Wednesday, April 19, 2017

Optimizing EVs Spreads in VGC: Part 1

Optimizing EVs Spreads in VGC: Part 1
(Oct 21, 2016)

Disclaimer: Before continuing on reading these tips regarding "optimizing" EV Spreads in any VGC format, I'm going to assume the reader knows the "basic knowledge" of Effort Values (EVs) / individual Values (IVs) and has a fair experience with EV Training. This guide is primarily based on making EV Spreads in VGC, meaning all Pokemon levels will be capped at Level 50, not at the usual Level 100 in standard Smogon play. Keep the "Level 50" cap in mind, stats distribution works far differently at Level 50 and most resources like Speed Tiers at NB uses Level 50 stats to assess the highest speed investment for Pokemon. Word of advice is not EV Spread is "perfect". Most Players might opt for the simple 4 / 252 / 252 spreads and don't have to make "over-complicated" spreads. Only use what spreads are deemed effective for the Pokemon and fits team composition as well. With all that said, let's get into Custom EV Spreads.

Preview: Why Use Custom Spreads?

Most beginners in the competitive VGC scene have seen these rather strange or abstract EV spreads from from RMTs, Pastebins, Team Reports, etc with some or nearly all of them not close to the "traditional" 4/252/252 set. What does this accomplish you ask? Why are they investing speed into a slow Pokemon? Why is there bulky investment ( either HP / Def / Sp. Def ) on this Pokemon? What's the purpose of not going full  252 Atk / Sp. Atk on a certain offensive Pokemon? These are all valid questions, sometimes answered in the Team Reports or RMTs, however new players might not grasp these intermediate concepts in one go.

Generally speaking, a good portion of Pokemon in most VGC formats have complex EV Spread to optimize their efficiency in battle. In terms of having "bulk" investment, note players want that Pokemon to stick around to keep offensive momentum, or support option knowing it can take a certain attack. In terms of offense, players might only invest enough to either OHKO, or 2HKO key benchmark targets while maintaining leftover EVs for others stats. For speed, maybe a certain benchmark investment is involved to be able to outspeed a target.

This guide will focus on how EVs Spreads are allocated, importance in Natures, and Speed/HP investments. Thought about having to work on the Offensive / Defensive side of optimizing, but since this will be a rather elongated guide, might as well split it up into parts.

Level 50 EV Spread Formulas and Notes

 Given how Singles-based battling (Smogon) is more popular than VGC Doubles, most entry-level players might be incline to use 4/252/252 spreads to hit as hard as possible, outspeed key speed tiers, or take as many physical / special attack as possible from either defensive stats. The problem is this specific EV Spread is geared towards Level 100 given how every 4 EVs yields +1 in any particular stat point. At Level 50, 8 EVs are used instead to increase 1 stat point, however there are some factors to consider with IVs of 31, 30, etc. Keep the following stat investment formula for level 50s in mind: (IV/2) + ( EV/8) = X. If general rule of thumb when using this if "X" is not a whole number, then stat investment isn't enough/efficient, or thus wasting EVs. 

While following a general rule of a specific investment divisible by 8 is reasonable, this isn't always the case. Note most IVs of of competitive Pokemon at Level 50 are going to be based at 31, sometimes resulting in a +1 stat increase if the player only invest +4 EVs. To explain this mechanic, essentially we use this formula above to show why this is the case. If a certain Pokemon has a value of 31 in one of its invested IVs, and only +4 EVs in a stat, here's what happens: 

(IV/2) + ( EV/8) = 31/2 + 4/8 = 15.5 + .5 = 20

Essentially the values even up to make a whole number, however what happens if the 4 EVs instead becomes 8 EVs? To keep the math short, the answer would be 16.5, or a non-whole number, meaning we are wasting Effort Values. Take into consideration the possibility of "investing" into a 30 IV value for Hidden Power in this scenario, players can use an investment divisible by 8 to get an efficient EV spread. Note these mechanics do work in a similar fashion with boosting EVs in a negative stat , but this isn't optimal for practical purposes (Note: SpA EVs in Adamant nature which decreases SpA growth). An example of this is raising the Atk EVs on an Modest Nature Pokemon like Machamp. This is wasteful especially if the a physical attacker with a high base attack of 130 like Machamp has strong physical attacks to utilize. Avoid "investing" in the negative natures, a particular minus stat of a Pokemon and focus on its highest stats. In short remember these notes:

  • To Determine an Investment isn't Wasting Effort Values, use the following formula: (IV/2) + ( EV/8) = X
  • If an stat has an IV of 31, the investment will be a multiple of 8 + 4 ; or 8m + 4 = EV Investment for 31 IV Stat
    • EX investments: 4, 12, 20, 28, 36, 44, 52, 60, 68, 76, 84 ..... 228, 236, 244, and 252. 
  • If a stat has an IV of 30, the investment will be a multiple of 8 only ; or 8m = EV Investment of 30 IV Stat
    • EX investments: 8, 16, 24, 32, 40, 48, 56, 64, 72, 80, ..... 224, 232, 240, and 248.

