking in. It's difficult to balance. Rightfully, Imp should be a blistering accurate attack, but should only affect 15% of non-boss enemies. So seeing it as just a blistering accurate attack that is rarely protected against the DL doesn't feel all that balanced to me (or reflective of in game use). I guess using Hilda again is a better example (solely because Celes isn't relying on Imp for duelling worth due to having a lot of other statuses that 85% of the enemy don't laugh at). Once you see exactly how few enemies Grand Slam can hit, it seems way offbalance to give Hilda that much credit. I would rather see it as 20% accuracy so it always has a chance, because 100% accuracy but only hitting 20% of enemies isn't something we can really translate.

Of course, under prorations/addition metrics, Hilda would get a boost because Grand Slam is also damage, meaning that it's better since Hilda can contribute to getting rid of enemies in the traditional way even if they are ID immune.

So maybe the key is to start developing the metrics for prorations/additions. Maybe first for base accuracy bonus/penalties. While I could split everything up into groups of 20% (0-20%...etc), that doesn't feel very nuanced, but it's by far the cleanest.

90-100%- +20%
80-90%- +15%
70-80%- +10%
60-70%- +5%
40-60%- No Change (I figure that everything in this game is in the vicinity of a 50/50 chance enough to make it feel relatively similar)
30-40- -5%
20-30- -10%
10-20- -15%
0-10- -20%

Should the effect be more potent? I'm not personally sure. Maybe skew it to make the effect size increase as you head towards the extremes. Or maybe I should skew it more towards bonuses.

[SnowFire]

While the DL isn't in-game, I do agree the DL should make respect calls that make the DL at least resemble in-game.  It's not terrible when the DL rejiggers the overall effectiveness a little.  If it puts it off a LOT, though, then that's a problem.  Notably, below a certain amount of accuracy, I'd be inclined to just plain discount the move as not working unless there's some kind of explicit vulnerability to it. 

So examples:
* If there's a flavor reason behind a status, then honor the flavor reason even if it ubers or nerfs a skill in the DL, e.g. Elf's PS4 examples.  (Also obvious with non-status examples like weakness-hitters like human-slaying on FF6 Man Eater or Symphonia weapons which aren't a big deal in-game but way better in the DL.  Or something like Jude's extra damage to nanomachines, or FE8 Sacred Twins extra damage to Monsters, that both come up practically never in the DL but quite commonly in-game.)

* Try to aim respect checks to make a status that doesn't have a solid flavor grounding for its effectiveness to make it hit close to a percentage of the DL that it does in-game.  This is the most obvious with "status equivalency" checks on blockers.  If Celes's Imp really has such a horrible hit-rate, that'd make me very inclined to consider practically any vaguely related form of status blocking as stopping it, to reflect that this is a commonly immuned status.  (This can get a LITTLE weird as maybe PCs have explicit problems blocking the same status in-game, and this is supposed to reflect PCs, right?).  Flip side, if a status is very good in-game, maybe be a bit more of a hardass about needing to directly block that status and not a kinda similar one.

* I agree with the view that if a status is either 100% or 0% in-game, it should ideally be 100 or 0% in the DL.  So, related to the above, Hilda's ID should be 100%...  when it's seen as working.  This is gonna get real vague, but if it's super-commonly-immuned, then basically rule that it doesn't work by default, and needs an explicit vulnerability to do so.  Defending game doesn't have ID?  Doesn't work.  Does have ID and DL-illegal ID blockers exist?  Doesn't work.  Defending game has ID, other characters can block it, but this particular character -can't- block ID?  Okay Hilda's crappy ID works, and 100% of the tim.  (For an example, perhaps a SMT game where one character has no Light/Dark resistance when other characters do, say.)


As a side note, "relevant enemies" is tricky, because this is often code for "do this game's randoms matter."  Justin's Bonesplitter Axe is great but non-essential since if you lose to G1 randoms, you're doin' it wrong.  It seems a bit weird to punish a game's status attacks because it's too easy, but so it goes.  I'll take 40% status in a game with badass randoms over 100% status in a game where all randoms die on turn 1 to an MT magic barrage with cheap refillable MP.

[Dhyerwolf]

Just to note, ToS anti-Human weapons actually suck. Pyro tested them are they are like a 12.5% bonus or something absurd like that.

That balancing for Hilda's case is definitely an legitimately interesting one. Of course, the question is how to make a metric out of it so it can comprehensively cover status of more varying levels.

For relevant enemies, it generally means that enemies have super piss poor HP. Our classic example is Hilda: there are other enemies that Grand Slam can hit...but they'll be OHKOed by it's damage anyways. If your status is ST and basically anyone can kill the same enemy for free, there's no reason to use the status.

