Data Attacking Type Effectiveness for ORAS OU

Status
Not open for further replies.

Ignus

Copying deli meat to hard drive
Hi Everybody! I'm here with numbers! The original, super old, thread is here. You can think of this as an unofficial update on the analysis X-Act did there, as this uses the exact same formula he did. There should be no differences in the math and this should be almost entirely an update of that thread for ORAS. (Edit: this is true for the order of the types in that cool list down there. The charts themselves, however, don't use this. make sense?)

Furthermore, if you want to wade through the spreadsheet yourself, Here's a link:

LINK.

More edits because things keep changing: The charts below are using a slightly edited formula from X-Act's; if I change the formula any more, I'll update it below.

Changelog:
  • Fixed incorrectly implemented immunity abilities. Now it works.
  • Gave chansey an eviolite.
  • Added Dual type chart. More numbers!
  • Dealing 4x super effective damage against a pokemon has been weighted differently. Now worth 2x instead of 4x
  • More Changes to the type effectiveness weighting (for dual type stuff only. Why? because I said so.)! You can find a detailed explanation Here from QxC4eva. He did a butt ton of work to help figure out a solid way to apply the average (and entered the new formula in to the spreadsheet himself), so if you have the resources send a cute girl to go slap his butt.
  • Added the "HEY WHICH COVERAGE MOVE SHOULD I TAKE GUYS" charts to the OP. Uses the most recent formula.



Bolded Numbers are below 72000. In other words, GOOD. The lower the number, the more effective the type combo. Normal + Normal is equivalent to having just a normal move.


Physical:


Special:


Mixed:



New type rankings:

Important: When it comes to weighting weaknesses, The below list is for single types. This list uses the original formula, Which means dealing 4x damage to an opponent is worth twice as much as dealing 2x damage. I kept this formula the same because it's more accurate for evaluating single type coverage.

Physical:
  1. Ice
  2. Ghost
  3. Fire
  4. Fairy
  5. Fighting
  6. Rock
  7. Dark
  8. Water
  9. Electric
  10. Flying
  11. Ground
  12. Bug
  13. Grass
  14. Steel
  15. Dragon
  16. Poison
  17. Psychic
  18. Normal
Special:
  1. Ice
  2. Fire
  3. Fighting
  4. Ghost
  5. Fairy
  6. Rock
  7. Water
  8. Dark
  9. Electric
  10. Flying
  11. Ground
  12. Grass
  13. Bug
  14. Dragon
  15. Steel
  16. Psychic
  17. Normal
  18. Poison
Okay, now I actually have to explain what these numbers mean, Which is actually pretty simple. If you read (or pretended to read) the original thread you should see a few interesting differences and similarities since 2008.
  • Ice is still really, really good.
  • Poison is still really, really mediocre.
  • Flying's rank has gone down drastically in the past seven years.
  • Ground's rank has gone up considerably in the past seven years. I Apparently don't know how to use the "times" symbol instead of the "plus" symbol.
What coverage move should I take on my set?
This is the real intent of a dual type effectiveness chart. If you've ever had issues deciding which coverage move to take, you can use this chart as a quick reference for which move, assuming each type has the same base power, will do the most damage when combined with your primary STAB! How cool is that? Oh, it's not? What are you, a plebeian? It's cool.

Physical


Special





Notes and Other Weird Stuff

  • Hey! These Numbers are way smaller then X-Act's! Surely the metagame hasn't changed THAT much!
You're damn right. The reason is because his uses every fully evolved pokemon as of 2008. I only used the 68 current OU pokemon. As these calculations are a SUM, Missing out on so many pokemon means the numbers are smaller. If someone wants to volunteer to fill out the other 300+ columns in the spreadsheet, volunteer and I'll put you to work. No problem.
I will still steal any volunteers willing to enter data, But the formula is currently "correcting" the scale of the math by dividing the usage of each the current pokemon entered by the sum of the usage of every pokemon entered so far. It works. You can trust Math.