So remember when people said not to use 4/252/252 EV spreads? As long as the stats being invested all have a value of 31, then technically speaking we aren't really "wasting EVs". Thing is there are still plenty of factors to cover in the next section involving bulk investments, notable HP EVs. The basic 4/252/252 are "fine" for generic battling immediately, though chances are its not efficient enough as detailed later.

For people who want the full formulas to determine a fully invested stat, here's the following formulas and variables for Total HP and Stat for level 50 from this EV Training Guide from Smogon for BW, which is still applicable.
  • HP = BaseHP + 0.5 x IV + .125 x EV + 60
  • Stat = (BaseStat + 0.5 x IV + 0.125 x EV + 5) x Nature
  • Notes: Base = Base Stat Value ; Nature = either x0.9, x1.0, or x1.1. (Multipliers for Natures)
  • The final value for the stat is always rounded down. 
Oh wait...we have Pokemon Showdown's Teambuilder to do this for interactive use and most damage calculators. How foolish of me to have this section. Well the point of this is to showcase the math involved during finding the right EV spread investments and final stat enhancements for a specific Pokemon. However, keep on these notes of using EV spread investments in mind for the future sections. Also promise, this is all the math we'll do for now.

 Why Natures are Important? 

For those who don't already know, each Pokemon is assigned one of twenty-five different combinations of "Natures" which affect the growth of at least two of its base stats. Twenty of these natures have a "positive" of x1.1 for a certain stat, while another is negative, or decreased by x0.9. With all these variations, why is it most Pokemon have certain natures over others? The thing about Competitive Pokemon is players want to bring out the maximum potential, or effectiveness of their Pokemon without losing worth.

Recall the "special" Machamp example. With its base 130 Attack stat and access to powerful STABs / coverage moves such as Close Combat, Ice Punch, Bullet Punch, etc Machamps benefits in having a high Attack EVs to abuse its movepool. Sylveon is often known for being a Special Attacker, or spread damage dealer with access to Pixilate Hyper Voice. In order to increase the damage output for Hyper Voice, Sylveon must have a positive SpA Nature like Modest, not most of the other Natures. Note both Machamp and Sylveon have abymsal Sp. Atk and Atk respectively as players always invest in a Nature which "decreases" a dispensable stat without much opportunity cost.

Conversely, the general tip for most "Custom EV Spreads" is to invest into a Nature that "increases" the highest stat of a Pokemon. Why might ask? With a positive nature and EV investment, there will be certain placements when investing the Pokemon will increase its stat by 2 points instead of the usual 1 point. People have referred to this specific allocation as "the bump" generally speaking when investing into a positive nature. This simple trick might sound obvious to do at first, but the thing is some new, or inexperience players who generally don't know much about optimizing EVs. Take this Politoed for example (going to be using the Damage Calculator either from Showdown, or Nugget Bridge).

Politoed @ Leftovers
Ability: Drizzle
Level: 50
EVs: 212 HP / 44 SpA / 252 SpD
Modest Nature
IVs: 0 Atk
- Scald
- Ice Beam
- Encore
- Protect

Original Stats: HP 192 | Atk 72 | Def 95 | SpA 127 | SpDef 152 | Spe 90 
* 44+ SpA Politoed Ice Beam vs. 4 HP / 4 SpD Salamence: 172-204 (100.5 - 119.2%) -- guaranteed OHKO

This Politoed has enough investment to OHKO a 4 HP / 4 SpDef regular Salamence with Ice Beam and the rest is dumped into HP and Special Defense. When looking at Politoed's base stats, notice its Base Special Defense is 100 compared to the Base 90 Special Attack. From the first EV Spread, Politoed's Special Attack will be at a value of 127, and the Special Defense stat at 152 ( Note: These are the actual stats, not the Effort Values ). Given there's already 252 Sp. Def Evs in EV Spread, why not just make this Politoed Calm Nature as it increases the Special Defense, while lower the rather useless Attack stat. Maybe we can adjust the Special Attack investment from 44 SpA to 132 SpA to 1) be at the exact number of 127 before, and 2) to allocate other EVs if possible. The result is the following:

Politoed @ Leftovers
Ability: Drizzle
Level: 50
EVs: 212 HP / 4 Def / 132 SpA / 156 SpD / 4 Spe
Calm Nature
IVs: 0 Atk
- Scald
- Ice Beam
- Encore
- Protect