[Dark Holy Elf]

As a side note, "relevant enemies" is tricky, because this is often code for "do this game's randoms matter."  Justin's Bonesplitter Axe is great but non-essential since if you lose to G1 randoms, you're doin' it wrong.  It seems a bit weird to punish a game's status attacks because it's too easy, but so it goes.  I'll take 40% status in a game with badass randoms over 100% status in a game where all randoms die on turn 1 to an MT magic barrage with cheap refillable MP.

Well, I will note that Dhyer used enemy durability, not overall enemy challenge, as the gauge. As noted by Tide, SH3 randoms can absolutely wipe you... ID vulnerable ones, even! However, the reason that Grand Slam is bad is that most enemies immune it, and many of the enemies who ARE vulnerable are OHKOed by virtually anything (physicals from higher-Str PCs, L1 spells from higher-Mag PCs), so Grand Slam isn't actually uniquely useful at killing those enemies when everyone can do it for 16 MP or less. (And of course Grand Slam has a recharge time that physicals and L1 spells don't, so it's even worse at killing those enemies!)

Similarly I do recall Grandia randoms who couldn't be OHKOed with low-cost stuff, because in G1 at least the low-cost stuff is actually fairly weak. So yeah status could certainly be relevant there... hell, I got some milage out of Death Sword as a quick way to take out some of the more durable randoms in some dungeons. (You'll still never -lose- to these enemies, but IMHO that isn't the point.)

I will also say that I'm... not sure I've played a single game where I feel that 40% ST status not attached to damage is useful. The lowest examples I can think of where I got meaningful milage out of such a skill (XF and FFT petrify) are 50% minimum. Of course, if the status is attached to useful damage (e.g. Pokemon Ice Beam's 10%, for an obvious example) or is MT (so the % chance of at least one enemy being statused is higher), then yeah low odds can still be useful. Otherwise it'd pretty much have to be used in a game against very durable enemies... 30% ST petrify would be worth a turn in significant sections of XS2, for instance.


Dhyer: For all my ruminating about the merits of Imp vs. 15% universally applicable status (and similar cases), I... really am not sure how I would factor it in. I do in principle like the idea of "status accuracy is more important than the number of enemies it works on, but both matter", but I'm not sure of how to factor it in. Your proposed pro-rating does feel like a decent idea, though.

[Tide]

I will say, I *have* gotten mileage out of status rates below 50% before. We're talking pretty bad odds like 20%. Do you know when I got the most mileage out of them? Speedruns >_>. Like I said earlier, you won't take unnecessary risks or gamble on bad odds normally, but in a speedrun, it's a different story. The risk and rewards shift to the amount of time saved for using such a gamble versus playing the most consistent/safe way. For example here's the stuff I have definitely used:

Shadow Hearts 3: Poison 1 on Jeb Nigas
Wild ARMS 3: Grav on Nidhogg
Wild ARMS ACF: Graviton on the Dual Mantis x2 boss and is also an option on Elizabeth

Basically, I don't see how status odds that are only work 50% of the time to be better than those work 100% but only hit half the enemies for DL purposes since the DL rewards consistency and not gamble options. I mean think of how respected Rafa and Malak are. Even if they dealt 2HKO damage with their abilities, you would still probably be iffy on using them. Flip side, if you knew that their abilities would always focus but were only effective on half the enemies due to Elemental properties, it would be really good.

[Dark Holy Elf]

To be clear, sub-50% status can still be useful against bosses, because the assumption is that bosses take way more hits. It's pretty reasonable for Black Mage on Fiesta to spam 10% petrify against Neo Exdeath Almagest part, for instance. Against randoms, though, I'm typically more skeptical.

[Dhyerwolf]

I think before digging into too many more games, I'm going to try and math out a proration/addition scale that feels sensible overall.

Having just obviously played Suikoden 1, I would say that a comparison between Tir's ST and Hilda's ST ID certainly serve as a good overall example of why I think this scaling makes so much sense. Tir can ruin any-non boss in the game and gets great use of even ST ID, while Hilda struggles to apply it to nearly anything (let alone anything that isn't killed in 1 hit anyways).

[Dark Holy Elf]

Agree with that.

Thinking about Barret's match today also made me wonder: how best to deal with OPB status effects? Like, I certainly got good milage out of Cloud's Cross-Slash paralysis, but since he can only activate it once per fight and it's apparently just below 50%, it could end up useless? Well, not totally since if there are other unpredictable elements to the fight, then the chance of paralysis could push him over the edge. (e.g. if he's facing someone with 30% evade, then his chance of paralysis is good enough that we can assume his opponent will never dodge, since the combined chance of even one dodge and Cross-Slash not working would be under 50%) I dunno.

