Curious about what we're currently doing about bots? Check out the latest updates from staff in this thread.
Welcome to Smogon! Take a moment to read the Introduction to Smogon for a run-down on everything Smogon, and make sure you take some time to read the global rules.
completely based on style design and how much it makes the mon more popular/better
something that I would mark
mega beedrill might be too high but it indeed does make beedrill a LOT more popular, like out of 10 beedrill fans 9 are actually mega beedrill fan, and the last one never knew that there is a base beedrill.
same for crabominable and mabye banette? I was banette fan for long as it was one of the mons I knew before I even really like know pokemon. Crabominable might be too high as well but I some what like it.
And I love gardevoir, but that can't change the fact that they gave it a super stupid and uncreative mega, it's just having a larger dress and that's it, like what the fuk?
here's mine! some of them i'm not too familiar with. i'm familiar with most of the Z-A ones but i haven't played through Z-A just watched my fiancé play through it (the battle style just isn't for me) so i don't know much about the end of the story and i haven't really gotten to see some of the megas from later on. :P
I didn't have any hard criteria but design probably played the biggest part in determining placement. Some other factors shifted things, of course. For example, I think Mega Sableye actually has one of the best Mega designs, but it was bumped out of S because of my disdain for it in ORAS OU before its ban.
S is the best of the best.
A is very good but didn't resonate with me as much as S.
B is decent.
C is indifference or mild dislike.
D is dislike.
You can guess why Mega Audino has its own tier. I have made no secret of my opinions on it.
Also, these are ordered only by Dex number within the tiers because I'm indecisive lol. I doubt Zard X would be in the front of A otherwise.
A lot of this comes down to aesthetics. I just don't like the overdesigned nature of Megas broadly. My favorite Pokemon tend to be those grounded in either a direct ecological framework or a mundane modern human reality. Mega Pokemon do not tend to work within those boundaries. Almost all of them are gaudy, over designed, and blatantly impractical looking. Many of them have action figure esqe poses they are designed to take on and then don't look natural when they are idle.
This leaves gameplay as the main avenue to appreciate Mega Evolution, and there I don't care much for them either. Being relegated to the realm of Competitive Play for all reasonable intents and purposes puts a very harsh spotlight on exactly the hard boundaries which Mega Evolutions work within. They are the gameplay designs in Pokemon which are most forced to be understood within the framework of a Deck Builder, with traits designed to fit into a Metagame with a dedicated Niche and maybe a surrounding Archetype. Through the implementation of their limitations, they are even pretty close mimics to many Player Unit mechanics seen in Trading Card Games; Magic the Gathering's Commanders, One Piece's Leader Cards, and Hearthstone's Classes come to mind as a very obvious comparison, but even more broadly Mega Evolution is the mechanic in Pokemon which most directly breaks the otherwise steadfast and unspoken rule that any set of Pokemon that exist together in a format can both be put on the same team, an irregularity in the world of TCGs which often include very broad restrictions on card combinations*.
I also liken the design of Mega Evolutions to Cards designed for Constructed Formats of Deck Builders because if they don't have relevance in a particular Metagame then they basically don't exist.
Any regular old Pokemon that gets designed gets to exist in it's own context. We might all be in agreement that Beedrill is bad in the metagames almost every Constructed Format it has ever been a part of, but that's OK because it is almost always the best Pokemon you can have in a party of Pokemon at Level 10, it can fulfill its role as a midgame filler just fine, and it can appear as a somewhat dangerous escalation on generic Bug Type using opponents. It is a gameplay design that works swimmingly in an RPG context, even if it completely falls flat in Constructed Formats of PVP. Over time, Pokemon has trended towards ensuring that individual Pokemon can always fit into Constructed Formats in some fashion, even if it ends up requiring some mad scientist surgery stitching to pull together what ends up feeling like two disparate gameplay identities into one Pokemon. But aside from a spare few odd cases like Mythical Pokemon, every new Pokemon gets a gameplay role to fill in the games where they are introduced, and even if that role is to be a mediocre but memorable member of a mobsters menagerie that tends to be a manageable mandate. This is the biggest advantage Pokemon have as gameplay designs compared to the filler gameplay designs present in many Deck Builders with a focus on Constructed Formats; the least impressive Pokemon designs have infinitely more depth and potential to tap into than the "Pack Filler" gameplay pieces of the average Deck Builder. Mega Evolutions, being endgame buffs stapled onto existing designs, don't really have this luxury.
So this brings me to the controversial(?) thesis statement I made before;
As a mechanic, Mega Evolution has so many potentially interesting design ramifications. A Pokemon basically morphing into something else at the player's command has wild design potential. I'll be clear that I think it makes sense to have plentiful factors reigning that concept in from being too overwhelming and flavorfully incoherent. The transformation always containing a massive straightforward statistical buff is... sensible given the overall power fantasy at play and aim to use the designs as marketing vessels, if nothing else, but definitely causes issues with how that power has to be repaid later. I don't know if I'd say I disagree with Mega Evolutions being disallowed from holding items or having exactly one ability, but these are blatantly things that make Mega Evolutions less interesting than standard Non-Mega Pokemon are. Even past that core conceit, however, the exact framework for how Mega Evolutions are designed and how Mega Evolution as a mechanic was balanced end up dooming the whole project in my mind. It's a bit of a cascading failure though, so to clarify my point I want to start with the biggest thing that really makes Mega Pokemon Boring to me.
Held Items are a Big Deal.
Real quick I need to mention here something about my own experiences. The form of Competitive Pokemon I play the most is actually Draft. There's a pretty big reason I've been distinctly referring to Constructed Formats this whole time; what that term refers to is formats where both players enter battle with teams that are designed to face any other possible legal team, with an agreed upon list of legal and illegal options for both players. This just describes the way most players engage with Competitive Pokemon in general, with the most popular exception being Showdown's Random Battles with Draft formats of various shapes and sizes a distant second.** Draft is a format where players bring a preplanned team, but unlike true Constructed Formats the legal options for both players are unequal by design.