  • How did you do Mega Evolutions?
I multiplied the usage of the base Pokemon by the percentage of how often the hold their respective Mega Stone. EZPZ.
  • How do you enjoy entering this much data?
Who knows. Maybe I enjoy stabbing myself in the brain, too.
Edit: I definitely enjoy stabbing myself in the brain. It makes me type the wrong things.

~~~~~~~~~~~~~~~~~

Big thanks to QxC4eva for coming up with the solid way to handle the weighting of type effectiveness. It works.


So that's that. Ready set discuss!

Approved by AM
 
Last edited:

Ignus

Copying deli meat to hard drive
'Kay. Only thing I'm worried about is if I plan to expand the number of data points after this (as in, add more Pokemon) I'm not sure where to draw the line for what prevolutions should be included and which ones shouldn't. If someone has a way to implement this, I'll do that.
 

Karxrida

Death to the Undying Savage
is a Community Contributor Alumnus
I'd stick to the only viable ones. So basically Chansey, Porgon2, Magneton (doesn't run Eviolite though), and Doublade.
 
I am wondering, is there a way to determine how effective two types are when paired together? I guess you could just do both of them seprately, but is there any way to account for the overlap between moves? For example, terrakion hits ice types super effectively with both its stabs, so calcing them seprately isn't exactly a accurate evaluation of the average damage it will do.
 

Ignus

Copying deli meat to hard drive
Good question. It's probably possible to do something like that. All I'd have to do is change the spreadsheet around a bit. I'll see what I can do!

Also, regarding "viable" Pre-Evolutions: I'll probably just start by entering everything UU or above, including those. Luckily the spreadsheet has a column for multiplying item percentages by usage percentages, so it shouldn't take that long to implement! Will report back with more (sort of) useful numbers.
 
How flying is so bad in this ranking when BirdSpam became popular ... ?
Birdspam is based around the concept of stacking stuff w/ common checks, so that once those common checks are gone, they can sweep uninhibited. It wasn't effective because they destroy the meta, but because of this.
 
Any chance of a similiar-ish analysis of defensive typing? I mean, we all know Steel is best and Ice is worst (or do we?), but I'm curious to see where things fall in-between.
 
Any chance of a similiar-ish analysis of defensive typing? I mean, we all know Steel is best and Ice is worst (or do we?), but I'm curious to see where things fall in-between.
That's actually a really cool idea that could highlight some less obvious points of interest. For example, fire has a reputation of being a rather poor type, defensively. However, it actually has the second most resistances of any type, iirc.. It's issue is how common the types that it's weak to (water, rock and ground) are.
 
Any chance of a similiar-ish analysis of defensive typing? I mean, we all know Steel is best and Ice is worst (or do we?), but I'm curious to see where things fall in-between.
You could do that, but the effectiveness of defensive typing would have to be based off of the list OP made because having a common weakness is much more debilitating than an uncommon one. The same (but the inverse) applies to resistances. You might say Steel is the best type defensively, but if you weight the fact that it is weak to some fairly common types which are also ranked high on OPs chart, it becomes less good. On the other hand, take water for example which is weak to some of the less common types and resists some of the most common. I'm kind of interested in this, but it looks like a lot of it has to be done by hand especially if you want to include some more obscure things, I might do something related to this, might not.

I don't really know how OP is calculating his values, but there a lot of things that could be taken into account for greater accuracy, like usage %, etc.
 
You could do that, but the effectiveness of defensive typing would have to be based off of the list OP made because having a common weakness is much more debilitating than an uncommon one. The same (but the inverse) applies to resistances. You might say Steel is the best type defensively, but if you weight the fact that it is weak to some fairly common types which are also ranked high on OPs chart, it becomes less good. On the other hand, take water for example which is weak to some of the less common types and resists some of the most common. I'm kind of interested in this, but it looks like a lot of it has to be done by hand especially if you want to include some more obscure things, I might do something related to this, might not.

I don't really know how OP is calculating his values, but there a lot of things that could be taken into account for greater accuracy, like usage %, etc.
Read the thread by Xact that the OP references. It explains everything, and it takes account for all that.
 

Ignus

Copying deli meat to hard drive
So 22032 Points of data later and we Have a multiple attacking type chart! This one is about 18 times larger then the original, So I'll just update the OP with the new charts. Also, I MAY OR MAY NOT have forgotten to put Zapdos in. Shouldn't effect much. We can blame somebody else for that.

