RE: Analyses EV Spreads

Eyan

sleep is the cousin of death
is a Top Tutor Alumnusis a Site Content Manager Alumnusis a Social Media Contributor Alumnusis a Forum Moderator Alumnusis a Community Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis a Battle Simulator Moderator Alumnus
This was cross posted from uu c&c hence the uu specific examples.
TDP (and other GPTLs) said it was fine to post this here, so don't kill me.


It seems as though that quite a number of writers (and even some QCers) aren't particularly familiar with how EV spreads should be like when writing analyses, so I'll just lay some things out for easy reference. Feel free to lay it on me if I'm wrong about something, and I'll edit accordingly:

1) Speed creep:

When deciding how much Speed investment you want on the analysis spread, keep in mind the simple guideline that you'd only Speed creep up to uninvested or fully invested Pokemon. EV spreads that are meant to outspeed other Pokemon with custom EV spreads shouldn't be on the actual set. At most, they should just get a mention in Set Details.

Example:
  • Timid Nidoking with max Speed hits 295 (just running full investment; nothing wrong with that)
  • Jolly Gliscor, esp SD sets, can run 176 Speed EVs to hit 297, outspeeding Nidoking (nothing wrong either, since it's EVed to outspeed a max investment Pokemon)
  • Timid Tentacruel can run 136 Speed EVs to also hit 297 (same as above)
  • Timid Tentacruel can run 140 Speed EVs to hit 298, outspeeding Gliscor that Speed creep Nidoking (this isn't allowed, so you're better off reverting it to 136 EVs)
The issue with using EV Spreads that creep other Pokemon that are already Speed creeping is that nothing technically stops them from investing that extra 4 EVs themselves; Gliscor can invest to hit 299 to outspeed 298 Speed Tentacruel, while Tentacruel can hit 300 to outspeed 299 Speed Gliscor...you get the idea. It's fine to invest to outspeed max Speed Pokemon (they don't have to be +Speed nature for the record), since it's not possible to have a back and forth like the example when they're literally already at max Speed). As for outspeeding uninvested Pokemon, that's just the baseline, but we don't go any higher, as the same logic would apply.

Lastly, for Pokemon like Reuniclus that would have an excess 4 EVs (assuming it's a standard 252 HP / 252 Def), it's fine to put them into Speed, although it's usually better to put them somewhere else to avoid confusion regarding Speed creeping Pokemon of the same base Speed.

2) HP numbers (tbh these should apply to teambuilding as a whole, not just analysis writing):

a. "Stealth Rock numbers":
An odd HP number does NOT mean it's automatically a Stealth Rock number, so don't operate under this assumption when crafting EV spreads for analyses (or even in your own teambuilder for that matter). Do some calculations and you'll realise that because of the way numbers in Pokemon round down, you're taking the exact same amount of damage from Stealth Rock but start out with less, which is objectively worse outside of rare cases where the numbers are actually accurate.​
Example:​
A Mantine spread of 248 HP / 252 Def / 8 SpD seems logical right? It's weak to Stealth Rock, so by lowering its HP stat by one, I'm ensuring it can survive one more switch in. Well, not quite, by doing some quick math...​
Damage taken from Stealth Rock with 252 HP EVs = 374 / 4 = 93.5 = 93 (rounded down; four switches into Stealth Rock with 2 HP remaining)​
Damage taken from Stealth Rock with 248 HP EVs = 373/4 = 93.25 = 93 (rounded down; four switches into Stealth Rock with 1 HP remaining)​
so why would you not just run 252 HP / 252 Def / 4 SpD? Writing the 248 HP spread in an analysis would just mislead readers.​
Example of an actual use of a Stealth Rock number:​
A Scizor spread with 248 HP. Doing the same math...​
Damage taken from Stealth Rock with 252 HP EVs = 344/8 = 43 (goes down on the eighth switch into Stealth Rock)​
Damage taken from Stealth Rock with 248 HP EVs = 343/8 = 42.875 = 42 (rounded down; eight switches into Stealth Rock with 7 HP remaining)​
while this may not be used often in practice, it's actually objectively helpful compared to the Mantine example, so it's acceptable in an analysis.​
b. Leftovers numbers:
There's little to no reason to run Leftovers numbers for HP (divisible by 16 / divisible by 16+1) in most situations. It's way too circumstantial to ever help, and the general logic of it is overrated. You're focusing on getting exactly 1/16 of its HP back when the raw HP gain is unchanged, unless of course you actually gain 1 HP by putting 4 extra EVs into HP. This operates on the premise or never taking any non-passive damage, which is unrealistic. If the EVs don't allow the Pokemon to survive any specific hits or the like, then just run max HP (not applicable for Pokemon that don't typically invest a ton in bulk). This is doubly so for EV spreads on analyses, as it'd show a more tangible increase in overall bulk.​
The main exceptions to this are any Pokemon running Substitute or a Berry. In these cases, you'd look into whether the HP number needs to be divisible by four to decide on the spread to use in the analysis. For example, Substitute + Belly Drum Kommo-o with Salac Berry would want to run an HP number divisible by four so that using Substitute and Belly Drum would activate the Berry. Thus, you'd use 4 HP EVs to achieve that. For something that's running Substitute with Leftovers, it's reasonable to run a HP number divisible by 16 so you gain back the HP used to create a Substitute in four turns instead of five. Finally, Belly Drum Azumarill with Sitrus Berry would want its HP to be exactly divisible by two so the Sitrus Berry gets activated after setting up.​
3) Jump Points:

