Which Attack Types Are Best In Standard, Underused And Ubers?

[X-Act]

Yeah, I know I had already made such a thread before, but now I actually have more reasons to do it:

1) Doug's statistics allow me to know exactly how many Bronzong are Heat Proof and how many are Levitate, for example. This gives me a much more exact calculation.
2) I now have access to Underused statistics as well.

The following chart answers the question: "How much damage would a 70 base power move of the listed type deal on average in the Standard metagame?" (including Garchomp and Deoxys-S):

Physical Ice       10.30%
Special Ice        10.25%
Special Fire        9.53%
Physical Fire       9.40%
Physical Rock       9.21%
Physical Flying     9.20%
Physical Dark       8.99%
Physical Electric   8.93%
Special Rock        8.79%
Special Flying      8.77%
Physical Fighting   8.72%
Physical Ghost      8.64%
Physical Water      8.52%
Special Dark        8.47%
Special Water       8.45%
Special Electric    8.38%
Special Ghost       8.38%
Special Fighting    8.30%
Physical Dragon     8.20%
Physical Psychic    8.05%
Physical Ground     8.03%
Physical Bug        7.92%
Special Dragon      7.81%
Special Ground      7.81%
Physical Grass      7.69%
Special Bug         7.53%
Special Psychic     7.50%
Special Grass       7.38%
Physical Steel      7.13%
Special Steel       6.81%
Physical Normal     6.62%
Special Normal      6.27%
Physical Poison     6.00%
Special Poison      5.46%

Nothing new here. Ice, Fire, Rock and Flying moves, in that order, are still top-notch, and Steel, Normal and Poison moves are still the worst types you can use.

But what about in UU?

Physical Rock      10.50%
Special Ice        10.47%
Physical Ice       10.42%
Physical Flying    10.40%
Special Rock       10.31%
Special Grass      10.25%
Special Flying     10.13%
Special Fighting   10.03%
Physical Psychic    9.99%
Special Ground      9.88%
Special Fire        9.81%
Special Psychic     9.74%
Physical Grass      9.73%
Special Water       9.45%
Physical Fire       9.40%
Physical Dark       9.39%
Physical Ground     9.25%
Physical Fighting   9.20%
Special Dark        9.19%
Physical Electric   9.17%
Special Electric    9.13%
Physical Water      9.06%
Physical Dragon     8.83%
Physical Ghost      8.77%
Special Dragon      8.75%
Special Ghost       8.48%
Physical Bug        8.45%
Special Bug         8.45%
Physical Steel      8.09%
Special Steel       8.06%
Physical Poison     7.62%
Physical Normal     7.62%
Special Normal      7.50%
Special Poison      7.40%

Notice how the percentages here are nearer to each other and higher overall. UU has a plethora of types that do well, in particular Rock, Ice, Flying, Grass and Psychic. Here, Steel, Normal and Poison types are still the worst types, as in the Standard metagame.

And ubers?

Physical Dark      11.05%
Physical Ice       10.85%
Physical Bug       10.79%
Physical Ghost     10.62%
Special Ice        10.42%
Special Dark       10.35%
Special Ghost      10.24%
Special Bug        10.13%
Physical Dragon     9.65%
Physical Rock       9.52%
Special Dragon      8.92%
Special Rock        8.81%
Physical Flying     8.80%
Physical Electric   8.41%
Special Fire        8.35%
Physical Fire       8.34%
Special Flying      8.20%
Physical Grass      7.88%
Physical Fighting   7.78%
Special Electric    7.57%
Physical Normal     7.50%
Physical Ground     7.44%
Special Grass       7.40%
Physical Water      7.27%
Special Ground      7.24%
Physical Steel      7.15%
Special Fighting    7.14%
Special Water       7.14%
Physical Poison     6.94%
Special Normal      6.92%
Special Steel       6.73%
Special Poison      6.18%
Physical Psychic    5.83%
Special Psychic     5.29%

Dark, Ice, Bug and Ghost moves are top-notch in ubers, while Psychic and Poison moves are the worst. This isn't really surprising because of the amount of Dragon and Psychic types overwhelming the metagame. The damage range here is even more accentuated than in OU, too.

 

[Caelum]

Nothing too surprising, some of them were obvious like Ice in OU and dark in ubers or that poison type attacks weren't that great (in terms of coverage) all around. I must admit I was a bit surprised to find Rock #1 in UU (although now that I think about it, not really). Nice lists.

 

[Winto]

Do these statistics take into account the usage of Pokemon in these tiers?

I think statistics in UU might be a bit skewed by the sheer number of Bugs and Bug/Flyings present in the tier... when in fact most of them are probably more NU than anything. With the plethora of resists dominating UU (Steelix, Claydol, Hitmontop) it does seem a bit surprising to see Rock as #1.

Other than that though, it's worthy of note that seven of the top ten in UU are Special typed. I don't know if these statistics have taken Weezing's entry into the tier in account yet, otherwise an even larger shift could be in the future.

Interesting charts. :)

 

[Taylor]

This looks like an interesting read. Thanks a lot, X-Act.

 

[ParanoidDrone]

When do you sleep, seriously? I lost count of how many of these kind of topics you've made. :P

I must express my fears that since you assumed constant base power, the list fails to take into account the fact that not all types are made equal when it comes to the power of their most common moves, and thus will not accurately reflect how much damage one can expect to do with, say, Meteor Mash.

 

[Wulfspade]

I was wondering about this not too long ago actually. I think it should be pretty useful. Thanks X-Act!