Furthermore, I've created a chart that lets you decide which coverage move to take!

Here's how it works:
Find your primary attacking move at the top (As in, the colored ones). Your best option for a coverage move is the type right below it, with the second best option right below that, etc,etc.

Edit: with the change to the formula, this isn't accurate anymore. Check below and you'll see something really really similar, but BETTER.

For a Physical attacker:



For a Special Attacker:



These aren't end alls, but as a general rule they should be accurate.

Some interesting stuff about these:
  • ICE IS REALLY GOOD
  • Water's second best option, behind Ice, is Fairy. Azumarill is Happy.
  • Special Dragon and Fire is the only type coverage to break from the grip of Ice. Physically, there's only a difference of 36 points between the two.
  • Despite Fighting + Ghost having perfect neutral coverage, Fighting + Ice still beats it by about 2000 points.
  • Despite ground sitting at 11th for both Physical and Special solo attacking types, as a coverage move it tends to set somewhere between 5th and 2nd.
  • Normal never moves above 16th with any type. It's BAD. Don't use Return for coverage.

For Defensive analysis, I'd probably have to spend two to three weeks on it. Here's how the calcs would work, if you're curious of how you would do it.

First, you'd have to create a sum for each type offensively. The basic calculation for that would probably look something like this:

Sum for all pokemon(pokemon1usage*pokemon1move1usage*move1BP * move1accuracy/100+pokemon1usage*pokemon1move2usage*move2BP*move2accuracy/100... for every move with that type)

Move1 is a move with the attacking type, as is move2, move3 etc.
pokemon1 is a pokemon that uses the attacking type move.
move1BP is the base power of move1, move1accuracy is the accuracy of move1.

Together, this would be the "threat rating" for each type. The more threatening the type the higher the number. After this you just multiply the threat rating by the amount the defending type resists it (2 for super effective, .5 for resist, etc) and you have a Numerical representation of how good the type is defensively. It might take some time to make the numbers look good, but this is the basic idea.
 
Last edited:
That's seriously sweet. Lol, the fact that you can even formulate an equation for that really impresses me. There's no way I could :P

Anyways, I think it is interesting that many of these combos aren't possible (barring Hidden Power), or at least viable.

Actually, are you sure this is 100% accurate? It shows physical fire's best partner as ice, which all water types resist. Well, not all, I guess, but you get the idea. Waters are super common in OU too? I'm not saying it is wrong for sure, that just seems really...off.

Some other interesting stuff:

As a general rule, you can find the combination that hits the most types with a set primary type by starting with your primary type, moving down to its best coverage option, then using that as the primary type and repeating. If a type that is already accounted for comes up, move down to the next one. For example:


Start with fighting (the best type)- it's best option is ice, so next start with ice- it's best is fighting, so move to the next down, ground. Grounds best is ice, then rock, then fairy. Ice is already accounted for, and rock and fairy have redundant coverage with its other moves. So the last move should be ghost.

The big thing pay attention to is redundant coverage. That's honestly the one thing that throws off the formula, as shown. It's not perfect, but it is a good guideline for determining your best move set. Theirs probably a better way to go about it tho, I'd just have to take a closer look.

Anyways, this is seriously way cool! It's honestly WAY more useful than the formula for each individual type. Good job!

Edit: speaking of redundant coverage, both of fires two best options are actually resisted by fire... Weird.
 

Ignus

Copying deli meat to hard drive
Any time you see Ice paired with something else, you have to understand what makes Ice so strong with this equation, and also have to understand what type of things this math scores super high for.

If you've read the original thread, the full equation for a type's effectiveness is this:



(HP*DEF is actually a lot more convoluted then that, but lets just say it's that)

Where T is the effectiveness against Pokemon "n" (as in 2 for super effective, .5 for resist, etc) U is the Pokemon n's usage, and HP and DEF are the respective defense and hp of that Pokemon at level 100.
The 68 represents the number of Pokemon in OU.


If you remember, the lower this number is, the more effective the type is, according to this calculation. In other words, the bigger the bottom of this fraction is, the better the effectiveness.



So we want this part to be really big. If that happens, our type is effective!

This Sum will get bigger if one of the four following things happens:
  1. The attacking type is super effective. Hitting Garchomp with a dragon type move is worth twice as much as hitting it with a normal type move. Hitting it with an ice move is worth four times as much as a normal type move.