The biggest distinguishing factor of Draft is that the planning of matchups on a case-by-case basis removes many of the defacto restrictions Pokemon face when seeing use in a Constructed Format. To me, Draft Formats represent one of the best ways to see the full potential a Pokemon's gameplay design can have, forcing strong Pokemon into edge-case situations where they need to explore their toolkits to overcome challenging matchups and allowing weaker Pokemon to play in formats where they can punch well above their weight. Much of this variety comes in the form of Pokemon exploring the Held Item Pool more thoroughly than they would; staples like Boots in Singles / Sitrus in Doubles persist as the most common options, but situational items like Air Balloon, Custap Berry, Resist Berries, Eject Button and so on see real play, while items that are widely understood to be strong but specific like Assault Vest, Rocky Helmet, and Choice Scarf see use on a much wider variety of Pokemon than they would ever be considered on in Constructed Formats. Held Items also often go hand in hand with specific moves, enabling them in a way that otherwise wouldn't be possible or reasonable. This sometimes comes with direct synergies like Loaded Dice + Scale Shot, Flame Orb + Facade, or Power Herb + Meteor Beam, but it also manifests in more subtle ways like Assault Vest and Choice Scarf encouraging Pokemon to differ from using standard setup moves or support tools and instead use extra supportive Attacks or type coverage, or with HP Recovery Items like Leftovers and Sitrus Berry enabling niche moves and strategies which require Pokemon to survive over the course of multiple turns. So beyond the things which having no Held Item strictly forbids or the things which are legitimately impossible to utilize (ie moves like Trick / Belch or abilities like Magician / Gluttony) , there is also a large amount of moves or pairs of moves which are simply never worth reasonably considering without a fitting Held Item to reinforce them.
So I'm sure some of my perspective here is coming into focus. If unique Held Items are a staple of how Pokemon can push their boundaries, then Mega Evolutions are by default missing a large portion of possibility space that average Pokemon have by default. Because of this baseline restriction, Mega Evolution movesets gravitate toward a consistent pool of good options for their given formats. In Singles, this is usually 3 Attack Setup Sweepers, sometimes exchanging the least important Attacking Move for Substitute, a Recovery move, or a Hazard (Removal) move. In Doubles, this is paradigm is squeezed even further by Protect, which usually outcompetes either the 3rd Attacking move or the Setup move itself, or sometimes ends up a Speed Control move like Tailwind, Trick Room, or Icy Wind. With Protect being completely Universal and there being very few Pokemon which lack prominent Setup options, Mega Pokemon don't just lack variety in and of themselves, they also end up homogenous as a collective.
This loops back to why I mention my preference in format, and Deck Builders for that matter. This lack of set variety I've been harping on about isn't as big of a deal in Constructed Formats. Players won't usually have much reason to explore the niche options available to any given Pokemon if it isn't already a large part of the relevant metagame, and a large portion of the Pokemon legal in a format will only really have one or two movesets they could reasonably run to begin with. But in Draft, the full range of sets a Pokemon could potentially run is much more important, and being relevant in a particular developed metagame ceases to be a prerequisite for a Pokemon having its toolkit fully explored. This is the biggest reason why I want to focus on Draft formats with my analysis. Constructed Formats can only have so many strong options that are explored in depth. Even with the stratified tiering brackets of Smogon 6v6 Singles, some Mega Evolutions have been perpetually stuck in Borderline tiers while in VGC the lack of any tiering system means that any Mega Evolution that has gotten power crept since being introduced will probably never be fully explored in a 4v4 Doubles Constructed Format. (Shoutouts to the Champions 4v4 Doubles UU folks if tis kind of thing would interest you)
So that is the big limitation of Mega Evolution. But it's also core to the central balance they are aiming to strike, a core part of the bargain at play with their design; get a large boost to Base Stats, and in return lose the flexibility afforded by a Held Item. Let's be clear; the big idea I have about Mega Evolution is not that they should remove the Held Item restriction. Being locked to a specific Held Item does make Mega Evolutions less interesting and less variable, but it also isn't a death sentence to creating interesting gameplay designs. Having very high base stats can be a point of interest in and of itself, and there are plenty of ways in which Held Items can't replicate the effects of raw stats. But that doesn't mean I am done finding bones to pick with the way Mega Evolution is done, and my next big gripe is with the methodology around the Mega Evolution Stat Changes.
BST Parity is Nonsense
Base Stat Totals. The number which represents the cumulative total of a Pokemon's Base Stats. Those raw stats I was mentioning. They've got a weird history. The developers have clearly been tracking this number when designing Pokemon from the very start. Many counterpart Pokemon were given equal BSTs, such as the Eeveelutions, Magmar and Electabuzz, Butterfree and Beedrill, Golem and Machamp, and even the Starters. That last one being especially fun to note because it is actually specific to Generation 1, as when the Special Split was applied in Generation 2 Charizard (and by extension Typhlosion) got an extra 9 total Base Stats compared to the other final stage Starters. I'd even speculate that one of the more notable examples of this not applying, the Legendary Birds, was a deliberate design choice.
In Gen 1, the birds have slightly stratified BSTs. Articuno at 485, Zapdos at 490, and Moltres at 495. These actually inversely correlate with how strong each bird is ingame, with Articuno the runaway favorite and Moltres the clear last place finisher - all this keeping in mind RBY's unsplit Special stat giving Articuno a full 125 SpAtk and considering JP RBY's absurd iteration of Blizzard with 120 BP, 90% Accuracy, and a whopping 30% Freeze rate. It also neatly accounts for the more obvious stat optimization done to each, as despite a higher raw stat total than Articuno and Zapdos, the lion's share of that difference is accounted for in Moltres' lesser used Attack stat. So despite having the highest raw stat total of the three birds, Moltres arguably has the worst stat spread of all three birds in RBY. The Special Split of Generation 2 would remove this disparity, and arguably leave the Birds in a worst state of balance than they had in Generation 1, but my belief that the developers had a much stronger grasp of game balance than anyone gives them credit for in Gen 1 specifically is a rant that can wait for another day.
