if we were to test a metagame with Lati@s but without Salamence, and that metagame ended up fine, then the same argument would apply both ways.
-
Welcome to Smogon! Take a moment to read the Introduction to Smogon for a run-down on everything Smogon, and make sure you take some time to read the global rules.
Lati@s Discussion Thread
- Thread starter RaikouLover
- Start date
True, but again, I think the most efficient thing to do would be capping the number of dragons on a team to 2, simply because of the argument applying both ways.
But we aren't going to test that. We can test close to infinite situations, but there's no point in doing that because people will always complain anyway. The metagame is fine at the moment, so we won't get rid of Salamence anyway.
Everyone's said it. At the same time, everyone's confident that Salamence is so much better than Lati@s, but so totally not broken, that the entire idea is ridiculous, without objectively identifying why, and I'd love to see that.
I think you're missing my point.
If a metagame with Lati@s is broken, that doesn't necessarily guarantee that Lati@s is the cause per se. If we decided that Lati@s "makes dragons too strong" overall, but it wasn't even the strongest dragon, then we'd be justified in testing whatever the strongest dragon happened to end up being (probably Salamence).
Obviously Salamence isn't broken right now. Of course, you can't exactly tell me that a metagame where we had, say, Latios over Salamence wouldn't also be perfectly playable.
Why are they being tested together? They are two different pokemon with different base stats and move pools. Testing them together is just going to skew the results, and you might as well just test all suspects at once if your only doing it this way to save time.
that was my first thought as well.
Latias and Latios are very similar. Little stat differences won't influence the results. Besides, if we were going to test them seperately, we'd need three tests.
It's very easy saying that, but I am sure Latios and Latias will go back to ubers if the metagame will end up being to centralized. Even though Salamence might be the biggest force in those teams, adding Latios and Latias is what caused the problem.
It's very easy saying that, but I am sure Latios and Latias will go back to ubers if the metagame will end up being to centralized. Even though Salamence might be the biggest force in those teams, adding Latios and Latias is what caused the problem.
Just because they were the last ones to be introduced in OU? Let's say Latios and Latias were OU since Advance, but Salamence had been deemed as Uber and now, we were going to test it. If we got that "dragon centralization", would it be Salamence's fault just because he was the last tested pokémon? The amount of usage it has won't matter, just when he was tested?
skew the results how
Jump, you won the internet. XD
Might as well be saying obvious stuff, but skewing is when a graph in statistics has a "tail". Yet, everyone uses it differently [dayum, so long since I took statistics]. It's like when people say there is a "correlation" between video games and violence.
The proper word is "confuse" or anything else similar. :P In that testing both might add too many variables, but anyways, I still support testing both at the same time. Not only is it shorter to test both, but then you'd have to test Latias, Latios, and Latias + Latios, since the metagame would change with whatever shift you do.
And even then, looking at usages in the start won't really help much because everything "settles down". Obviously usage is going to skyrocket when you introduce Latias/Latios, Hell, I'm gonna use them XD.
Bleh, begin nitpicking at my horribly constructed and oh so flammable post.
Might as well be saying obvious stuff, but skewing is when a graph in statistics has a "tail". Yet, everyone uses it differently [dayum, so long since I took statistics]. It's like when people say there is a "correlation" between video games and violence.
The proper word is "confuse" or anything else similar. :P In that testing both might add too many variables, but anyways, I still support testing both at the same time. Not only is it shorter to test both, but then you'd have to test Latias, Latios, and Latias + Latios, since the metagame would change with whatever shift you do.
And even then, looking at usages in the start won't really help much because everything "settles down". Obviously usage is going to skyrocket when you introduce Latias/Latios, Hell, I'm gonna use them XD.
Bleh, begin nitpicking at my horribly constructed and oh so flammable post.
This is only true if you honestly believe that an Uber is "default Uber" even if we never should have made it Uber in the first place. Which is what a lot of people actually do seem to think, but that doesn't mean it makes any more sense.Shiny Crobat said:
kind of an unfair question lol, I mean I couldn't tell you how testing, say, Manaphy and Mew in this manner would necessarily skew our data, just that it would.Jumpman16 said:
This applies to all of our suspects (again, let's just say Manaphy and Mew) and is pretty much exactly what Stage 3 is for.Superstar said:
That said, I'm not agreeing with TVboyCanti that we somehow won't be able to make enough sense of the data for it to be valid or whatever, just that we're definitely making a judgment call here because I'm sure there's something that could potentially go wrong by testing with "too many variables" as Superstar put it.
