Theorycrafting! (Because I like competitive metagames too much)

[Grefter]

For FPS if you are interested in skill vs tactics and how it all interplays, you can't go past Counter Strike, there is a reason it was so ridiculously successful.  Can't find a super good video of a live match easilly (Don't know the players or teams to look for anymore...)  but http://www.youtube.com/watch?v=6tu__AtSJtM&feature=related this certainly is all about tactics and gives you a good cross section of skill as well.

Of course as always, you cannot go past Quake 3 if you want pure skill twitch gaming that impresses.

The DL actually has some Painkiller fans, but uh as one of them, the multiplayer having a following honestly suprises me.  It is a good oldschool shooter, but what its multi brings is purely inferior to Quake 3.

Edit - Still not great, but a good idea of where to start compared to the other one.

http://www.youtube.com/watch?v=dcNPTyw5KQk

Team 3D vs SK Gaming.  Both pretty insanely huge in FPS scene in general.

[metroid composite]

Yeah, I'll be honest that I don't know much about CS specifically (other than the basics--no respawning, plant/diffuse bomb, use low mouse sensitivity for sniping (1-2 range), buy weapons, everyone runs faster with a knife, boom headshot).  Specifically, I wanted to write something about why 1.6 is still more popular than Source (as that seems like exactly the kind of example that would be useful) but I realized I didn't actually know what I was talking about.

(Reading some random forum...Source has crate physics and larger hitboxes, and 1.6 has more exploitable bugs and more recoil.  So...I guess 1.6 is the more twitch-skill game?)

[Grefter]

That I can't really tell, I am not into CS enough to know how it is playing more than kind of tangentially.  My exposure is that 1.6 is awesome and Source is shit because everything is wrong  (The hit boxes and the recoil differences are likely a large part of this).

They are both very twitchy with good players being able to spray down whole teams in matters of seconds.  At a really high level of play though there is a pretty crazy level of tactics and skill being put into play to get yourself in a position to do that.

Part of why 1.6 is still more popular than Source is really just because it came first.  CS had been sitting at 1.6 for like 3 or 4 years between when it was "finished" and when CS:S came out.  The journey to 1.6 was a pretty long one as well and CS had been dominating as a tournament FPS since before I started playing (Beta 6 from memory?  Just a bit before they removed the scope from the M4 and actually decided to balance T and CT weapons), so yeah people are fairly attached to it.  I guess it is the Smash to Brawl jump or something played out over a much more granual set of changes (It isn't here is a pile of bug fixes and some bullshit exploits changed, it is different hit boxes and everything that looks pretty much the same just not quite functioning in the ways you expect them to).

But yeah, the point was that the game that was far and away the most mind bogglingly successful tournament wise in FPS was huge on both factors of twitch and tactics.

[metroid composite]

I'm posting from a cell phone, and thus may randomly cut this post short but...

A conversation I had recently: why does species rule exist in competitive Pokemon?

Having had the question circulating in my mind, it occurs to me that this is a common rule in a lot of competitive setups.  In TF2 you can only use 2 of each class at most.  In MtG you can only use 4 copies of any one card.  In most Fft Tournaments you can only use one copy of each skill/class on your team.

Some games have no such rules.  Wolfenstein Enemy Territory does not (and competitive play tends to be 5x Medic and 1x Engineer).  Hell, TF2 hasn't always had the restriction (and used to be something like 3x Medic, 2x Soldier, 1x Demoman).  Some games have a restriction clearly for balance purposes only, like Starcraft where double-zerg is not legal in 2v2.

But it seems to me there's more than just balance concern in such species restrictions.  One of the things I heard cited for TF2 (when discussing only Medics being restricted) was "swinginess", which is to say, one team overloading on one class would crush certain setups, and crumple to others.  At this point you have a guessing game of rock paper scissors where player skill in-battle is less important.  The next argument is just a general increase in variety.  When you have to be creative to fill your last two slots rather than just throwing in a couple more of your allstars, you get a bit more team variety.  An argument I've heard from a Magic the Gathering developer is that it's just a good safeguard: okay, so some card may end up stupidly good, but there will still be variety in decks, because deck designers have another 8 cards to pick.

As it happens, analogs to the species rule pop up in game design too.  Pokemon lets you learn 4 attacks, with no duplicates.  FFX lets you set four abilities on a weapon/armour with no duplicates.  And, on the other hand, there are games that do the oppisite: a lot of Starcraft strategies involve attacking with masses of one unit when your opponent isn't expecting that unit (a notable difference with Starcraft being that scouting the opponent and changing your build accordingly is a huge part of the game...whereas information scouting and setup adjustments aren't built-in to most games).

[metroid composite]

So...going back to a previous discussion, I made the statement that Warcraft 3 is more designed around micro than Starcraft.  While that's probably true about being more designed around micro, here's an interesting video that caught my attention:

http://kotaku.com/5580080/korean-gamers-are-faster-than-a-speeding-bullet

In the interview, the korean pro says that you don't have to be as fast APM-wise for WC3 as you do for SC.