All of that is to say, Base Stats are not created equal. There's reasonable enough evidence that the designers have known it since the very first iteration of the battle system before the turn of the millennium. As is the truth of many under-the-hood statistics about Pokemon, the designers were not really expecting players to thoroughly pick the system apart to the bones. I will not be doing the internet archeology to try and find out when fans first started calculating Base Stats, mostly because I can't read Japanese anyway, but I don't think anyone out of specific competitive circles was really digging into these numbers back in the day, and they certainly weren't documented in anything as forward facing as Pokemon Showdown! itself. Even you don't know the axioms, if you have been looking at the numbers for long enough you probably know some of the short rules about how Base Stat Optimization goes.
The Primary Attacking Stat is more valuable than Defensive Stats, which are more valuable than the Secondary Attacking Stat.
An extra point of Attack is canceled out by an extra point of Defense, so any boost to one Defensive Stat is half as efficient as a boost to an Attacking stat.
HP and Defenses are cumulatively optimal when they are all roughly equal as Base Stats#, as excess Defenses don't apply to taking hits from the opposite side of the spectrum and excess HP makes Defenses less efficient.
A single extra Base Stat in Speed can be game defining in one case while a double digit boost can be almost negligible or even actively detrimental in another. Benchmarks are primarily contextual.
Simple way to understand is Equal Base HP and Defense is the optimal distribution for taking hits on that side of the spectrum, but this advantage thins out and eventually flips when combined HP and Defenses get really high. I'm talking beyond Arceus Bulk here - with no investment and 31 IVs 120 HP / 120 Def / 120 SpDef takes hits about equally well to 200 HP / 80 Def / 80 SpDef - and past that point excess HP relative to Defense has incremental advantages to optimal distribution. But it'll be apparent why discussing cases with extreme HP values won't be super important to my point pretty soon.
What I do know for certain is that everybody was talking about BST when Mega Evolution came out. Because everyone knows the rule around the two. Mega Evolving gives a Pokemon exactly 100 Extra Base Stats. It's a 100% consistent truth, except that one time it barely wasn't for a bit with Alakazam. It's much more accurate and statistical to say than just "they get a big boost to their stats". It gave people a way to quantifiably talk about how strong Mega Evolutions were, and how they dwarfed other minor Legendary Pokemon.
Now I can't entirely blame the developers for choosing a specific stable number to increase BST by. It certainly makes deciding a lot of default values a lot easier, and doesn't give room for complaints about favoritism from people who are at just the wrong point of understanding about the game's systems. But at the same time... we know BST Parity is Nonsense. And if you are trying to create relatively optimized Base Stat Spreads for Mega Evolutions, then you'll end up with Stat Changes that focus on buffs to One Attacking Stat, secondarily buffing Defense, and orienting Speed Stats post Mega Evolution around metagame relevant benchmarks. And that would lead to a set of Mega Evolutions which, in cases where they there aren't bountiful Base Stats to work with, will inevitably funnel themselves into the same kinds of stat gains.
The Mega Evolution Stat Gains and Optimization
Shoutouts to the Mix and Mega Page for being the best resource on the Internet for information on the exact Stat Boosts of each individual Mega Stone? They had a few wrong but this data wasn't anywhere else easily so I'm not complaining. Also, for the calculations I do below, I am only considering Mega Evolutions that have their Abilities assigned at the time of writing, which is in the wake of Champions Regulation MB's release.
So a reasonable thing one might do after making the statement "The Mega Evolutions are All The Same" is to back this up with some Empirical Data. If you run the numbers with no alterations, you get this when calculating the Average Mega Stat Change;
Average Mega Stat Change : +27 Atk / +25 SpAtk /+20 Def /+18 SpDef / +10 Speed
So yes, Mega Evolutions are predisposed towards gaining Attacking Stats over Defensive Stats and Speed. That is a trend I think folks would be able to guess. But let's do a bit of tweaking to what data we are exactly looking at. Because remember little ol' RBY Moltres? With those extra points in an Attack stat it seldom uses? We need to care about which Attacking Stat is being raised.
Enter, Primary and Secondary Attacking Stats. A Pokemon's Primary Attacking Stat (PAS) is The One It Uses, and the Secondary Attacking Stat (SAS) is The One It Doesn't Use. I don't really like thinking of Pokemon designs in this way broadly but the fact that we can talk about Mega Evolutions in this framework at all without losing much precision is approaching my point, so I will continue. There are cases where this is ambiguous, and that could have been a problem, but thankfully with this group it really isn't for reasons I will highlight later. (For the sake of my calculations, the Primary Attacking Stat is the stat used by the Mega Evolution, not the Base Form. So for Starmie Physical Attack is the Primary Attacking Stat, while for Dragonite it's Special Attack.)
The only case where it's genuinely awkward to assign AND the stat boosts aren't equal on the Mega Evolution anyway (meaning me being wrong in this assignment has no consequence) is Sableye, which gets +10 Atk / +20 SpAtk and ends up wavering between Knock Off as a generic utility move that deals existing damage and Dark Pulse setup sweeping sets backed by Calm Mind depending on the format. I considered Special Attack to be Secondary in my calculations, which shouldn't end up being especially impactful on the data output but is a hedge against biasing the data toward the trends I'm going to be proposing.
With just that change alone, our new result is notably different.
