Pokemon Uranium

Full Version: Type Coverage Analysis Ver. 2
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2
When I released my initial analysis of type coverage in Uranium, I knew it had some issues. Luckily, Cataclyptic brought my attention to this thread, which addressed many of the problems I had with my own work as well as having some practical applications for building movesets. I won't bother you with the fine details of exactly how the math works, as it's explained pretty well by Ignus and X-Act before him. The TL;DR of it is that the spreadsheet takes the Pokemon of a metagame (preliminary OU list from our own poweroftiban) and looks at each one's type, ability, and defensive stats, then does a lot of summations in order to generate an overall score not only for each individual type, but also type combinations in movesets. Now, let's get into the data and how to use it, but first I'd like to thank X-Act, Ignus, and QxC4eva of the Smogon forums for doing the real leg work in this and allowing me to use their work to help advance the Uranium metagame.

Without further ado, I give you the Linearized Uranium OU Type Analysis. Feel free to look around.

Let's start with the more esoteric side so that those not reading all the way through can find the useful stuff at the bottom of the post:



Data

Table 1
[Image: ZL0ivRt.png]
Table 1 contains the coverage scores for movesets with two physical moves.

Table 2
[Image: BgEhtgd.png]
Table 2 contains coverage scores for movesets with two special moves.

Table 3
[Image: jLvfiay.png]
Table 3 contains an average of the values in Tables 1 and 2.

Table 4
[Image: 0xwJyHC.png]
Table 4 contains the PATE, SATE, and MATE for each type.



Results
"But Twisted," you ask, "look at all these tables full of big numbers! What could they possibly mean?"

Well, as intimidating as this all may look what you really need to know of it is actually pretty simple. The score is actually a special factor that lets you calculate the minimum damage of a given move against an "average" Pokemon in the metagame by the following formula:

Attacking Stat * Base Power * Other Modifiers (STAB, Abilities, etc.) = Average Minimum % Damage
Attacking Type Effectiveness
                                                
What this means is that the lower the ATE, the better coverage the type has. This gives us the following rankings for physical, special, and overall:

[Image: We5tg3g.png]          [Image: VbmKhoQ.png]          [Image: BBJgD2O.png]

As an example, let's take a look at poweroftibarn's SubPunch Empirilla with an Adament nature's Focus Punch:
361 * 150 * 1.5 = 88.5933%
91683  (Table 4)                       
Thus, unboosted Empirilla's Focus Punch does, on average, about 88% minimum damage (roughly 32% chance to OHKO). Does this mean that it deals 88% damage to any given Pokemon? No, but if you were to average out the minimum percent damage to everything in the OU meta, it should come out to that. Thus, it give you an idea of how much coverage the move has and around how much damage it'll do.

The dual charts are less directly practical, but have the same system of lower numbers being better. If you look at the chart, you can pick the row for the type your interested, and then pick the column of the other primary type on your Pokemon's moveset, and you get a number similar to one you'd get in the single type chart. This gives you an idea of how good the coverage of those moves together is. This allows for the creation of a coverage pairing ranking for each type, both physical and special:

Physical Next Best Coverage Type
[Image: LJpQ0cc.png]

Special Next Best Coverage Type
[Image: dNCAiCi.png]


To end, I'd just like to thank Ignus, et al. once again for making this all possible. I hope this was helpful!
Bloody hell, when I have more time, I think I will poor over the data you presented a bit more, and try to understand the greatness you brought to us. This will help guide makers who like to work numbers, rather than theoreticals (like me!)

You the hero we need and deserve~!
Why, thank you! The spreadsheet has usage built in as well, so it'll be even more accurate once we get usage statistics. I will, of course, update the tables here when that happens.
Oh yes, this is very nice Twisted. Thank you so much for these data, they'll help everyone out immensely!
I just got home from the roadtrip and I see this. Awesome. And not so awesome. XD I have yet to digest many things I have left unattended on forums when this came up. Looks like I will be spending more time on the forums today....

Thanks TL ^^ Putting this up on forums wasn't a bad idea, now wasn't it? :p

EDIT: Now that I've actually read through everything word by word (which im guilty of not doing as often as I should), I'm immensely grateful for you doing this for us, setting aside your busy schedule to meet up with the peeps of showdown to compile this organized mess for us. I thank you from the bottom of my heart ^^
Well, this is great!  I'm gonna have to take this into account in my analysis on Seikamater (it will hopefully be up some time soon)...that bug has so much dang viable coverage that most of the sets look like a bit of a mess, tbh...not much to do about it though.  At least the main STABs she's got can rather easily suffice on their own...

Oh, and I'll definitely be using this for future analyses too! This is much easier to look at than trying to simply judge what coverage might work best.