I can think of a few reasons why this might be: even though WC3 units require more individual attention, you can have more units in SC, so it might just be a numbers issue.  SC units also have less HP in general making flanking more important and a mis-step more costly.

[DjinnAndTonic]

I was hoping to hear more from you on the species restrictions rules.

It seemed interesting, but you really didn't come to a conclusion about their purposes or implementations in various game genres.

[Dark Holy Elf]

In Pokemon, the species rule always felt designed by the same people who didn't ban ubers; the best way to control Mewtwo without banning him outright is to ensure each team only has one, thus minimising his overcentralising effect on the game.

The only other reason I can think to have the rule continue is to avoid situations where players can mislead others by making it unclear if this Garchomp that has just been switched in is the same as one that has already seen play. It'd be cruel and rather annoying misdirection for a player to fool someone into thinking that this is a Swords Dancing Garchomp and then BAM, Scarfchomp kills Weavile before he can move.

[metroid composite]

I was hoping to hear more from you on the species restrictions rules.

It seemed interesting, but you really didn't come to a conclusion about their purposes or implementations in various game genres.

Sure.

My main conclusions on their purpose are buried in the third paragraph, namely:
1. As a metagame safeguard, in case something is unexpextedly overpowered (to prevent, say, everyone using Magneto/Magneto/Magneto teams in MvC2).
2. To keep matches from being decided before they start when teams are so extreme it becomes rock-paper-scissors.
3. To increase variety a little; instead of "now that I have my four core team-members...I'll round out the team with two more of that guy" you have to stretch a bit more.

To elaborate a little on the rock-paper-scissors: when used right it's a construct that's very, very good for competitive games; good pokemon matches have prediction mindgames going on every single turn.  Fighting games usually have something built-in like block>punch>grab>block.  What doesn't work well, however, is when a single decision is too important, especially if it's a single decision before the match even starts.

For example, it would be bad in Starcraft if, say Protoss always beat Zerg at high level.  However, Protoss going DT-Sair always beating Zerg going Ultra-Ling?  That's fine, especially since Zerg can change their strategy mid-fight (and maybe catch back up).

In the case of single player games, I think there's an added benefit that it opens up game design options.  Imagine if, in FFT, you could stack copies of Magic Attack Up in your movement, support, secondary, and reaction slots.  While 4xMAU may or may not be an optimal setup, there's no question it would break things.  The fact that this isn't possible allows the designers to worry a little less when making stackable effects.  (You see this a lot in MtG).

I'm starting to tangent here, but FFT doesn't use a species rule per-se (WAXF is the FFT-style game that did).  FFT instead used a slot system; my kneejerk is that WAXF has it right and FFT has it wrong, though that may just be a kneejerk negative reaction to equipment in most JRPGs ("man, what am I going to equip on my left elbow?  The elbow pad with 8 defence or the elbow pad with 10 defence???")  Mathematically the number of possible setups goes doen a lot with a fixed-slot system, though.  Like, let's say there's 4 slots and 16 pokemon that can go into those slots.  Species rule gives you 16*15*14*13/1/2/3/4 = 2*5*14*13 = 1820.  Fixed slot rule gives you 4*4*4*4 = 256.  No-rule gives you 1820 + 16*15*14/3/4 + smaller terms = 1820 + 4*3*13 + smaler terms = 1820 + 156.  ...Or in other words, the difference between fixed-slots and species-rule is about 1000%, a huge increase in theoretical combinations.  The difference between no species rule and species rule is small; maybe 10% of theoretical combinations.

Since I'm on this tangent anyway, looping back to the things I listed species rule doing well--like the fact that it's less of a problem if one particular piece is overpowered...fixed slot rule doesn't handle this nearly as gracefully.  If a support ability is overpowered in FFT, for instance, everything else in that slot just automatically won't see play.  One hypothetical advantage of fixed-slot is that it may be less rock-paper-scissors swingy; like...inagine competitive pokemon where there's exactly one "steel slot" on each team; this would make the metagame less "about steel".  However, fixed-slot would probably ultimately fail to produce more metagame diversity, due to the sheer mathematics of so many fewer teams, on top of the less graceful handling of overpowered options.

So...yeah, goooo tangents (still posting from a cellphone; reordering paragraphs into logical essay orders doesn't happen).  I hope I answered your question somewhere in there; feel free to pressure me on specific points if you want to hear more about them.

[metroid composite]

Starcraft 2 (Or: how some Zerg units look kinda weak compared to other races...)

Now, don't get me wrong here--I'm not implying Zerg is necessarily weak as a race (maybe they have better macro than other races--that was certainly the case in SC1).  But looking at some of these units cost-for cost compared to similar units in other races...

Let's start out with the relatively solid...