For this, I'll leave this from the GP standards
A special mention goes to "jump point", which often isn't really relevant information for in-battle scenarios (if your benchmark allows you to hit some significant KOs, just get to the point and state those KOs) and on its own tends to be a pretty arbitrary benchmark to begin with.
Essentially, if the jump point does something other than just being a jump point, state it. Otherwise, don't bother with them. "Increasing overall Defense more" is pretty flimsy in an analysis. In some cases, it can even be detrimental; for example, you lower a Pokemon's SpD investment and invest into Def to hit a jump point "for the sake of it" even though it doesn't change what physical hits it can take, yet you're sacrificing its ability to take special hits, which is what it's supposed to.
 

McGrrr

Facetious
is a Contributor Alumnus
This needs a bump because people have obviously not read it. Maybe it is too verbose, so let me summarise:

Residual damage rounds down

Belly Drum: 1/2 (even HP activates Sitrus)
Shed Tail: 1/2 (odd HP enables 2 passes, even HP activates Sitrus)
Substitute: 1/4 (multiple of 4 +1 enables 4 consecutive subs, multiple of 16 +1 with Leftovers enables 5 consecutive subs)
Rocky Helmet: 1/6 (avoid multiples of 6 for Pokemon with contact moves)
Poison/Stealth Rock/Spikes: 1/8 (avoid multiples of 8, adjust accordingly for Rock resistance/weakness)

These are the main factors to consider, but obviously the same logic can be extended to Burn, Life Orb, Salt Cure etc.

Mitigants:

50% recovery moves: 4 additional EVs for an even HP is worth up to 9 HP per battle (after SV PP nerf)
Leftovers: usually not worthwhile to hit a multiple of 16 unless the Pokemon also has a 50% recovery move

Example sub-optimal HP EV spreads

https://www.smogon.com/dex/sv/pokemon/skeledirge/

411 vs 412 HP makes no difference to residual damage, especially with Heavy-Duty Boots covering for 1/4 from Stealth Rock. The suggested 248 HP EVs costs up to 9 HP per battle due to Slack Off

https://www.smogon.com/dex/sv/pokemon/iron-treads/

252 HP EVs on Iron Treads is at least somewhat understandable with Leftovers, but with Assault Vest?! 384 divides by everything relevant; it is even divisible by 32 so Stealth Rock does 12 instead of 11

Rule of thumb

Single defensive bulk is optimised when HP = Def or HP = SpD
Mixed defensive bulk is optimised when HP = Def + SpD

Use this to allocate spare EVs. HP and Speed are the only stats where 4 EVs = 1 absolute point, so 4 HP EVs should be preferred (subject to the above residual damage caveats) to 4 Def/SpD EVs because it covers both sides, while the latter will not always impact a damage roll.

Examples of 252/252 Pokemon where I would put 4 spare EVs into HP: Iron Treads, Flutter Mane, Iron Bundle, Chien-Pao, Breloom, Iron Moth etc.

Dragonite is a curious example where you can go either way: 4 HP EVs benefits Roost, but 324 divides by 4 and 6, which is sub-optimal if Heavy-Duty Boots gets Knocked Off. Earthquake is non-contact and divisible by 4 no longer matters after Tera Normal, so I would take the HP point, but 4 Def has merit because it does impact many common damage rolls.
 
Last edited:

Users Who Are Viewing This Thread (Users: 1, Guests: 0)

Top