[Dhyerwolf]

It would technically. I will note that at least for my limit scalings, FF 7 L1's could be used twice in a DL match sometimes. Granted, it's going to be pretty rare and given his damage, Cloud likely won't be winning or losing on turn 2 Paralysis. I'm also trying to compile a list of games where characters defend status based on some kind of consistent stat. There could be a few characters that would be weak enough to make Cross Slash turn 1 (which would reflect in game well!).

[Nephrite]

I don't know if it matters, but from my perspective, if I try a status and it doesn't work on something, I'm really not ever liable to try to use that status again "just to see if it might work this time!" I mean, Final Fantasy 5 bosses are pretty notorious for having status holes -- but I don't think your average person is going to sit and try and figure out which statuses work on them and instead just blow them up with magic or whatever.

I realize bosses aren't exactly fair, but I didn't go around trying out Imp on every random in FF6 to see if it might work. I don't know that I ever had any reason to cast the spell in the first place. I'm not really sure if that makes any significant difference, but the same probably applies to FF4 for me too, which had like 9 different statuses, none of which I feel like I ever went out of my way to actually test.

I think ultimately I just threw a bunch of words together and didn't really say anything at all here, but I guess my point is that the status rates should probably reflect how good they are if you choose to use them, not how good they are if you know to use them.

[Dhyerwolf]

It took me quite a while before I could write on this topic and have it make sense! It sounds like compared to Elfboy, you would put different weights on different prorations or additions (He values base rate hit more, you value transparency more. FF 4 and FF 6 likely get the worst ratings under transparency, which is to say: many statuses, monsters immune individual statuses (as opposed to blanket immunity/resistance), you need to remember which statuses work on which enemies).

I would say that the whole re conceptualization is to make status reflect how good it is on a whole as opposed to just if you know exactly how to use it.

Anyways, I've at least come up with a few metrics (Starting with either 3 (Base Hit Rate, MT?, Damage) or 5 (Transparency and Spammability as well, although Spammability has been pretty constant in most games. I'm thinking of just altering it to a penalty if spammability is particularly bad). Is there any way to make tables on the forum? That would definitely be the cleanest way since I would like to put in full math notes (and not just final results), but I seem to remember that the only way was extremely difficult.

[Dark Holy Elf]

Hi	Dhyer
I am	a table.

Quote this post and you should be able to figure out the tags from there. What I usually do is make my tables in Excel/OpenOffice/whatever, then use a decent text editor which can actually search for tabs and returns to find replace tabs for [ \td ][ td ] and returns for [ \td ][ \tr ][ tr ][ td ]  (without the spaces), works pretty well.

[Dhyerwolf]

I gave up on the board spreadsheet. Too unwieldy. At worst, I'll put my the grand spreadsheet in google docs and link it.

I figured that before I finished that up, I should note the specifics of the proration/addition catergories so I can hopefully get more feedback.

1. MT/AoE- The most important facet of in game worth of a status.

I'm considering giving ST moves a slight penalty. If so, I'd be looking at 5%-10% likely. I'm not sure about this.

MT Attacks are taking as the % of hitting either enemy if you are attacking 2 enemies prorated to ¾ of the gained benefit (to reflect that sometimes you may miss the enemy you really want to hit with the status, which this masks a little bit). AoE attacks...I'm thinking between 1/4 to 1/2 the benefit of hitting 2 enemies

For true MT, an 80% status would go to 92%, a 50% would go to 68.75%, and a 10% would go to 16.75%.

All other categories are straight bonuses. So MT/AoE that can't break 100%, but it also has a much more potent effect. I could switch it so that none of the other categories could do that, but I would likely need to alter the formulas for balance (or not, since MT is still far more important for in game potency)

2. Base Rate against non-immune enemies- A measurement of how accurate the attack is against non-immune enemies. Base Rate would generally be seen as the old DL hit rate, although sometimes it might be a little lower because some level of enemy resistance (not immunity) is not factored in. For example, in FF X, a lot of enemies have 20%-50% resistance to some statuses, so the base rate wouldn't be 100% in this category.
90-100%- +20%
80-89.9%- +15%
70-79.9%- +10%
60-69.9%- +5%
40-59.9%- No Change (I figure that everything in this game is in the vicinity of a 50/50 chance enough to make it feel relatively similar)
30-39.9%- -5%
20-29.9-% -10%
10-19.9%- -15%
0-9.9%- -20%

3. Transparency- Base Rate and Transparency balancing are probably where you'll see the largest differences between formulations. For example, it sounds like NEB would put more focus on Base Rate while Neph would put more on transparency. Do you want a reliable status that you constantly have to be testing to see what you can use it against or do you want a less reliable status where you can easily remember when it might be useful?