Average Mega Stat Change : +37 PAS / +15 SAS / +20 Def / +18 SpDef / +10 Speed
Where before the Attacking Stats were both slightly above the Defensive Stats, now it's clear that the Primary Attacking Stat has the lions share of these buffs, with the Secondary Attacking Stat ending up lower than both Defenses. In fact, there is not a single Mega Evolution which has anything lower than a +10 Boost to their PAS within in this framework.
Here, I've charted Mega Evolutions by their deviation from the above Average Mega Stat Change. A higher number here represents a higher deviation from the norm. We'll call this Average Stat Change Deviance. (ASCD? If I even use this acronym again?)
Overall, this statistic basically represents how uniquely transformative a Mega Evolution is for a Pokemon in terms of raw stat changes alone. There are a few important trends to lay out with how to view Pokemon within this measurement. Because the PAS / SAS as a statistic is only considering the Mega Evolution's preferred stat, this measure doesn't capture the transformation of Mega Evolutions which switch their PAS / SAS compared to their Base Form, namely with Charizard X, Mewtwo X, Dragonite, and Starmie. Because every Stat increases in the Average Mega Evolution, any Mega Evolution with a stat that doesn't increase gets a notable deviance score, with it being especially strong in the cases where any stat actively decreases. This measure doesn't cover the proportionality of changes, so the 50 point increase to Defense on the Base 180 Aggron and the Base 75 Sableye are considered equal here despite having different impacts. Lastly, without considering Abilities in any way, this does provide an incomplete picture of the changes Pokemon undergo in several places, especially when it comes to Abilities which increase Damage.
Even with those caveats in mind, that's... pretty tame variance, right? With 5 stats being changed here, a Pokemon with a 50 on this ASCD scale only deviates from the standard stat change by about 10 points per stat. For the Pokemon below, their variance is even lower. For many Pokemon above that threshold, the extra variance they have represents increased optimization more than anything else. Afterall, the +37 PAS boost is only an average, many things get even more and when half or more of the allotted bonus stats are allocated to PAS a Mega Evolution inevitably won't be gaining many other stats due to the 100 BST Limit. Most of the cases that don't explicitly raise PAS more than average come with some kind of various obvious caveat. Salamence, Metagross, Groudon, and Kyogre all have absurdly high Stat Totals and Damage Boosting Abilities to begin with, and seem to if anything be purposely made unoptimized to not be completely broken. Chesnaught (and to a lesser extent Aggron, and to a much lesser extent Slowbro) gains an absolute boatload of Defense rather than purely gaining PAS, but this boost functionally serves as an offensive optimization anyway because of access to Body Press. Several of these Pokemon get damage boosting abilities that could work as exceptions but rather than muddy those waters just yet I'll just point out Mawile newly acquiring Huge Power.
A high ASCD on this scale doesn't entirely correlate with optimization. It's only towards the very extremes where we start seeing Mega Evolutions that don't act as optimizations of an existing playstyle. Diancie is an entirely unique case for several reasons, but I do look to Sableye, Skarmory, and Scolipede as great examples.
But given this statistic is an overall poor measurement of optimization, how can we more closely look at just how optimized most Mega Evolutions actually are? What are some of the biases that color the baseline used to generate this variance? In the spoiler below I go into a bit too much detail about the ways that the +15 SAS boost I list for the Average Mega Stat Change is, if anything, an overestimate of how relevant those boosts actually are and how much optimization there is.
Pokemon that have 600+ BST Prior to Mega Evolving have notably different Mega Evolution buffs compared to others. Where the majority of Mega Evolutions have relatively optimized Stat Buffs, some of these especially high BST Mega Evolutions decidedly aren't. Consider the dichotomy here;
Average Mega Stat Change (All Megas): +37 PAS / +15 SAS / +20 Def /+ 18 SpDef / +10 Speed
Average Mega Stat Change (<600 BST): +37 PAS / +13 SAS / +21 Def / +19 SpDef / +10 Speed
Average Mega Stat Change (>=600 BST): +36 PAS / +26 SAS / +14 Def / +11 SpDef / +13 Speed
Average Mega Stat Change (^-Diancie): +34 PAS / +23 SAS / +19 Def / +16 SpDef / +8 Speed
While removing the 12 (out of 83 total) Mega Evolutions which start with over 600 BST doesn't have a major impact on the average, it's pretty clear that in isolation they absolutely have an inflated boost to their SAS compared to the majority of Mega Evolutions. I include the stats for this group with and without Diancie because it acts as a massive outlier for Mega Evolution stat changes, with by far the most polarized stat changes of any Mega Evolution. With the direct comparison from the all inclusive measurement being a series of small decreases to stats across the board funneled into a significant increase in SAS, this pretty directly just says that the 600+ BST Mega Evolutions are directly unoptimized compared to their more standard counterparts. To me, this makes sense as a conscious design choice. These are Pokemon that are on the whole already expected to be a step above their peers, making the optimization of their Mega Evolution Stat Boosts less necessary for them to thrive in a metagame setting.
Another factor which influences the need for Stat Optimization is Abilities! On top of 100 Extra Base Stats, plenty of Mega Evolutions have Abilities which further increase their damage. The ways this can manifest are actually fairly complex, which I elaborate on in the spoiler, but to keep things simple here I am only focusing on Abilities which have a plainly stated and reliably active Damage Multiplier attached to them. All in all, about 1/3 of Mega Evolutions are going to count as having a Damage Boosting Ability in my strict criteria.
Front and center, for the sake of the data used in my assertions, only the Direct Damage Boosts (and Huge Power clones) are considered. The consistency of access to the Damage Boost was considered more important than the quality of the Damage Boost in my assignments here; Primal Groudon and Mega Aerodactyl might run a fair amount of sets which never even utilize the Damage Boosts their abilities provide, but they don't require a turn of setup, a partner, or an opponent to facilitate that Boost. I was very tempted to also include Megas like Meganium, Pidgeot, and Froslass as well given the extent to which their abilities' newly granted access to consistent forms of Solarbeam / Fire Weather Ball, Hurricane, and Blizzard respectively inform their core gameplay identities through the raw increase in power they have. Getting into the weeds of exactly why those do represent large damage buffs while Abilities like Hawlucha's No Guard or Heracross' Skill Link don't quite have the same importance isn't something I feel like focusing on here.