[c]Roaches[/c]

Roaches aren't bad.  100 cost 2 supply for a ~150 HP 1 Defence unit that can only attack ground.  The obvious equivalent unit in Protoss is the Zealot, which attacks 50% more often for the same damage, but is range 1 instead of range 3.  There isn't an as obvious parallel for Terran, but Marauders kinda fit--20 less HP (40 less HP if they stim) and 25 extra cost are the Marauder downsides.  The upsides?  Marauders deal more damage to Light units if they stim...and over double damage to Armoured units if they stim (overall around twice as much damage on average, assuming about half of your opponent's forces are armoured).  Oh, and as an extra bonus, Marauders have 6 range instead of 3, and slow down opponent movement speed. >_>

So anyway, Roaches are decent; Marauders look better, but that's okay--Marauders are probably the best unit in the game so being worse than them isn't insulting.

[c]Hydralisks[/c]

Compared to stalkers...they cost about the same (stalkers are 17% more expensive).  Thanks to a higher attack rate they deal about twice as much damage to Light targets, but they have half the HP and no defence.  So far this is relatively even--slight advantage to Hydras even, thanks to a lower cost.  But Stalkers have more movement, 40% bonus damage against armoured targets, teleportation, and are much lower on the tech tree, while Hydralisks have...the ability to upgrade their range to 6 (the range Stalkers have by default).

So...comparing to Terran--there isn't an exact comparison, but Marines are very similar to Hydras--low HP units that have a fair bit of damage, but don't have any specific damage bonus.  They're also 1/3 the cost of Hydras.  Even without Stim or the HP upgrade, three marines kill one Hydra (only losing one Marine themselves).  Although to be fair: on the other end of the spectrum, technically five Hydras kill one Thor (equivalent cost), and yet in practice groups of Hydras don't do so well against equivalent cost groups Thors thanks to higher range and initial burst damage from the Thors.  And compared to Marines, Hydras have the range and burst advantage.  Then again, on the third hand, if you mix in Stim and HP upgrades for the Marine, then on paper two Marines take out one Hydra (when three Marines is the equivalent cost), so I'd hazard a guess that Marines probably are efficient against Hydras.  And Marines do come much earlier on the tech tree.

Corruptor

Remember the good ol' days of Brood War, when Acid Spores had splash effect, and let you deal up to 10 extra damage per hit?  Now Corruption is a 75 energy single-target spell that lets you deal 20% more damage...which for most Zerg units means 2 extra damage per hit.  Okay, well...maybe Corruptors have good combat stats?  Well...when one of their uses is apparently countering Vikings, a less expensive unit which deals 24 damage per second to corruptors (compared to the 7 DPS corruptors deal back; maybe close to 10 DPS if they use the corruption spell) and the HP gap is only 200 to 125.  Yeah, corruptors soundly lose the 1v1 against a cheaper unit that liquipedia suggests they are used to counter, even assuming they cast a 75 energy spell to help.  Though I'll give them some credit: Corruptors do handle Phoenixes.  And uncharged Void Rays (for all that Corruptors killing slowly means void rays might have time to reach full charge).



Granted, a lot of the power of Zerg arguably comes from units that don't have a direct analogue in other races (Zerglings, Mutalisks, Banelings).  And there's other comparisons I could make, but they're a bit of a stretch (yeah, Banshees horribly outdamage Brood Lords, but arguably BLs are for spawning free meatshields that stop the enemy ground army from walking forward, and yeah, Ultralisks are probably overall weaker than Thors and Colossi, but they do move quite a bit faster and have more durability so it's not a totally cut and dried comparison).  Still though, some of these unit stats do make me raise an eyebrow....

[Laggy]

Stalker: 10 vs light, 14 vs armored with 1.44 cooldown = 6.94~ / 9.72~
Marine: 6 vs anything with 0.8608 cooldown (0.57387 stimmed) = 6.97~ / 10.46~
Hydralisk: 12 vs anything with 0.83 cooldown = 14.46~

In practice, hydras kill pretty much all Terran bio and Protoss gateway units on the ground barring caster use when going against them. They cleanly outclass stalkers (over twice the damage output vs them, same range, Blink doesn't really help) and kill zealots fine once they've reached a semi-respectable critical mass. Against Terran, stimmed marines (even with the shield upgrade) are taking a hit before they get in range and die to 3 hits, while killing in 8. Even though in one-on-one the average marine is getting 4-5 shots off before dying, in practice it's very hard to get a proper surround (you need a wide open field and to be able to circle the hydras quickly), and that eats up valuable time when you're taking considerably more damage than you're doing. They are probably the most efficient unit to fight hydras, but actually still generally lose to them (and infestors vs medivac for support units is -cleanly- in favor of the infestor because Fungal is that amazing). It is also a rare situation where marauders are cost-inefficient - against ANY OTHER UNIT they still have use (the target is either armored or melee, i.e. zerglings, roaches, zealots, stalkers, even immortals/colossi) but hydras have the rare distinction of avoiding both weaknesses.

Hydras have mobility issues where they aren't easy to move around off creep and are later tier than the aforementioned units, but at appreciable army sizes they do quite soundly beat out other ground forces that don't have splash. The general accepted counter to them is either tanks (and not so much with the advent of the latest patch), colossi, or storm.