  2. The defending Pokemon's Usage is higher. Garchomp is #1 in usage, and so hitting it for super effective damage is worth more then hitting Manectric or Celebi.
  3. The defending Pokemon has a low HP.
  4. The defending Pokemon has a low Defense (or special defense)
If you can't tell by the way a bold random words, dealing super effective damage means a whole butt ton. The fact that Ice hits so many Pokemon for 4x damage means it gets a lot of points. On the other hand, it doesn't give you much of a penalty for hitting something that resists you when compared to the super effective's massive bonus for overkill.

Think of it this way:

I get 100 points for hitting Garchomp with a neutral move.
I get 200 points for hitting Garchomp with a super effective move.
I get 400 points for hitting Garchomp with a 4x effective move.
I get 25 points for hitting Garchomp with a 4x resisted move.


Now, lets talk about that in the context of Fire + Ice:

Ice hits Garchomp, Landorus-T, Dragonite, and Gliscor for 4x damage, and hits multiple other threats for 2x damage.
Fire hits Scizor and Ferrothorn for 4x Damage and hits many steel types that resist ice for 2x damage.
Therefore, it gets a buttload of points.

Ice and Fire gets a bunch of points because it hits a lot of high usage threats for a lot of damage. That's why it scores high. I might be bad at entering numbers, but I'm pretty sure I got this right.
 
Any time you see Ice paired with something else, you have to understand what makes Ice so strong with this equation, and also have to understand what type of things this math scores super high for.

If you've read the original thread, the full equation for a type's effectiveness is this:



(HP*DEF is actually a lot more convoluted then that, but lets just say it's that)

Where T is the effectiveness against Pokemon "n" (as in 2 for super effective, .5 for resist, etc) U is the Pokemon n's usage, and HP and DEF are the respective defense and hp of that Pokemon at level 100.
The 68 represents the number of Pokemon in OU.


If you remember, the lower this number is, the more effective the type is, according to this calculation. In other words, the bigger the bottom of this fraction is, the better the effectiveness.



So we want this part to be really big. If that happens, our type is effective!

This Sum will get bigger if one of the four following things happens:
  1. The attacking type is super effective. Hitting Garchomp with a dragon type move is worth twice as much as hitting it with a normal type move. Hitting it with an ice move is worth four times as much as a normal type move.
  2. The defending Pokemon's Usage is higher. Garchomp is #1 in usage, and so hitting it for super effective damage is worth more then hitting Manectric or Celebi.
  3. The defending Pokemon has a low HP.
  4. The defending Pokemon has a low Defense (or special defense)
If you can't tell by the way a bold random words, dealing super effective damage means a whole butt ton. The fact that Ice hits so many Pokemon for 4x damage means it gets a lot of points. On the other hand, it doesn't give you much of a penalty for hitting something that resists you when compared to the super effective's massive bonus for overkill.

Think of it this way:

I get 100 points for hitting Garchomp with a neutral move.
I get 200 points for hitting Garchomp with a super effective move.
I get 400 points for hitting Garchomp with a 4x effective move.
I get 25 points for hitting Garchomp with a 4x resisted move.


Now, lets talk about that in the context of Fire + Ice:

Ice hits Garchomp, Landorus-T, Dragonite, and Gliscor for 4x damage, and hits multiple other threats for 2x damage.
Fire hits Scizor and Ferrothorn for 4x Damage and hits many steel types that resist ice for 2x damage.
Therefore, it gets a buttload of points.

Ice and Fire gets a bunch of points because it hits a lot of high usage threats for a lot of damage. That's why it scores high. I might be bad at entering numbers, but I'm pretty sure I got this right.
I actually fully understand the concept: I read the original thread and everything, all of that makes sense to me. It just seems like bulky waters and various fire types should be popular enough to bring that combo down a bit, but maybe not. I guess I have kinda seen less rotom and Gyarados and stuff recently (I haven't checked usage in a while, but I guess that makes sense).
 


Think of it this way:

I get 100 points for hitting Garchomp with a neutral move.
I get 200 points for hitting Garchomp with a super effective move.
I get 400 points for hitting Garchomp with a 4x effective move.
I get 25 points for hitting Garchomp with a 4x resisted move.
First off I'm aware you're following what X-Act did, so if I disagree with the formula I'm actually disagreeing with the work of Smogon's best mathematician (which kinda puts shivers down my spine >_<) but it happens to be so.

I don't think that formula is a fair way of measuring type effectiveness, and that could be why Ice seems to be overrepresented on your charts. The problem is, an average is taken at the end which puts a huge bias on 4x SE hits. For example:
  • vs Gliscor (+400) then Heatran (+25) --> 425 points for ice
  • vs Chansey (+100) then Tyranitar (+100) --> 200 points for ice
The first example had ice hitting 4x super effectively, then it's 4x resisted. The second one had it hitting neutrally both times. Now, I hope you see that both cases should technically get the same score. The disadvantage of being 4x resisted is supposed to cancel out the advantage of hitting something 4x effectively, right? How about add this example:
  • vs Latios (+200) then Breloom (+200) --> 400 points for ice
So with the current formula, hitting 4x SE then being 4x resisted is considered a better attacking type than one that hits 2x SE twice! And I think you kind of said it yourself anyway. :P

I think a more accurate way is put a log around that fraction, though that tends to mess up on immunities (log(0) is undefined).
 
This is cool and all, but on a realistic level, does it actually mean anything? I mean, your calculations says ice is the best attacking type to pair with fairy, but if you're running say an Altaria with only fairy + ice coverage, you get walled by a lot of common shit (Heatran, Skarmory, Ferrothorn, etc). Choosing a fire type move or ground type move is way more useful in a realistic setting.

There are a lot more examples too (i.e.: poison being the "worst" special attacking type by your calculations, while in reality Gengar is one of the premier special attackers in the tier because of his poison STAB), but I think they should be obvious. Numbers don't tell the whole story.
 
Last edited:
A thought -- like the posts above me mention, the formula seems to be heavily biased towards super-effective hits as opposed to neutral coverage. For instance, despite the advent of Fairies, Dragon's best quality as a type has been that it hits almost everything neutrally. Similarly, Flying, despite having several resists, is harder to wall in practice because its resistors all have common weaknesses. So it seems that the emphasis is put too strongly on SE hits, without regards to just powerful neutral attacks -- so stuff like Ice/Fairy is valued disproportionately because it hits a lot of stuff super-effectively, even though you'd never run it because they have a ton of redundancy in resists and SE's.
 
First off I'm aware you're following what X-Act did, so if I disagree with the formula I'm actually disagreeing with the work of Smogon's best mathematician (which kinda puts shivers down my spine >_<) but it happens to be so.

I don't think that formula is a fair way of measuring type effectiveness, and that could be why Ice seems to be overrepresented on your charts. The problem is, an average is taken at the end which puts a huge bias on 4x SE hits. For example:
  • vs Gliscor (+400) then Heatran (+25) --> 425 points for ice
  • vs Chansey (+100) then Tyranitar (+100) --> 200 points for ice
The first example had ice hitting 4x super effectively, then it's 4x resisted. The second one had it hitting neutrally both times. Now, I hope you see that both cases should technically get the same score. The disadvantage of being 4x resisted is supposed to cancel out the advantage of hitting something 4x effectively, right? How about add this example:
  • vs Latios (+200) then Breloom (+200) --> 400 points for ice
So with the current formula, hitting 4x SE then being 4x resisted is considered a better attacking type than one that hits 2x SE twice! And I think you kind of said it yourself anyway. :P

I think a more accurate way is put a log around that fraction, though that tends to mess up on immunities (log(0) is undefined).
I get what you are trying to argue, but this is flawed in that it doesn't seem to quite comprehend the purpose of the calculation. The point of this is calculating the average damage done by a move of X type, so the bias towards 4x super effective attacks is just because, overall, this is just how it works. For example, given they have the same BP and everything, a ground type attack will deal twice as much damage to Heatran than water. This doesn't add damage, it multiplies it. So having the ability to hit one thing for 4x damage and one thing for .25x you WILL do more total damage than hitting two for 2x damage.

This is cool and all, but on a realistic level, does it actually mean anything? I mean, your calculations says ice is the best attacking type to pair with fairy, but if you're running say an Altaria with only fairy + ice coverage, you get walled by a lot of common shit (Heatran, Skarmory, Ferrothorn, etc). Choosing a fire type move or ground type move is way more useful in a realistic setting.

There are a lot more examples too (i.e.: poison being the "worst" special attacking type by your calculations, while in reality Gengar is one of the premier special attackers in the tier because of his poison STAB), but I think they should be obvious. Numbers don't tell the whole story.
Poison is NOT what makes gar so dangerous- that's honestly kinda ridiculous to suggest. It's access to unresisted coverage between shadow ball and focus blast, which is complimented by poison, that makes it so good.