Average Mega Stat Change (All Megas) : +37 PAS / +15 SAS / +20 Def / +18 SpDef / +10 Speed
Average Mega Stat Change (No Boost) : +39 PAS / +12 SAS / +21 Def / +19 SpDef / +9 Speed
Average Mega Stat Change (Yes Boost) : +33 PAS / +19 SAS / +20 Def / +16 SpDef / +12 Speed
Mega Evolutions which have Abilities that boost their damage tend to have smaller boosts to their PAS and larger boosts to their SAS, as well as a slight extra focus on Speed at the expense of Defenses. The change in PAS and SAS boosts points to another example of accounting for optimization; if a Pokemon newly gains access to a Damage Boosting Ability alongside the standard Mega Evolution Stat Boost, then it gets a disproportionately high amount of extra Stats that can have their impact reduced by diluting them more equally across both Attacking Stats. There are a fair number of Pokemon which gain access to a Damage Boosting Ability that enables both Physical and Special Attacking sets and create a gameplay design which utilizes Mixed or SAS focused sets. But there is also another trend that solidified this as a relevant stat to me. Of the roughly 2/3 Mega Evolutions which don't have a Damage Boosting Ability, 22 of them have no change or a decrease (only Dragonite, -10) to their SAS. Among the other 1/3 that do have a Damage Boosting Ability, only 4 of them have no change or a decrease in their SAS. Within those 4 is Mega Beedrill, a huge outlier due to having a BST so low that even the Mega Evolution Stat Boost barely brings it into the range of average Pokemon, resulting in the developers actively reducing its SAS by 30 (by far the largest decrease to any attacking stat on a Mega Evolution) to allow it a massive PAS and Speed Boost that could make it reach competitive stat thresholds without breaking the +100 BST Rule. Even with that kind outlier in the smaller Damage Boosting Ability dataset, the trends of PAS Boosts decreasing and SAS Boosts decreasing still remain intact.
Lastly, I want to mention Speed. As opposed to all of the other stats which are more effective when they are higher, Speed is a stat which is effective relative to other Pokemon in a metagame context. Thus, there are a portion of Mega Evolutions where their role is solidified by them having their Speed remain the same or even decrease rather than increase. Without a benefit to raising Speed in the optimization process, there ends up leftover stats that can be distributed across the rest, which should on paper increase all of them.
Average Mega Stat Change (All Megas) : +37 PAS / +15 SAS / +20 Def / +18 SpDef / +10 Speed
Average Mega Stat Change (No Speed Drops) : +37 PAS / +14 SAS / +18 Def / +17 SpDef / +14 Speed
Average Mega Stat Change (0 or Dropped Speed) : +38 PAS / +16 SAS / +30 Def / +22 SpDef / -5 Speed
Average Mega Stat Change (All Speed Drops) : +37 PAS / +18 SAS / +37 Def / +27 SpDef / -19 Speed
But those leftovers are not distributed equally. The most obvious observation is that decreases in Speed are correlated with much larger than normal increases in Defenses, and especially Physical Defense. The change to SAS boosts is relatively minor but still fairly notable, as is the lack of any change to PAS boosts. Even if greatly increased Defenses aren't as unoptimal as possible, the lack of any PAS boost increase alongside the slightly higher SAS boost increases is a suboptimal distribution of stats that mostly makes sense because of the extra Stat increases afforded by drop in Speed.
Combining all of these together, keeping in mind there is overlap between some of these categories;
Average Mega Stat Change (All Megas) : +37 PAS / +15 SAS / +20 Def / +18 SpDef / +10 Speed
Average Mega Stat Change (<600 BST): +37 PAS / +13 SAS / +21 Def / +19 SpDef / +10 Speed
Average Mega Stat Change (No Ability Boost) : +39 PAS / +12 SAS / +21 Def / +19 SpDef / +9 Speed
Average Mega Stat Change (No Speed Drops) : +37 PAS / +14 SAS / +18 Def / +17 SpDef / +14 Speed
Average Mega Stat Change (<600 BST & No Ability Boost & No Speed Drops) : +39 PAS / +9 SAS / +20 Def / +18 SpDef / +13 Speed
The SAS increases absolutely crater proportionally, being cut by over a third. Speed takes a portion of this increase as would be expected, but the other portion feeds directly into PAS rather, with Def and SpDef seeing such insignificant changes that they don't even change the results of rounding.
The point I make there in a very lengthy and verbose way is fairly simple; a significant portion of the stat gains to SAS on Mega Evolutions are present as a result of not needing to account for other baseline factors that make optimized Base Stats less relevant, such as a high initial BST, a powerful Ability that boosts damage, or an optimizing decrease to Base Speed that gives a Mega Evolution more extra Stats to spend elsewhere. Accounting for these factors, the already stark difference between PAS and SAS boosts gets even more polarized towards optimization. Combined with the lack of Held Items to potentially specialize their kits towards their SAS, the result is that over half of all Mega Evolutions cannot run a reasonable SAS focused / mixed set. That proportion increases to over 2/3 of all Mega Evolutions when excluding Mega Evolutions which start from above 600 BST. Outside of hyper-specific Draft settings involving 4* super effective moves or Blissey / Avalugg levels of Defensive Stat polarization, these Pokemon will never find a local optimum that involves using attacks from their SAS to deal damage independent of utility.