Corruptors are extremely niche but serve well in their one niche: they counter colossi, as they can effectively snipe them and not care about stalkers shooting at them thanks to having so much HP. No one uses Corruption really and it is a spell badly in need of overhaul. They do not even remotely counter vikings, Liquipedia is full of it (vikings are the best AA unit in the game), though they do counter BCs should it ever be relevant. Hydras are actually a better counter to void ray (relevant) and phoenix (not so much). Hell, QUEENS are better to counter void rays in the context of most games.

[Yoshiken]

Ooh, wonder if you could do a specific analysis. Since you looked at why Garchomp was moved from OU -> Ubers, any notes on why Salamence was? I still can't quite get my head around that one. >.>

[metroid composite]

They cleanly outclass stalkers (over twice the damage output vs them, same range, Blink doesn't really help)

Do they?  Sure, twice the damage output, but half the durability should make for a relatively even fight.  Movement speed and blink are relevant for stuff like chasing down and killing a smaller force, or running from a larger force.  (And blink can abuse cliffs and avoid walking in single file).

Although sure: Stalkers are not built for taking out Hydras; it just strikes me that they have a relatively even matchup despite Hydras very much not being the kind of enemy they're built to fight.

In practice, hydras kill pretty much all Terran bio and Protoss gateway units on the ground barring caster use when going against them.

So...counting out all casters that leaves...six units?  Most of which win in some scenarios (like Dark Templar and Reapers)?

Hydras have mobility issues where they aren't easy to move around off creep and are later tier than the aforementioned units, but at appreciable army sizes they do quite soundly beat out other ground forces that don't have splash. The general accepted counter to them is either tanks (and not so much with the advent of the latest patch), colossi, or storm.

Well yes--the way to kill masses of low HP units is units is to outrange and splash them.  I'm pretty sure the list of Marine counters would also be Storm, Tanks, and Colossi.

Ooh, wonder if you could do a specific analysis. Since you looked at why Garchomp was moved from OU -> Ubers, any notes on why Salamence was? I still can't quite get my head around that one. >.>

I might.  Actually, what happened is that I was planning to write about a lot more SC2, and then Pokemon BW started getting revealed all over the place, so I'll probably end up writing about my expectations for new bannings from BW.

Briefly though: Salamence does fit into the pattern of other 600 BST pokemon that were moved to uber--which is to say high speed, and high damage.  Now if you want speed and damage on a 600 BST pokemon you do like...normal Shamin...complete with that 100 special (but hey: Seed Bomb--it's shiny).  If you want an actually good attack stat on a 600 BST pokemon, your fastest option is...Dragonite, at 80 base speed stat.  Not that speed+damage are the only reasons for banning, but now that Latias and Salmence are in uber there is a certain amount of consistency.  (The more in-depth reasons also have something along the lines of "Salamence's intimidate can force a switch, and Salamence's type coverage of Fire/Earth/Dragon means there isn't a safe switch in").

[Laggy]

Uh, actually, by "caster units" I really mean "sentries and high templars" (and to some extent, ghosts). Reapers lose badly so badly to hydras it's not even funny (4.5 range vs 6 with 50 HP is a terrible combination). DTs aren't even comparable because they flat out lose if there's an Overseer (which is pretty trivial for Zerg to morph in) and are not a kind of unit you use as part of an army composition.

The actual number of units they're good against hardly matters. Marine/marauder/zealot/stalker is the backbone of nearly all army compositions in SC2 and come into play extremely often. Additionally hydras do fine vs a variety of units beyond that (hellions, immortals, basically anything in the air). Being able to handle those number of crucial units certainly makes them one of the best Zerg units on the field right now to me, considerably more than, say, roaches.

Marine counters do include storm, tanks, and colossi, but that list is also expanded to banelings (which hydras kill just fine due to their superior range and lack of need to stim), as well as chargelots.

Stalker/hydra is interesting. While you're right that the numbers in practice make it look close, I find in practice hydras have a decisive edge in engagements. It may be because it's so much easier to clump hydras together and shoot en masse whereas stalkers don't have that advantage (despite being faster, they are larger). If you want to test this out on some map at some point, I'd be more than happy to do so.

[metroid composite]

Marine/marauder/zealot/stalker is the backbone of nearly all army compositions in SC2 and come into play extremely often.

I thought Colossi (or in some cases Immortals) were supposed to be a significant part of Toss armies?  And High Templar late in the game.  (Unless you mean against random people on battle.net, in which case yes: Zealots/Stalkers/Void Rays >_>).

It may be because it's so much easier to clump hydras together and shoot en masse whereas stalkers don't have that advantage (despite being faster, they are larger).

Hm, yeah, that's a good point I didn't take into account.  (I wonder how the size varies between Marine and Hydralisk too--Marines were a lot smaller than Hydras in SC1, but I think marines are larger now).

If you want to test this out on some map at some point, I'd be more than happy to do so.

Sure.

[Laggy]

They're mid-game and late-game units. Both storm and colossi require a significant amount of time to reach, and while they certainly are major factors the main army composition is still going to be gateway units (unless they're rushing or their opponent sucks and lets you expand/mass without much issue). In particular, colossi and HT take roughly the same amount of time to reach (Twilight Council/Templar Archives + research Storm, or Robotics Facility/Robotics Support Bay + research Thermal Lance), and are massive gas investments.