New Stats: HP 192 | Atk 72 | Def 96 (+1) | SpA 127 | SpDef 154 (+2) | Spe 91 (+1)
132 SpA Politoed Ice Beam vs. 4 HP / 4 SpD Salamence: 172-204 (100.5 - 119.2%) -- guaranteed OHKO

If anyone compares the two Politoed sets, the two different EV Spreads equate "about" the same exact stat totals, except now with a few leftover EVs to increase the a couple stats by one or two point at least. Notice the second Politoed has an two extra Special Defense points given around the 154 mark is where the "bump" occurs and nets the extra values. The leftover 8 Effort Values can be sorted into 4 points in Defense and Speed to gain a slight stat boost. This is what happens when Players invest into the Nature which increases the highest stat can result in salvaging a couple EVs here and there. Of course, there are obvious exceptions in this rule such as speed tiers (next section), and "specific" offensive/defensive benchmarks achieved for a purpose in competitive battles.

To end with some valuable info, here's are the "points" when the actual stat increases to +2, instead of +1 if this is a positive nature (exception being the HP Stat). These will start at around the 100+ range:
  • 108->110, 119->121, 130->132, 141->143, 152->154, 163->165, 174->176, 185->187, 196->198, 207->209, 218->220, 229->231, 240->242, 251->263, ... etc 
  • Think anyone can figure out the pattern...

 Speed Investment and Tiers

Speed is one of the most important factors in Double Battles given the offensive-oriented nature of the VGC format as the ability to strike or setup first can be advantageous. For most offensive incline Pokemon, usually they are measured with 252 Speed EVs in order to reach key "Speed Tier" benchmarks given the specific metagame. For defensive support users, sometimes speed isn't allocated, however Speed Creeping is done to at least outspeed the same Pokemon in or near a similar base speed tier or investment. Regardless of the reason, always consider the if there should be speed investment for any Pokemon first. Even if this is blandly obvious examples like max speed Jolly Garchomp, or 40 Spe EVs Sylveon, these will likely consume most, if not cap one of the stats (Spe) at around or close to 252 Effort Values of the 510 remaining. If the Speed EVs are going to use up most of the Effort Values cap allowed, then might as well ensure those Pokemon reach those specific benchmarks.

Going near or the full 252 Spe EVs investment might be necessary to allow the Pokemon to outspeed key targets without any worry. If players want to "maximize" or make efficient investments, always research, or look into a Speed Tier Guide / Damage Calculator to see how much does a Pokemon need. At the moment, Nugget Bridge's VGC 16 guide is the one of the best speed tier guides out there as it include key benchmarks, the actual speed stats, positive/neutral/negative Natures, and modifies (Tailwind, +1 Speed Boosts, etc). Consider reviewing the "Base Speeds" of all Pokemon, which are essentially detail how fast can a Pokemon be if at max, or min. Damage Calculators like the Smogon / NB versions can work, but this doesn't go as far in-depth as the guide, however still helpful in allocating Speed EVs.

For custom speed tierimg, always have a benchmark speed tier in place before dumping EVs. One point is enough for any Pokemon to outspeed its target following the game mechanics as long as their speed stat is greater than the others. Lets say we have a base 100 Speed Naive Salamence who wants to "outspeed" a base 91 Jolly Landorus-T for Level 50s. For this Naive Salamence to "outspeed" max speed Landorus-T, it will need at least 188 Spe Evs to reach an actual stat of 158.  Notice this is one point faster than Landorus-T's actual cap of 157, meaning this Naive Salamence always outspeeds bar any speed modifiers. Assuming one of the offensive stats capped at 252 EVs, this Salamence now has 68 EVs leftover to use for any stat to distribute equally!

When benchmark speed tiering key Pokemon, assume the opposing target will always have a positive Speed Nature like Jolly. Though there's a possibility the Pokemon won't have such a nature, always "assume" the Pokemon is max speed, unless proven wrong in actual battle. Players want to have a guaranteed to outspeed these specific targets, unless they are Scarfed or have other speed modifiers. If for whatever reason, both Pokemon speed stats are "tied", or exact value,  the result will be a 50/50 "Speed Tie" odds for either one to strike first, a rather even, yet unfavorable odds. Some players prefer to "risk" the speed tie when going for max at certain Speed Tiers, but generally speaking consider using Speed Control methods, or use "faster" Pokemon in these situations.

Regarding Speed Creeping itself, players have to know what Speed Evs people are running either on the ladder, or in tournaments for a respective Pokemon to figure out the speed investments. If there are any "leftover" EVs, or just paranoid in wanting to be the first to "outspeed" a target with the same base speed, then by all means go for it. However don't over commit to Speed EVs, as this will mess up any benchmarks made like Offensive, Defensive, or HP Values. The best way to speed creep is to know the trends of the metagame, know how much players are using in their Speed EVs, and ask yourself if this investment is a comfortable amount to stay at.