The calculation DOES take usage into account. I don't get what you're getting at with the Altaria with ice coverage comment. Yeah, and? That's not the point of this. This chart gives averages. I will admit that fairy and ice seems redundant to me. Ice hits flying, grass, ground, and dragon, whereas fairy hits dragon, fighting and dark. Grass is pretty uncommon (bar venusaur, which takes neutral damage from ice), dragon is weak to both, and even then, the other types are at least neutral to fairy. They are both walled by fire and steel, which are both quite common, so this just seems... Off. I feel like ground is way too low for fairy. Barring abilities, the combination is only walled by flying and fire/steel/poison, which are not that common overall (only skarmory, Tflame, and crobat, of which the first two are pretty common, but that's just two mons...). Again, that doesn't include abilities, like gengar, but it is still VERY good. Fairy/ice is pretty redundant, which is why azumarill doesn't run usually ice punch this gen.

I do understand how good ice is overall, but I also get where virgin is coming from in this specific example (I don't fully understand its relevance to the argument presented though). Perhaps the dual type chart doesn't take into full account redundant coverage. This would account for ice being so high for so many. There are several high-ranking combinations that are HARD walled by common things due to this redundant coverage, as well as many combinations that hit multiple things for SE damage with both types. For example, rock has ice listed as its best option, however, they both hit flying for super effective damage, they are both walled by steel, and the only things that are resisted by one and weak to the other are fire and ground. This SE coverage is very limited, and it isn't made up for by great neutral coverage either. Rock/ fighting hits far more things for SE damage than rock/ ice. I get that Garchomp is all the rage and hippo is pretty common, but this doesn't make up for the numerous steel types that resist it or take neutral damage from the combo at best. Basically, i feel like this chart basically says "ok, you have this move, now which is the best type," which would explain why ice is the first option for LITERALLY everything other than itself. Dragon/ fire, for example, is only resisted by tran and azumarill, whereas dragon/ice is resisted by pretty much all steels, which I'm not going to take the time to name, and azumarill. How then would it make sense that ice is ranked higher than fire for dragon? This would also explain why ground is as low as it is; ground doesn't hit flying types, which are very common, but this is pretty much irrelevant when you consider that flying types are weak to ice for the most part. The ones that take neutral damage from ice also take neutral damage from fighting.

Tl;dr, I don't think this chart fully accounts for overlaps in coverage, and therefore leaves obvious holes in coverage, while double covering specific targets (garchomp is weak to ice, so it brings ice up for fairy and dragon, despite being redundant coverage). This could probably by accounted for by excluding Pokemon of types that overlap weaknesses. Especially since stab makes a significant difference, meaning azumarill will do nearly as much to garchomp with play rough as it will with ice punch. I know that BP isn't accounted for, however, if this is to be fully accurate, perhaps the best way to accurately demonstrate the best types to use together would be to account for these would be to remove Pokemon weak to a type from the calculations for types to pair it with.

Sorry, that is really confusingly worded, but I can't think of any better way to say it.
 
Poison is NOT what makes gar so dangerous- that's honestly kinda ridiculous to suggest. It's access to unresisted coverage between shadow ball and focus blast, which is complimented by poison, that makes it so good.
I would argue that shadow ball + sludge wave is a better attacking combo than shadow ball + focus blast. Sludge wave hits 20% harder than shadow ball and destroys the common fairy types, while shadow ball hits for neutral damage on the types that resist or immune sludge wave. The only type combos in OU who resist poison + ghost are Bisharp and Tyranitar, who can be bopped with focus blast, however their most common sets (scarf Tar + sucker punch Bisharp) will beat Gengar unless focus blast hits them on the switch, so it's generally better to play it safe with Gengar if those two are around rather than hope you get lucky with focus blast.

The 1825 usage stats agree with me here and show sludge wave as his 2nd most common move, ahead of focus blast:

| Moves | | Shadow Ball 83.849% | | Sludge Wave 77.743% | | Focus Blast 60.032%
Point being: you can call one type the worst, and on average it is, but I fail to see how that helps. The best special attacking poison type is going to use his poison moves, and type combinations versus specific threats in the meta matter far more than the "average" effectiveness.
 
Last edited:
I said that wrong; ghost typing is what makes him good, but his SECONDARY poison typing is what makes him an A+ threat instead of a B threat. Shadow ball + focus blast being unresisted is cool and all, but killing fairies is Gengar's biggest niche in ORAS. The 1825 usage stats agree with me and show sludge wave as his 2nd most common move, ahead of focus blast:



Point being: you can call one type the worst, and on average it is, but I fail to see how that helps. The best special attacking poison type is going to use his poison moves, and type combinations versus specific threats in the meta matter far more than the "average" effectiveness.
The idea of this isn't "how good is x type in practice, rather, it's a mathematical equation to determine the average damage X move would do if you slapped it on something. basically, and I can't remember if this came from the original thread or this one, but say you have a sandslash, and you're trying to determine between brick brake and night slash for coverage. This chart tells you which will usually be the most useful, in terms of damage dealt.
 
The idea of this isn't "how good is x type in practice, rather, it's a mathematical equation to determine the average damage X move would do if you slapped it on something. basically, and I can't remember if this came from the original thread or this one, but say you have a sandslash, and you're trying to determine between brick brake and night slash for coverage. This chart tells you which will usually be the most useful, in terms of damage dealt.
And my point is: who cares about the math equation if it's not good in practice? Math equations are fun and all but if it's not going to help me beat someone, I don't see the point.
 

Ignus

Copying deli meat to hard drive
Suddenly people are worried about the equation, which is perfectly reasonable.

Let's talk about how to fix these problems, shall we?


This current formula rewards overkill way, way too much.
This is obviously the largest problem with the current formula. When I say overkill, I'm mostly talking about when we run in to that fairy/ice situation y'all were talking about. If I have a fairy STAB and a ice coverage move as my second type, both will still kill Garchomp. So do I need ice in this situation? Definitely not, and the formula should reflect this. There is absolutely no reason to reward doing 200% of something's health over dealing 100%. Both of these kill the opponent, right?

How to fix:
This is an easy problem to fix in the formula. All I have to do is change some of the resist multipliers for the equation:

Old Multipliers:
  • Neutral: 1
  • Super 2x: 2
  • Super 4x: 4
  • Resist .5x: .5
  • Resist .25x: .25
New Multipliers:
  • Neutral: 1
  • Super 2x: 2
  • Super 4x: 2
  • Resist .5x: .5
  • Resist .25x: .25
What this does to our coverage list

Attack



Special Attack



See some pretty important changes? Looks more reasonable already. Most notably, Ice Electric receives the fame it deserves. If people think this change is for the best, I'll update the charts using this instead.

Let's talk about how to better use this, too. There's a few things that I've been noticing when you guys are talking about which coverage moves you should be using on what Pokemon.

Regarding the base power of moves
This chart does not take in to account base power of moves. At all. It assumes all types are equal in damage. Fortunately, there's an easy way to take base power in to account, with the power of MATH. Let's talk about it using the Gengar Example.

Shadow Ball + Sludgewave, with the new equation, have a SATE of 16013. Assuming we use both of the moves an equal amount, the two moves have an average base power of ((80*1.5)+(95*1.5))/2 = 131.25 Damage.

Shadow Ball + Focus Blast, with the new equation, have a SATE of 14644. Do the same thing as above, and we get ((80*1.5)+(120))/2 = 120 Damage.

Now, because Poison+Ghost does more damage from these moves, we divide 131.25/120 to get 1.09375, the percentage of extra damage the combo will do.

16013/1.09375 =14640.4571429, which is less than the 14644 SATE that Ghost Fighting has. Therefore, Shadow Ball+Sludge Wave is better against the whole of the metagame, although just barely. When it's this close, it really doesn't matter.
 
Last edited:

AM

is a Community Leader Alumnusis a Community Contributor Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis a Battle Simulator Moderator Alumnusis a Past WCoP Champion
LCPL Champion
I haven't been really following all the math, and not going to act like I am regardless since I don't understand it. However from a team-building perspective I can assure you Ice and Electric are indeed two of the best offensive typings in the tier, Ice being a staple now with Kyurem-B and Weavile, while Electric has always maintained its consistency from the likes of Thundurus, Rotom-W, and Mega Manectric.

Also Ghost + Poison is more deadly because of the meta itself with emphasis on Fairies. On paper Ghost + Fighting would be the go to but realistically this is why many Gengars will use Sludge Wave as opposed to not using it. A non-Sludge Wave Gengar is a super team specific asset.

Obviously there's way too many variables to say that it will always be the most effective type but the last comment speaking as a builder has some truth in it, outside of all the formulas involved.
 
Status
Not open for further replies.

Users Who Are Viewing This Thread (Users: 1, Guests: 0)

Top