A surprising number of games, whether in Bronze League or Diamond, end before tech even gets that far. Believe me, I can attest to that >_>

This is significantly later than hydra, which just requires Lair and a Hydralisk Den and is much more gas friendly. Incidentally, corruptors are often mixed into hydras just to deal with colossi. No race, on the other hand, has an easy answer to storm other than Ghost's EMP (some things don't change).

Basically, if you're going to pick on any Zerg unit, pick on the roach. Or zerg tier 1 in general - they basically have speedlings as a bread and butter option, with banelings as a niche choice, and.... I'm just far less impressed with roaches (die to marauders, die to stalkers - the two most popular units in the game for their respective races - die to immortals, die to any sort of air rush ala banshee/void ray, and even reapers and hellions can outrange and kill them with impressive micro, while sentries also get a field day because Force Field makes them cry horribly.) This is generally why they're regarded as underpowered early on since at professional levels of play there's often enough pressure that you don't have TIME to get to Lair while expanding (which, yes, Z still needs to be 1 base up to be competitive just like in SC1) and you almost have to use roaches to counter certain builds, but they really peter out quickly (matching Hydra food and being a generally inferior unit the longer the game goes on does not help).

[metroid composite]

They're mid-game and late-game units. Both storm and colossi require a significant amount of time to reach, and while they certainly are major factors the main army composition is still going to be gateway units (unless they're rushing or their opponent sucks and lets you expand/mass without much issue). In particular, colossi and HT take roughly the same amount of time to reach (Twilight Council/Templar Archives + research Storm, or Robotics Facility/Robotics Support Bay + research Thermal Lance), and are massive gas investments.

Colossi are still a lot earlier--they can do stuff without research (for all that you probably won't time your push before research, you can defend with them alright) and even after you train a templar and research storm it only spawns with 50 energy, so there's extra waiting there.

A surprising number of games, whether in Bronze League or Diamond, end before tech even gets that far. Believe me, I can attest to that >_>

Sure.  Often rushing with Zealots or Lings, cheesing with cannons or bunkers, getting hit with a unit that you're not prepared for (Dark Templar, Mutalisks).

Is there a timing push with Hydras, though?  (I know there was in SC1 versus toss, but they're tier 1 units in SC1, and even in SC1 it was possible to get defencive Storm up in time with scouting).

This is significantly later than hydra, which just requires Lair and a Hydralisk Den and is much more gas friendly.

Lair is 100 vespene.  So...maybe that starts around the same time as Robo (you could probably start Lair earlier if you delayed your Queen, but >_>).  Build time for Lair + Hydra Den is 120.  Build time for Robo + RoboBay is 130.  Hydras build in 33 seconds; a chrono-boosted Colossus builds in 37 seconds.  So...it's not much later for producing one Colossus (which will probably be backed up by Zealots to save gas).  This is dependent on scouting, though--in that you must scout "not Mutalisks".

Basically, if you're going to pick on any Zerg unit, pick on the roach. Or zerg tier 1 in general - they basically have speedlings as a bread and butter option, with banelings as a niche choice, and.... I'm just far less impressed with roaches (die to marauders, die to stalkers

Versus stalkers?  Roaches are like...half the cost, while actually being almost able to one-on-one...(ignoring range issues, obviously).  Not as good as Zealots are  against stalkers when Zealots get into range, but still decent (and get into range much easier).  Although sure: crazy micro can give Stalkers an advantage.

and you almost have to use roaches to counter certain builds, but they really peter out quickly (matching Hydra food and being a generally inferior unit the longer the game goes on does not help).

Well yes, I think that's most of why they exist.  They're one of the only heavy armor zerg units, and the only early range, which makes them the only sensible way to fight early Reapers, for instance.  They're one of the only durable zerg units, and if your entire army lacks durability, you open up vulnerabilities to stuff like storm.

Although sure: if other races had access to roaches, I doubt those races would use Roaches.  Roaches just fill a niche for Zerg.

[Laggy]

There's not really a timing push with hydras that I know of, no. They are a little late for that; they're more of a "if you've got solid macro, this is a great unit to mass up" option.

Colossi do ultimately come out faster than high templar, though bear in mind that aside from needing Thermal Lance to push out, they cost 300/200 a pop. Assuming fast tech, you can only support 2 gate / 1 robo producing colossi off 1 base and that's with significant unit cutting early on (so no expanding or anything) whereas Zerg pretty much auto-expands before Lair. You will also be gas blocked from constantly producing Colossi AND getting Lance at the same time, and you won't be able to build sentries beyond the mandatory 1 for your ramp.

On the flipside, hydras at 100/50 means you can pick up 6 to one colossi in terms of gas, and if Protoss is producing zealots as his mineral sink, that certainly works in Z's favor.

The main advantage of templar is not that you get them fast, but that you get to pick up useful upgrades on the way (Blink is cool, Charge fundamentally upgrades zealots almost as much as ling speed does).