In short, of a Pokemon has a specific speed tier, or target Pokemon they need to outspeed, make sure when EV training it can. For furthe tips regarding Speed Investments, Tiers, and other advice, here's a list down below. 
  • Outspeeding Pokemon "naturally" won't be enough at times, so consider Speed Control methods as detailed here.
  • Keep in mind Choice Scarf will boost the Pokemon's speed by a multiplier of 1.5x.
    • Unless the Scarf user needs the "full" max speed, or benchmark, ensure that the resulting number is even and not a non-whole number. Thing is the game will always register say 185.5 as 185 unless it becomes 186. 
  • Speed Boosts, or multipliers tend to go by the following: +1 (1.5x), +2 (2.0x), +3 (2.5x), +4 (3.0x), +5 ( 3.5x), and +6 (4.0x).
  • Trick Room Speed Tiers are generally Pokemon using a Negative Speed Nature (Relaxed, Quiet, Sassy, Brave), but maybe with some "custom" IVs to underspeed key Pokemon in this field condition. 
    • Since most Trick Room Pokemon aren't investing in Speed at times, the EV Spreads reflect a more "Bulky Offense" as TR teams want to make use of every turn possible.

HP Optimization Tips

Aside the massive amount of tips discussing HP investment list down below, here's some advice for investing in HP. First off, don't think of adding HP as a "supplement", or a middle-ground of running any Defense / Special Defense effort values. Remember HP stands for Hit Points, so you are only increasing the amount to health to receive a hit, not take one decently well. Its rather best to minimize as much HP investment as possible and mix between moderate Def / Sp Def investment, unless of course a support or defensive Pokemon needs to be close to either physically, or specially defensive as possible. There are some exceptions to this rule as if a Pokemon's Defenses are naturally higher than the HP Stat itself, best given it near or max HP as possible. If a Pokemon has massive Base HP stat than the overall defense stats, allocate EVs in both stats.

Alright the following information below are useful pointers to consider when investing in HP EVs wisely without throwing a whole 252 in at once. Some of these benchmarks are to reduce the damage of residual damage, increase for a slight HP recovery, or achieve a Substitute number. Obviously players don't have to follow most, if any at all, however consider these values as they can sometimes determine the momentum of the battle. Some also have below the actual HP stats players should "achieve" when investing in HP for Level 50 for convenient sake. 

  • To reduce residual damage of Sandstorm/Hail, the actual HP stat should be 16n - 1, or one value less than a multiple of 16.
    • (Ex. Values of 16n - 1): ....239, 223, 207, 191, 175, 159, 143, 127, 111, 95, ... etc.
  • To reduce the residual damage of Burn, Leech Seed, regular Poison, etc, the actual HP Stat should be 8n - 1, or one value less than a multiple of 8.
    • (Ex. Values of 8n - 1): ... 239, 231, 223, 215, 207, 199, 191, 183, 175, 167, 159, ... etc. 
    • Some of these values like 239, 223, 207, 191, 175, 159, etc share the same optmization worth of taking less Sand/Hail damage.
  • To increases the amount of HP recovery gained via Leftovers / Black Sludge, have the HP stat setup to a multiple of 16, or 16n.
    • Ex. Values of 16n:  ... 112, 128, 144, 160, 176, 192, 208, 224, 240, etc
  • For Life Orb users: To decrease Life Orb residual damage taken, the Actual HP stat should end with a value of 9, or 10n - 1. This is so the LO user will take one point less, and potentially get off an "11th" LO hit before fainting, if possible.
  • For maximum Sitrus Berry recovery (25%), the value should be a multiple of 4. 
    • An even number can be useful to activate Sitrus Berry when the opponent uses Super Fang.
  • For maximum Substitute usage to produce 4 instead of 3, make sure the HP value is 4n + 1.
  • For maximum Regenerator recovery on the switch, the HP stat should be divisible, or a multiple of 3. 

EV Spread Optimization Guides & Threads
Here's some other resources used from various sites which basically explain the concept of EV effeciency, with most of the info found here. Here's the links down below.

[spoiler] [/spoiler]

The next two parts will be regarding Offensive and Defensive EV Spreads, though some of this might have been covered, or reviewed again. For a future TLDR note, always consider specific goals and benchmarks for making either type of EV Spreads, just don't over do it and over complicate the EV Spread. Remember the goal for Custom EVs is to always achieve a certain goal first, before taking any considerations. Didn't want to include "everything" as this post is already lengthy as is and most of this information is moderately difficult to take in all in one. Will work on both Offensive and Defensive EV Spreads with examples from the VGC 17 metagame with Su/Mo. Hope this suffice for now. 

No comments:

Post a Comment