Transparency measures how easily you can gauge and remember whether an enemy is suspectible. Note that Grand Cross style moves from games that get a negative penalty move up to 0%. Transparency has a larger maximum penalty than maximum increase because the last category can get a pretty bad (For example, FF 4 has 14 statuses, and if you exclude enemies that either immune all statuses or immune no statuses, the average enemy immunes 5.7 statuses (including enemies who immune 1 status and 2, 3, 4, 6, 9, 11, 12 or 13 statuses). Good luck remembering which enemy can be hit by what status as you run through the game).

+10%- You know what enemies will be hit because basically everything can be hit (ATL 2, Brig, Child of Light)
+5%- Not all enemies are hittable, but either resistance is pretty light or you can determine pretty who to use against (FF 9, MK 2, PS 4)
+-0%- Either it takes a little work to figure out or there is some light remembering involved. Games that groups status resistances into just a few categories and there aren't varying levels of resistance go here. Games with decent Scan spells go here. Games where it's not really relevant come here (SO 3 would be an example; status is a nice bonus, but it's basically always attached to damage and you really only can use 1 thing for damage) (BoD 7, FF X, SO 3)
-10%- A moderate amount of statuses that all hit separate checks or grouped statuses (ala Mind/Death/Status in BoF) that have many levels of resistance (BoF 3, BoF 4, SMRPG)
-20%- Too many statuses that all have separate checks and enemies have random immunities (FF 4, FF 6, FF 7...although due to the nature of limits, they might climb 1 category up potentially. Unsure there since there are limit options))

4. Damage- The bonus is 25% x % of damage you retain compared to your best damage attack (rounded down to nearest 2.5%). So Yang's Sleep gets a 25% bonus, while  Bowser's Terrorize gets a bonus of 7.5%

5. Spammability- Penalty only category. Does a PC suck at being able to spam a status attack? This can be major since it means that not only can you not utilize it most of the time while you could in the DL, but it also means that it's much harder to determine status weakpoints.
Can Use 10 Times in a Row +0%
Can use 8-9 times in a row- -5%
Can use 6-7 times in a row- -10%
5- -15%
4- -20%
Less than 4- -25%

PCs in games with storebought MP healing are generally exempt from the category if the restoration is worth anything. Attacks that are item dependent get the lowest tier.

6. Synergy- Completely undone. This will just be a penalty I believe. Does your status work particularly badly with whatever the rest of your team would be optimally doing. Examples:
Freya's Luna- Forces mages not using anti-Berserk to be use shitty physical attacks
Sleep/Confuse cancelled by damage- In a turn based game, this type of status might end up being cancelled out by an ally attacking later than you since you won't know whether the status will hit or not before you assign actions.

How much of a penalty? I'm thinking up to 25%, although I could see higher if it clashes that badly.

Questions I have:
Does anyone remember how good FF X, WA 4 and WA:ACF scans were in terms of revealing enemy status resistance data? Do you happen to remember any other games where a Scan move explicitly tells you status immunities or susceptibilities?
Does anyone remember any games that specifically differentiate between Miss/Immune (I believe FF 9 does, for example)?
Are there any categories that you feel should be less or more potent?

[Dark Holy Elf]

FFX scan explicitly lists status immunities (though not resistances). WA4's and ACF's do not.

FFX and FF13 both explicitly differentiate between immuned status and missed status. Pokemon does as well, for what little status immunity it has. Probably more. Most SRPGs (FFT, XF) let you differentiate immunity by way of the spell's hit rate projection being listed as 0%.


Other notes:
-I don't personally agree with any credit for MT/AoE, the DL is about duels for me. MT is just one of those things that doesn't get DL credit, much like things like Turn Shift. (Except against multitarget DL bosses of course.) The MT moves can shine in team fights, which are generally more of a thing than they used to be these days anyway!
-I don't agree with credit for the attack dealing damage. The PC already gets credit for that in the DL with... the damage they deal while using said attack. No need to exaggerate this further.
-I'd like to see some sort of extra discredit for status which requires you to set up for in advance (STATUS WEAPONS. Freya's Luna, as you mentioned). Because if you bring this setup into a fight with immunity in-game you've wasted your time. In principle you could vary the penalty with the setup required, so games where you can't change your weapons/relevant skills in-battle would be hit worse. I'm not sure.