Roach really doesn't beat stalker, though. Off creep, they're criminally slower and basically picked off by stalkers with zero issue (2.25 speed vs 2.95). Even on creep they're only marginally faster with 6 range vs 3. I mean, yes, it's not a hopeless cause since they cost less, but they are at a clear disadvantage (and they're equal food units).

One of the main reasons I disdain roaches so much is that almost all of their supposed niche uses can be circumvented with speedlings in most situations (speedlings kill stalkers fine, and while it takes 4 to take down 1 zealot that's still equal cost - and if P is going zealot heavy all of your other tech options look so much better, since ANYTHING zerg has kills zealots - roaches, hydras, mutas, banelings, etc.) In the TvZ matchup, you MUST get banelings to contend with Terran bio, so lings are going to be mandatory regardless and also do better against the MM ball than roaches do. If Terran mech was more prevalent I would agree with seeing more roach play, but... well, it's not, and will be even less so after the recent tank nerf.

(Admittedly, a lot of this anti-hype is because many Z I see aren't using roaches the way I feel they're intended to: you probably shouldn't build then en masse, but use them as meatshields in conjunction with lings, the same way Protoss utilizes chargelots or Terran utilizes marauders.)

As a side note: Zealots are actually horrible against stalkers. The firing animation is considerably faster than dragoons from SC1, so kiting is of minor issue (the shot goes off relatively fast).

[metroid composite]

(Admittedly, a lot of this anti-hype is because many Z I see aren't using roaches the way I feel they're intended to: you probably shouldn't build then en masse, but use them as meatshields in conjunction with lings, the same way Protoss utilizes chargelots or Terran utilizes marauders.)

Well yes.  Meatshields are useful when you're actually shielding something.  Not so useful when you're not shielding anything....  (Although, exception: a Korean game I saw a while back had the Toss going mass high templar, and Zerg using a mass of roaches against that army and just burrowing and +10 regenerating in response to storm).

As a side note: Zealots are actually horrible against stalkers. The firing animation is considerably faster than dragoons from SC1, so kiting is of minor issue (the shot goes off relatively fast).

They can do fine in an earlygame rush--have one Zealot chase after each Stalker, and the remaining Zealots kill Probes.  If the Stalkers go defend the probes, then the chasing Zealots catch up and kill.  This does require a rush build, though--obviously an opponent with equal numbers of stalkers to your Zealots would be bad.

Zealots also do fine against Stalkers in big mixed army collisions mid-late game.  In that case...sure you could micro the Stalkers, but it's more important to micro the Colossi.

Roach really doesn't beat stalker, though. Off creep, they're criminally slower and basically picked off by stalkers with zero issue (2.25 speed vs 2.95).

Roach speed can be upgraded to 3 >_>

Although sure, that upgrade is Lair only, and won't be part of a Roach rush.  (Roach rushes versus Toss have to operate more like Zealot rushes--make the stalkers run or corner them to the base walls, and then kill the probes with Lings).

[hinode]

and then Pokemon BW started getting revealed all over the place, so I'll probably end up writing about my expectations for new bannings from BW.

If you do theorymon on this, bear in mind that the tenatative plan is to start with everything unbanned and rebuild the ubers list from scratch.

[Yoshiken]

Doesn't surprise me, given how much the metagame is likely to change - can see Sand Stream suddenly moving even further up the tiers, thanks to the damn mole thing. >.>
It's funny how many people have complained about BW making the old Pokemon completely redundant, though - a few people have said that old Pokes won't even be worth using given the BW ones.

[metroid composite]

and then Pokemon BW started getting revealed all over the place, so I'll probably end up writing about my expectations for new bannings from BW.

If you do theorymon on this, bear in mind that the tenatative plan is to start with everything unbanned and rebuild the ubers list from scratch.

Well right--there is a question of "where do you draw the line".  If they end up with a line drawn in the same location as last game, though...




Starting with the top of the pokedex, as it should have the highest concentration of likely ubers....

http://www.serebii.net/pokedex-bw/649.shtml

600 BST Bug/Steel (one 4x weakness) with 99 speed, 120/120 Attack/Special (which get further boosted by the ability: Download).

Abilities: STAB U-Turn, Explosion, reasonable special STABs (90 Bug Buzz, 80 Flash Cannon), excellent all-round coverage (Flamethrower, Thunderbolt, Ice Beam, Energy Ball (80 power Grass), and "Techno Buster" an 85 power move it can make any type with the right held item).  Support includes Thunder Wave, double Screens, and Rock Polish.

Verdict: Hmm....lacks anything like Swords Dance or Dragon Dance that would just be...absurd; so it's only fast+high power+download.  Heatran resists much of the likely moves.  (Hidden Power Earth or Techno Buster Earth can threaten there).

Hmm...leaning towards yes: this would be Uber in the gen 4 Tier list.


http://www.serebii.net/pokedex-bw/648.shtml

600 BST Serene Grace pokemon with two forms.

Steppe Forme:
128 Attack, 128 Speed Normal Fighting with STAB Close Combat, STAB Return, has Cheer UP (Atk+1, SpAtk+1).  Type Coverage is...Shadow Claw, Stone Edge, U-Turn.