I tried to be generous in looking at Mixed Attackers but things are really really bad here; consider that the majority of all standard Pokemon would fall into the "Very Niche SAS / Mixed Sets can exist" category. Those in the higher 600+ BST tier all fit in the "Strong Use Cases" tier at minimum. (Also I'd probably bump Emboar down a tier in this graphic in hindsight)
Also, Mega Floette is really in a whole other league in this regard. It's Physical movepool is Exactly Endeavor, Petal Blizzard, Seed Bomb, Facade, and Trailblaze. Despite having a PAS / SAS Gap smaller than 90, it is arguably worse at utilizing a dedicated SAS set than Mega Beedrill. Which has F I F T E E N base SpAtk. If Mega Floette was put in Little Cup but could only use Physical Attacks it would not be worth using because it could be walled by a Gulpin.
Look at these Level 5 calcs
0 SpA Adaptability Beedrill-Mega Sludge Bomb vs. 0 HP / 0 SpD Gulpin: 5-7 (21.7 - 30.4%) -- 98.1% chance to 4HKO
0 Atk Floette-Eternal Facade vs. 0 HP / 0 Def Gulpin: 6-8 (26 - 34.7%) -- 0.9% chance to 3HKO
Gulpin resists Poison by the way.
This is the direct result of optimization efforts made to squeeze maximum value out of the allotted 100 Base Stat budget. For the vast majority of Mega Evolutions, especially those with a 60+ Point Gap between their PAS and SAS, an increase to their SAS (that would inevitably break the BST Parity alone) would only have the potential to make the Pokemon a better gameplay design. For differences that exist entirely within Foul Play, Confusion, already weak Utility Attacks, and other even more miniscule tools at the very margins of the game's mechanics as far as constructed play is concerned, the return would be for these Pokemon to have a significant increase to their set variety for Draft formats increased, along with just making them better able to use bad sets for Casual play. In the absolute worst cases, for the few Pokemon where an absurdly large SAS boost wouldn't make any impact at all on the sets they use, the difference would just be a bizarre footnote on a spreadsheet to 90% of player. I don't think every Mega Evolution needs this kind of change, but it would undoubtedly be a positive change for the majority of them. That's all I'll wax on about Mixed Attacking, which is a fun but not at all a necessary element of gameplay design. But I do want to bring my focus back to Optimization in other ways, namely because of how it interacts with another core trait of Mega Evolution. That Button you Press.
Mega Evolutions are Composed of Two Pokemon and that Almost Never Matters
One of the core unique things about Mega Evolution is the fact that a Pokemon that can Mega Evolve is effectively undergoing mid battle metamorphosis. There are Two Pokemon crammed into One Gameplay Design. You technically can't separate Mega Evolutions from their Base Forms. Even the fact that they can lead a battle and Mega Evolve instantly doesn't change the fact that being pressured into such a lead can be predicted around, and should a Pokemon not Mega Evolve instantly on the first turn a player may find themselves in a position where they need to switch in a Pokemon that hasn't yet Mega Evolved to use it's defensive traits.
That is the absolute bare minimum of interactions which occur with such a transformative mechanic. That is the kind of complexity they could only keep out of Mega Evolution by inventing an entire secondary splinter mechanic that seemingly does nothing but ensure it doesn't take place, and I will only have more scathing things to say about Primal Reversion as it comes up. And most importantly it is a practically non-existent fraction of the overall potential of a mechanic like this. That isn't to say that Mega Evolution was ever really going to explore the very fullest depths of possibility space here. It's more than understandable that Mega Evolution is doesn't go completely off the rails with transformation as a feature - the flavorful concept of Mega Evolution is as an enhancement of an existing Pokemon rather than any kind of complete overhaul. And once again the flat +100 Base Stat boost does play a significant part in how Mega Evolution plays out mechanically, as it's hard to create any kind of balance between two Pokemon when one of them just has so many extra Base Stats. But there is so much possible within the framework of transforming into another Pokemon that Mega Evolution just doesn't tap into, doesn't even try to tap into. For most Pokemon in most situations where it is available, The Button to Mega Evolve your Pokemon is a Checkbox, not a Decision.And that suuuuuuuuuuuuuucks.
Let's do a quick look at the kinds of decisions players do make around Mega Evolution.
Bluffing is when a player aims to leverage hidden information dynamics to display a potential threat they don't have. Mega Evolution can be the subject of bluffs in a few different ways. In some cases, players can use multiple Pokemon which could Mega Evolve and make it unclear which one of them actually will Mega - in others, players can utilize Pokemon that could Mega Evolve with a different held item like Choice Scarf or Focus Sash, taking advantage of how an opponent might make assumptions of what will work in any given situation. Pokemon with multiple potential Mega Evolutions can also bluff being one of them instead of the other as well, but these dynamics are very exclusive for obvious reasons.
There are two things I want to point out with bluffing that make it something I don't really want to put too much focus in. This first is that bluffing is incredibly format specific. Many formats, including Competitive VGC, use an Open Team Sheet system that removes the teeth of most bluffing attempts right off the bat, while others like Draft often stipulate that Mega Evolutions and Base Forms are separate as a form of balance which hamstrings the potential depth of bluffs heavily. It's only really possible to play within this design space in Competitive Singles or on the Closed Team Sheet VGC Ladder, the latter of which is fun but isn't especially serious. The second is simply that Mega Evolutions themselves don't really make much use of bluffs. Think of it this way; for a Mega Evolution to itself be bluffed, the Base Form has to be the expected outcome. That just isn't the case for the vast majority of Mega Evolutions across formats, so most kinds of bluffs around Mega Evolution involve the Base Form bluffing the threat of the Mega Evolution rather than the other way around. But there definitely are exceptions; in singles there are the likes of Clefable and Glimmora which have extremely valuable and unique traits in their Base Form but can Mega Evolve to flip certain matchups, in doubles the presence of Sableye or Kangaskhan can be deceptive from Team Preview, and across all formats Garchomp is not ever really expected to utilize its (original) Mega Evolution.