 

[matty]

Interesting; hard evidence for BoltBeam usage as well if you think about it.

I'm most surprised at Fire and Flying tbh

 

[AJC]

Physcial ice is really good...so good theres probably a reason why there isn't a stronger no drawbacks Physcial ice move than ice punch (avalance doesn't count)

 

[ultimifier]

because nintendo knew that physical ice would be the most powerful average type on smogons server... yes

 

[Soma]

Hmm this could be useful when deciding on moves for certain sets and things like that when you're going for coverage. Thanks X-Act.

 

[McLovinn]

Thanks X-Act. But I don't see many Physical Ice types more, certainly not more than Special Ice - the only ones I can think of atm are, Ice Shard and Icicle Spear. Lol.

 

[Stathakis]

Thanks X-Act. But I don't see many Physical Ice types more, certainly not more than Special Ice - the only ones I can think of atm are, Ice Shard and Icicle Spear. Lol.

yeah, ice punch, avalanche, and ice fang are so unviable. who would ever use those?

 

[gorm]

When do you sleep, seriously? I lost count of how many of these kind of topics you've made. :P

I must express my fears that since you assumed constant base power, the list fails to take into account the fact that not all types are made equal when it comes to the power of their most common moves, and thus will not accurately reflect how much damage one can expect to do with, say, Meteor Mash.

multiply by 70/base power then lol

 

[McLovinn]

Stathakis; I'm feeling dead, and my brain wasn't working :p. Forgot about them.

 

[SMZ]

Flying surprised me a bit, but otherwise either obvious or a bit under obvious. Yet again nice job X-Act.

 

[thecrazylazyguy]

Thats pretty interesting information. This will definitely help me out I am pretty rusty at team building

 

[Cynthia]

Rock is probably so high in UU due to Ninetales being so common, also Froslass/Lapras and more rock weak pokes in general.

Flying is actually a pretty good attack type and being strong against fighting types is a big boon. Most Pokemon just have poor flying STAB.

 

[Bourbon]

Maybe it's time to throw Drill Peck back on Xatu... and HP Flying on Nasty Plot Croak. Thanks again X-Act, you never disappoint!

 

[X-Act]

Now the last thing I have to implement is give all Pokemon their proper defensive EVs as per Doug's statistics.

It would need a herculean effort from me to wade through all that data though... but that's why I'll write a program that does it for me! With the program, it would be easy to update the excel sheet I'm creating when other statistics are generated every month.

"Make computers work FOR you... not against you." :)

 

[billymills]

X-Act, is it really necessary to go through each pokemon's possible permutations?
Some are never used. And it would provide negligible difference from the current statistics. IIRC doug's statistics include all types of battles (random included).

But nice guide, i can see it being used for single attack pokemon (blissey, and other walls) and pokemon such as giratina which can run either physical or special spreads. Otherwise, type coverage and stab are more important.

 

[ODDish]

I think statistics in UU might be a bit skewed by the sheer number of Bugs and Bug/Flyings present in the tier... when in fact most of them are probably more NU than anything. With the plethora of resists dominating UU (Steelix, Claydol, Hitmontop) it does seem a bit surprising to see Rock as #1.

Well of the top 39 UU pokemon

Swellow, Ninetales, Froslass, Altaria, Scyther, Drifblim, Venomoth, Lapras, Glaceon and Rapidsh are all vulnerable so its not that surprising really ...

Flying is actually a pretty good attack type and being strong against fighting types is a big boon.

Not to mention the three big grass types (Leafeon, Vileplume, Meganium) and the bugs (Scyther, Venomoth).

 

[X-Act]

X-Act, is it really necessary to go through each pokemon's possible permutations?
Some are never used. And it would provide negligible difference from the current statistics. IIRC doug's statistics include all types of battles (random included).

But nice guide, i can see it being used for single attack pokemon (blissey, and other walls) and pokemon such as giratina which can run either physical or special spreads. Otherwise, type coverage and stab are more important.

It's important to collect data for all Pokemon, yes. The usages of each Pokemon will then take into account the weighting of how much that Pokemon's information is important. If Sandshrew was never used, the damage done to it would be multiplied by zero, whereas if Garchomp was used 5.1% of the time, its damage dealt would be multiplied by 0.051. But maybe Sandshrew WILL be used in August, so it's important to take care of every Pokemon.

 

[Darkflagrance]

Wow! If it was possible for someone to earn a PhD in pokemon, I'm sure you'd still be ridiculously over-qualified.

Perhaps people seeking to [radically] change a metagame (perhaps along the lines of AA's or Captain's metagames) could use this information to decide to strengthen or add moves of certain types or change type match-ups to make moves more equal.

 

[billymills]

Yes, i understand weighting, but that wasn't my point (It has to do with non-existant EV spreads, but i can't explain it very well)

Doug's statistics have the problem that they do not record what type of match they were used in. If you were going to weigh the EV spreads, it would be a mix of all metagames and would flaw the current system. If the next statistics are metagame restricted, then it would work better.

Ie: if there are 24 bronzongs total: 20 in OU, 4 in random

of the 20 OU:
19 levitate
1 heatproof

of the 4 random:
2 levitate
2 heatproof

Therefore in the statistics there are 12.5% heatproof. however, there are only actually 5%.
Ground seems (albeit incredibly negligibly) better, and fire seems slightly worse.

 

[X-Act]

Yeah, but what can I do about that? Can I do better? If I can't, why complain?