Serene Grace is...basically wasted on this pokemon.  No flinch move.  No big effects that I can see.  On the other hand...ouch.  And Shadow Claw makes Ghost switch-ins somewhat questionable.  On the third hand...Hippowdon is the current OU physical wall, and is only weak to ice, water, and grass, which aren't available to #648 without dipping into its 77 special.  On the third hand, Grass Knot is 120 power against Hippo's lowish special defence, and Cheer Up boosts it.

verdict: Honestly, doesn't use Serene Grace as well as Shaymin-S, but higher attacking stat, better typage (both on attack and defence) and more durable (90 defence instead of 75).  Yeah, uber.

Voice Forme:
Normal/Psychic with 128 SpAtk, 90 Speed, 128 SpDef

It's bulky, but only specially bulky.  Offence is...okay (STAB options are...Hyper Voice and Psychic.  Non-STAB options are Grass Knot, Thunderbolt, Shadow Ball, Focus Blast).  With that bulk it wants Recover and...doesn't have it.

verdict: Honestly...looks counterable.  No super effective move on Scizor (outside of Hidden Power Fire).  Pursuit.  Hit that bad defence stat super effective.  Countered.


http://www.serebii.net/pokedex-bw/647.shtml

580 BST Water/Fighting with "Heart of Justice" (Attack rises if hit with a dark type attack)

*snort* wow, lame ability and below 600 BST.  Okay, not sounding uber so far.


129 Sp Atk, 108 Speed

...oh well, I guess I should take a closer look anyway.


Surf+Focus Blast make good STABs.  Sword of Mystery gives it STAB physical damage off of its special stat.  And...that's pretty much it for special moves it can learn.  (Lots of physical moves, it can use with its 72 attack stat though!)  Support includes Calm Mind (and Swords Dance!!) but not, say, Recover.

So um...Starmie says "'sup.  I'm water/Psychic and resist all your special moves (outside of hidden power), can recover stuff anyway, plus hit you with super effective Psychic."

verdict: Not uber.


That's all for the moment....

[Taishyr]

* Taitoro looks in. Meloia/whatever can't start in Step form, has to transform in battle by using an attack (Ancient Song IIRC) according to the info I've been reading.

[metroid composite]

Oh, well, just a few more before getting back to work...

http://www.serebii.net/pokedex-bw/646.shtml

660 BST Ice Dragon with Pressure

95 speed.  130 in both attack stats.  Tanky.  No W4 (in fact, Ice resists Dragon's Fire weakness, and Dragon resists Ice's Fire weakness, so only four W2s).  Obviously nice STABs, but they overlap a little.  Both Ice and Dragon are stopped by Steel, but Focus Blast exists for that.  Nothing hugely outstanding in support moves.

verdict: uber.  Less automatic than you'd expect for a 660 BST pokemon, but y'know, 95 speed, 130 attack stat with good STABs would be suspicious on a 600 BST pokemon.


http://www.serebii.net/pokedex-bw/645.shtml

600 BST Ground Flying with Sand Strength (or in the Dream World: Encourage)

101 Speed.  125 Attack.  115 Special.  Swords Dance.  Explosion.  Rock Polish.

Sand Strength is 1.5x the power of Rock, Ground, and Steel moves under Sandstorm.  So...2.25x STAB Earthquake, and 1.5x Stone Edge.  Other random moves include U-Turn, Outrage, Hammer Arm (100 power fighting move, lowers speed).  For use with the high special stat there is Earth Power, Psychic, Focus Blast, Grass Knot.

So...it's basically like Garchomp right down to the Ice W4.  Somewhat less durable, but with mix/special setups possible.  Oh, and bonus Earthquake damage, and Explosion.

verdict: uber.


http://www.serebii.net/pokedex-bw/644.shtml

680 BST Electric Dragon with Terra Voltage (which means: can't be stopped by opponent abilities like Volt Absorb).

Oh look, 150 attack stat, excellent durability, 90 speed, and its own unique physical electric moves that are better than all other electric moves.  Oh, and Electric is an excellent type defensively too.

verdict: uber


http://www.serebii.net/pokedex-bw/643.shtml

680 BST Fire Dragon with Turbo Blaze

Same as above, but replace "physical" with "special" and "electric" with "fire".  Just to be cute, if you're playing a Sunny Day team it has Solarbeam!  Oh, also note that with Turbo Blaze it can hit every opponent with at least a neutral STAB (as Flash Fire is not an issue).

verdict: uber


http://www.serebii.net/pokedex-bw/642.shtml

580 BST Electric Flying with Mischevous Heart (non attacking moves have priority) or (Dream World: Competitive Spirit)

Two obvious comparisons jump to mind: stats are very similar to Azelf.  And naturally there is Zapdos being another 580 BST flying electric.