From here out, I'll be focusing on Mega Evolutions which specifically utilize their Base Forms rather than just bluffing them.
Inbuilt Setplay :Sharpedo::Scolipede::Glalie::Scovillain::Audino::Slowbro::Greninja:(:Sableye: and :Glimmora: On Paper?)
Some Pokemon have powerful abilities that passively give them huge benefits. Abilities that give passive Stat Scaling like Speed Boost or Moody innately reward a Pokemon for existing on the field turn over turn. Regenerator Pokemon recover HP by switching which is incredibly valuable in Singles. Sometimes Pokemon have a notable downside to their ability once Mega Evolved. Modern Protean can make using moves like Protect or Swords Dance awkward, so being able to use them in the Base Form and then Mega Evolve afterwards to keep STAB on your first attack opens up lines of play for Greninja. These are cases where the interactions are baked into the premise of being a Mega Evolution for a Pokemon that has certain kinds of abilities in their Base Form, and they tend to be fairly simple in execution but can still be engaging to execute upon when the need to rely on them is disruptive.
Specific Counterplay :Froslass::Excadrill::Scrafty::Kangaskhan::Mawile::Manectric::Chandelure::Gardevoir::Alakazam::Meowstic:
There are plentiful situations where timing a Mega Evolution is important contextually due to the advantages of the Base Form's Ability or the disadvantages of the Mega Evolution's ability. Many of these interactions revolve around Intimidate, which is an ability that swings from being a defining feature of a Pokemon to being useless to even being an active liability in the wrong matchups. Choosing when to start or stop having Intimidate and choosing when to start or stop being immune to / benefitted by the effects of Intimidate is a common theme with ability interactions, and is the most broadly common situation among these. Other situations tend to only arise in far more specific situations. Mega Evolutions which gain weather setting abilities often play mindgames involving when they Mega Evolve in relation to opposing weather, while Base Forms with abilities that benefit from weather can utilize those abilities in a pinch against opposing teams aiming to utilize that weather. Base Forms with immunity abilities like Manectric or Chandelure can find themselves utilizing those immunities on odd occasions to protect allies or withstand an attack. Trace is an ability for which timing is nearly everything, and whether the case is like Gardevoir which starts with Trace and loses it or is like Alakazam or Meowstic which gain Trace upon Mega Evolving, there is usually something worth considering about how Trace will effect abilities when deciding whether or not to Mega Evolve.
No matter how broadly beneficial a change in Typing is, it is almost impossible for it to come without any downsides. Staying in a Base Form to be able to access the beneficial properties of a Typing Profile lost upon Mega Evolution is naturally a distinct calculation on a case by case basis. For Pokemon like Aggron and Meganium, the benefits of the Base Form's Type Profile might be completely overshadowed by the faults, making considerations about Mega Evolution nearly non-existent. For Pokemon like Gyarados, Charizard (X), and Barbaracle, utilizing the defensive benefits exclusive to the Base Form's Typing is a core part of their gameplay, especially because they centralize around Scaling Strategies which see them taking turns where they don't Attack and thus only care about minimizing the damage they take by any means necessary. Most Pokemon fall somewhere inbetween, meaning... they still only rarely need to consider the value of staying in their Base Form.
Trying to put actual rates to how often these situations come up is both difficult and meaningless. There are more small situations I haven't listed here, and infinitely more situations that technically exist where not Mega Evolving is optimal because any increase in Stats can technically have a downside in the right context. The overall trend though is obvious. You just Check the Box on your turns with Mega Evolution. As much as these edge cases do exist they more than thoroughly prove the existence of the underlying rule.
No I didn't forget about Multi-Mega it just barely exists in optimal play
There's one extra rule of Mega Evolution; you can only Mega Evolve once per battle. While this can technically come up with only one Mega Stone through Revival Blessing, (or even ingame with just. Revives.) this manifests primarily in preventing players from using two different Mega Evolutions in the same battle. This is, by all accounts, a reasonable rule. Mega Evolutions have tons of stats and utilizing a team full of them would be an optimal strategy without some kind of limit. It does have an important consequence for decision making. But that decision making isn't centered on battles; it's centered on the Teambuilder.
Players have optimized this rule to just say "Don't build teams that play multiple Pokemon holding Mega Stones in the same battle" and that just holds true the runaway majority of the time in constructed formats. Prior to Pokemon Champions Reg MA / MB, this rule was much more strictly optimal. The few Multi-Mega teams which did exist prior were gimmicks that relied heavily on hidden information dynamics of their respective games, which especially held true for VGC where Open Team Sheets were not yet established as a tournament standard. Champions only saw this rule break down somewhat because of the sheer quality difference that existed between the Mega Evolutions and the standard Pokemon after the game removed a significant portion of available items, and even the Double Mega teams which do exist have tended to function as teams with separate modes where players make a decision in Team Preview on which Mega to bring but not an in battle decision around what Pokemon of two to actually Mega Evolve with.*** Teams that don't optimize away their battle to battle internal Mega Evolution conflicts end up suboptimal at absolute best. In Constructed Formats, engaging with this difficulty in any capacity is an unnecessary ankle weight.
In Draft though. Maybe you can get away with this. So I tried to prove it.
Yep Hidden in this Massive Post is a Mini Draft Team Report! Psyche! This was just my Social Experiment to get you to engage with VGC Draft Gottem Ggs
...Right. So this was an Champions Regulation MA VGC Draft. Each player was allowed to draft 8 Pokemon hosted by the Unown Draft League (UDL). There were 8 teams drafting across 13 Pools, so competition for individual Pokemon wasn't super high. The draft budget was 95 points with 20 point top tiers and the pricing for Pokemon was super favorable to Top Picks as they were barely more expensive than comparable High Tiers they blow out of the water.****
My primary goal entering this Draft was to maximize my engagement with inter-team Mega Evolution conflicts while still having a competitive team. This notably does not mean I wanted to only maximize the quantity of drafted Mega Evolutions - not only is having access to multiple Pokemon that can hold items like Choice Scarf or Focus Sash pretty important for making a team "good" at a baseline, I also didn't want to do a strategy like taking one Top Tier Mega Evolution that would be a focus for my team and then a bunch of cheap picks to hold decorative Mega Stones.