I'm not really surprised by Nuclear being best complimented by fighting, either. That combination can hit every mon in Uranium for super effective damage with the one exception of S51-A. And what does Nucleon have for that (and every non-dark dual typed nuclear mon as well)? Shadow Ball. Yeeeeaaaahhhhhh.....perfect super effective coverage is a scary thing...
(10-03-2016, 06:40 AM)Dragonstrike Wrote: [ -> ]Well, this is great!  I'm gonna have to take this into account in my analysis on Seikamater (it will hopefully be up some time soon)...that bug has so much dang viable coverage that most of the sets look like a bit of a mess, tbh...not much to do about it though.  At least the main STABs she's got can rather easily suffice on their own...

Oh, and I'll definitely be using this for future analyses too!  This is much easier to look at than trying to simply judge what coverage might work best.  

I'm not really surprised by Nuclear being best complimented by fighting, either.  That combination can hit every mon in Uranium for super effective damage with the one exception of S51-A.  And what does Nucleon have for that (and every non-dark dual typed nuclear mon as well)?  Shadow Ball.  Yeeeeaaaahhhhhh.....perfect super effective coverage is a scary thing...

Oh yeah, that's why we need to suspect Nucleon for sure. As for Seikamater, I think that a set of Quiver Dance, Ice Beam, Thunderbolt and Surf would be best. The chart backs me up on this.
(10-03-2016, 11:14 AM)Cataclyptic Wrote: [ -> ]
(10-03-2016, 06:40 AM)Dragonstrike Wrote: [ -> ]Well, this is great!  I'm gonna have to take this into account in my analysis on Seikamater (it will hopefully be up some time soon)...that bug has so much dang viable coverage that most of the sets look like a bit of a mess, tbh...not much to do about it though.  At least the main STABs she's got can rather easily suffice on their own...

Oh, and I'll definitely be using this for future analyses too!  This is much easier to look at than trying to simply judge what coverage might work best.  

I'm not really surprised by Nuclear being best complimented by fighting, either.  That combination can hit every mon in Uranium for super effective damage with the one exception of S51-A.  And what does Nucleon have for that (and every non-dark dual typed nuclear mon as well)?  Shadow Ball.  Yeeeeaaaahhhhhh.....perfect super effective coverage is a scary thing...

Oh yeah, that's why we need to suspect Nucleon for sure. As for Seikamater, I think that a set of Quiver Dance, Ice Beam, Thunderbolt and Surf would be best. The chart backs me up on this.

Don't forget the utility it can have with other sets~. After all, sweeping isn't the only thing this given Pokémon can do...
(10-03-2016, 11:43 AM)Lord Windos Wrote: [ -> ]
(10-03-2016, 11:14 AM)Cataclyptic Wrote: [ -> ]
(10-03-2016, 06:40 AM)Dragonstrike Wrote: [ -> ]Well, this is great!  I'm gonna have to take this into account in my analysis on Seikamater (it will hopefully be up some time soon)...that bug has so much dang viable coverage that most of the sets look like a bit of a mess, tbh...not much to do about it though.  At least the main STABs she's got can rather easily suffice on their own...

Oh, and I'll definitely be using this for future analyses too!  This is much easier to look at than trying to simply judge what coverage might work best.  

I'm not really surprised by Nuclear being best complimented by fighting, either.  That combination can hit every mon in Uranium for super effective damage with the one exception of S51-A.  And what does Nucleon have for that (and every non-dark dual typed nuclear mon as well)?  Shadow Ball.  Yeeeeaaaahhhhhh.....perfect super effective coverage is a scary thing...

Oh yeah, that's why we need to suspect Nucleon for sure. As for Seikamater, I think that a set of Quiver Dance, Ice Beam, Thunderbolt and Surf would be best. The chart backs me up on this.

Don't forget the utility it can have with other sets~. After all, sweeping isn't the only thing this given Pokémon can do...

Well yeah, but he was asking specifically for a sweeper set. At I think he did...
After looking through the dual tables, I've found that the following Pokemon have the best STAB spreads (PATE below 80000) for this metagame:

Physical
  1. Herolune
  2. Lanthan
  3. Cocancer
  4. Anderind
  5. Gliscor
  6. Majungold
  7. Terlard
  8. Navighast
  9. Drilgann
Special
  1. Syrentide
  2. Archilles
  3. Escartress
  4. Jerbolta
  5. Winotinger
It should be noted that this doesn't take stats into account, but I thought you guys might find it interesting. It definitely confirms that Syrentide and Archilles are going to be major threats. Lanthan is also scary. Another thing I found kind of interesting is that Blubelrog almost made the list, but didn't because it's a special attacker rather than a physical attacker, leaving it JUST above 80000.
Pages: 1 2