Priority Taunt.  Well then.  Priority Substitute.  I see.  Priority Nasty Plot.  Well...priority doesn't help here, but like Azelf this is nice.  Priority Toxic; not bad.  On the other hand, doesn't have Explosion or Stealth Rock unlike Azelf, and doesn't have Roost, unlike Zapdos.

Fantastic lead what with the Priority Taunt.  Kinda not the stats you want for Mischevous Heart, though (all speed and attack stats...but when taking advantage of Mischevous Heart speed and attack stats are useless).  Most of the Mischevous Heart tricks are dealt with by "attack the ****er, don't mess around" (like priority sub and priority taunt, and...hell, Nasty Plot too).  Azelf already proves that glass cannon 580 BST pokemon with Nasty Plot can be dealt with, and Azelf brings more cannon to the table with stuff like Explosion.

verdict: OU, but not uber.  Quite highly used in OU as a lead, probably (no Stealth Rocks itself, but can taunt everyone else).  Fundamentally can be dealt with, though (hell, Aerodactyl leads are pretty popular, outspeed, and hit super effectively with STAB).  Actually, I'm not liking the lack of a priority attacking move, as it does get outsped by Azelf and Aerodactyl, who will both have focus sashes and surely kill in two hits.


NINJA'D

* Taitoro looks in. Meloia/whatever can't start in Step form, has to transform in battle by using an attack (Ancient Song IIRC) according to the info I've been reading.

Oh, well, that's fairly awkward given how one of them is all special attack and one of them is all physical, and you have to waste a move slot and a turn (on a 75 power Normal special move with 10% chance of adding sleep...raised to 20% by serene grace) to get a transform to the big bad physical sweeper.  Probably OU as a whole, then.

[Dark Holy Elf]

660 BST Ice Dragon with Pressure

95 speed.  130 in both attack stats.  Tanky.  No W4 (in fact, Ice resists Dragon's Fire weakness, and Dragon resists Ice's Fire weakness, so only four W2s).  Obviously nice STABs, but they overlap a little.  Both Ice and Dragon are stopped by Steel, but Focus Blast exists for that.  Nothing hugely outstanding in support moves.

verdict: uber.  Less automatic than you'd expect for a 660 BST pokemon, but y'know, 95 speed, 130 attack stat with good STABs would be suspicious on a 600 BST pokemon.

I kinda suspect they'll auto-ban him for 660, but that really seems shockingly unimpressive for an uber. 660 isn't as high as it sounds once you realise the 130 atk is pretty much wasted (well, there will probably be some nasty surprise Outrage sets, but the physical coverage is quite poor otherwise, 100% walled by steel), and Focus Blast is never a good thing to have to mention as a possibility on a pokemon's set. Also weak to Stealth Rock, assuming that abomination has persisted unchanged. Why is the ice dragon the weak one?


Also WTF physical electric moves that don't suck? This will take some getting used to. I started reading the entry for the electric dragon (not even realising it was an uber at first) and was immediately thinking "Oh, MC, they tried electric with super-high atk in gen 4 (Luxray) and it bombed horribly then, this won't be any different~" until I actually went in and looked at some of those attacks. The 5-PP electric EQ clone was impressive enough, but a move with 110.5 average power? Seriously? That shatters the previous title of strongest drawback-free move (ignoring crits); heck, even RBY Blizzard fell a bit short of that.

Well, it's uber, guess they can do what they want. And yes, I noticed fire is the same after the fact, although special fire wasn't as hard up for attacks as physical electric (in particular, Fire Blast was the co-holder of the aforementioned strongest drawback-free attack title before).

Oh, power creep, you're so silly.

[metroid composite]

I kinda suspect they'll auto-ban him for 660, but that really seems shockingly unimpressive for an uber. 660 isn't as high as it sounds once you realise the 130 atk is pretty much wasted (well, there will probably be some nasty surprise Outrage sets, but the physical coverage is quite poor otherwise, 100% walled by steel)

I don't think you'd want to go pure physical, but mix-sets seem fine, and free you up to Draco Meteor indiscriminantly.  In addition to Outrage there's Dragon Claw, Shadow Claw, and Stone Edge, all of which have proven useable before.  Consider: would you think of this as an impressive possibly Uber pokemon if it was 600 BST with 70 atk instead of 130?  My gut instinct is "probably; that's borderline."

Also weak to Stealth Rock, assuming that abomination has persisted unchanged.

The bad news: yeah, looks unchanged.
The good news: none of the pokemon so far have it in their learn list (I've been checking), and some pokemon who could learn it in older games can now cannot learn barring trades with older gens (Azelf, for insance).

Well, it's uber, guess they can do what they want. And yes, I noticed fire is the same after the fact, although special fire wasn't as hard up for attacks as physical electric (in particular, Fire Blast was the co-holder of the aforementioned strongest drawback-free attack title before).

Yeah.  They just had to give it a higher burn rate than Fire Blast too.  What I don't like is that it creeps into Overheat territory--Maaaaybe if you have a choice item and plan to switch out next turn anyway you'd still go overheat, but 130 with 85% accuracy is damn close to 140 with 90% accuracy.