This is what I ended up on. (For those with no image; Sneasler, Dragapult, Mega Scovillain, Mega Gyarados, Mega Meowstic (M or F), Mega Aggron, Krookodile, Rotom-Fan) Meowstic in this draft was allowed to start as either Gender, only being required to hold the Mega Stone.
The picks here are not in order of how I chose them, but in order of price. (except Krookodile was more expensive than Mega Aggron) I was the only player in this draft who took either Mega Scovillain or Mega Meowstic, and I was also the only player to take more than two Mega Evolutions. Mega Aggron and Rotom-Fan in particular were absolute steals for their value. Krookodile was an early pick up I am very glad I had with how uncommon Ground types were, though fitting in on teams was a fair bit more difficult than I would have liked.
I had a team that at least had three flexible Mega Evolutions. Scovillain and Meowstic have relevant supportive functions they can take on without Mega Evolving in Prankster Screens and Rage Powder, Gyarados has Intimidate and enough raw stats to work decently as a non-Mega pick, and while Aggron definitely does want to Mega Evolve almost every time it hits the field it at least makes very good use of support from all three non-Mega picks into it's poor matchups. Meanwhile Sneasler was absolutely the glue that held the team together, utilizing sets with tons of investment into bulk to tank hits unreasonably well and clear the path for the others. I brought every Mega Evolution multiple times, even if actually Mega Evolving Meowstic-F was a bit of a forced ordeal that never quite worked out as planned, and was overall extremely happy with how versatile it felt to play. The team's matchups into standard Top Tier Mega teams was somewhat shaky, but options like Mega Aggron, Prankster Rain Dance, and Krookodile helped a lot into the less optimized Floette, Zard Y, and Gengar teams while the two runaway second string Megas in Delphox and Froslass both have a bit of an uphill battle into Mega Gyarados.
As for the result, I eked out a Top Cut in the Battle Pool stage with one of the most absurdly lucky and exciting victories I have ever gotten in a Draft game [link] before taking the team to T5th Overall. (Top Cut was Double Elimination so 5/6 tie) For an initial pool of ~100 players, that result feels OK. The Draft felt like a liability in two of my four set losses, but both of those losses were to the same player whose team absolutely bricked mine in addition to being generally very strong. (I also lowkey told him the exact adaptation I'd make against his team when I lost in pools and then mindgamed myself out of bringing it to the rematch where he seemingly forgot to account for it anyway.)
Playing with these team seriously almost flipped this entire post. It was a ton of fun to build and play with a team that had so much power potential and had so many avenues of play. I considered scrapping this entire post briefly. (Partially because I literally didn't expect to make it so far with this Draft and preparing for those games took up time I would have spent making it) But the more I thought about it the more I realized how I would probably never have a draft be shaped like this and actually be good again without double dipping on some of these exact same pieces.
Anyway I'm not too keen on keeping this yapfest in the oven lest it crumble to ash. There's still a fair bit I haven't gotten to with this. I didn't even get to talking about Diancie as a design and why I like it so much. But oh well I'll come back to this in another month or so.
---
*I'm talking about Mana / Color / Energy type stuff here more than archetypes.
**In this framework I am narrowing my scope to Player vs Player games. So this does not include, for instance, Nuzlocke Challenges, Battle Factory, fangame Rogue Likes, and so on. This tangent is already long and I can't be reaching this wide with my point but be assured that I do still mostly think the rest of my thoughts on Mega Evolution apply to these contexts.
***And I only really soften that from an "always" because I have seen some teams recently (ie between the time I initially wrote this and when I am revising it) utilize select Pokemon as secondary Mega Evolutions which shine when the primary Mega is benched but still see play in the main mode involving them. Aerodactyl on Charizard teams has done this throughout Reg MA, and a draft pal of mine also made a team (report here) for Reg MB using Blastoise with Blastoisinite and two Mega Launcher moves that doubles as a Fake Out support to set a hard Trick Room mode for Mega Camerupt. His team had a crazy tournament run in Skraw's Encore event so check it out!
**** For example, Charizard Y and Mega Floette were 20 pts to Torkoal and Sylveon's 14. Expectedly, Charizard Y and Mega Floette were picked in 11/13 and 13/13 Pools respectively while Torkoal and Sylveon were collectively picked only 5 times across all pools. (significantly lower than anything else at their cost) The thing with having a hard cost cap of 20 even for Top Tiers that could blatantly cost 5 points extra just seems to be a common thing with Drafts hosted by folks adjacent to Smogon. I guess it's to make the final drafts comparable in power to teams being played in the constructed format to let folks build ~80% of a constructed team in their draft? Whatever the reason, the result is that the Drafts orbit extremely heavily around the handful of defining Top Tiers and the cost effective ways to enable / counter them and exactly them. An attempt to actually optimize your draft (read; not what I did) would see you aggressively picking top tiers out the gate just to ensure your opponents lose access to broken combinations of them. It also means not introducing inter-team Mega Evolution competition in this case, as the Top Tier trio (the third is Gengar) all rely heavily on their Mega Evolution to actually do anything and as a result the only Mega Evolutions in the draft below 18 points which were picked more than 3 times were Delphox / Dragonite (new toy syndrome + good MU into top picks) and Aggron / Crabominable (optimized low cost Mega options for a TR mode to supplement fast mode) while most weren't picked even once. I have far more thoughts beyond this but the tangent is on long enough.