Manaphy and Mew are inarguably different, Latios and Latias are not.
They're different pokemon, but they're obviously similar enough that we were able to make the decision that "the risk of us being significantly affected by any skewed data as a result of this unconventional test is unlikely."
If you're going to act as if we're not at all threatened by skewed data then I have a problem with that; it's silly and pointless to suggest that when it's just a fact that testing two different pokemon at the same time is not going to give us an exact representation of a metagame with each of the tested pokemon individually. And I'm not even suggesting that anyone believes that; I'm just being nitpicky with the question "skew the results how" which I thought wasn't a great way to put it.
If you're going to act as if we're not at all threatened by skewed data then I have a problem with that; it's silly and pointless to suggest that when it's just a fact that testing two different pokemon at the same time is not going to give us an exact representation of a metagame with each of the tested pokemon individually. And I'm not even suggesting that anyone believes that; I'm just being nitpicky with the question "skew the results how" which I thought wasn't a great way to put it.
Jumpman was making fun of his use of statistics terms, not on what he meant. Similarly to how I make of fun of someone who uses "correlation" for something other than 2 quantitative variables [in which case it's a correlation, not a causation].
Correct me if I"m wrong about that though. If I recall, a "skew" is simply a tail in a bulk of data. When you skew the results, it's that you collected data wrong that the tail appeared. Or it might also mean "something far out there". As in extended or some other thing.
Either way, skew was the wrong word and was just to sound smart. The correct word is confound the data [kinda..., I can't phrase it well]. It's like in statistics, if you test too many things at once, it's pretty difficult to tell what exactly was the cause.
Either way, correlation does not prove causation. I could put a graph of # of fireman on one side, and destruction in $ on other axis, with a positive correlation [correlation is right here]. The more firemen, the more damage. Are the firemen causing the damage?
Edit: Half a year without statistics and I forgot. AND I got a 5 on the AP test? What the hug is this? Well, google is good when I want to figure out the standard deviation or something. :P
Correct me if I"m wrong about that though. If I recall, a "skew" is simply a tail in a bulk of data. When you skew the results, it's that you collected data wrong that the tail appeared. Or it might also mean "something far out there". As in extended or some other thing.
Either way, skew was the wrong word and was just to sound smart. The correct word is confound the data [kinda..., I can't phrase it well]. It's like in statistics, if you test too many things at once, it's pretty difficult to tell what exactly was the cause.
Either way, correlation does not prove causation. I could put a graph of # of fireman on one side, and destruction in $ on other axis, with a positive correlation [correlation is right here]. The more firemen, the more damage. Are the firemen causing the damage?
Edit: Half a year without statistics and I forgot. AND I got a 5 on the AP test? What the hug is this? Well, google is good when I want to figure out the standard deviation or something. :P
I think the worry is that when it comes time to vote, people will think "Latias really saved my butt in many battles with it's Fighting-resistant Wishpassing, but Latios totally destroys teams, to the point where I would switch in to stop Outrage, only to have my Steel wall receive Choice Specs. Then it switched out and came back in later to Dragon Dance.
I think Latios is unbalancing, but Latias is a boon for the OU metagame by providing crucial support. Lati@s should stay."
In other words, people keep an unbalancing pokemon because you grouped it with the less offending one.
This is easily remedied by separating the two in the final vote, but to suggest they are basically the same thing is ridiculous. It is like comparing Snorlax to Blissey. One sweeps, the other tanks hits and statuses, yet both are normal types with solid durability and good stat distributions otherwise.
Just throwing in that I've been in full support of testing them separately. They are two different Pokemon and generally do two different things.
Yes, my understanding of it is that both will have a separate vote, just at the same time.
Guys, are we really that afraid of Triple Dragon? All those weaknesses.... Yes, it sounds potent offensively, but what about defensively??? Your entire rest of the team has to be steel just to make it balanced...
Guys, are we really that afraid of Triple Dragon? All those weaknesses.... Yes, it sounds potent offensively, but what about defensively??? Your entire rest of the team has to be steel just to make it balanced...
Except Scizor and Snorlax share no common weakness (lol ice/ghost resistance/immunity), have absurdly different stats and movepools, etc
Latios and Latias are even more similar than the Eevolutions. Are you really saying that the line between "Uber" and "Not Uber" could possibly fine that Latias is on one side, and Latios the other? That's like saying Skymin's Air Slash is broken, but Togekiss's (and Jirachi's Iron Head) isn't. Oh wait...
Latios and Latias are even more similar than the Eevolutions. Are you really saying that the line between "Uber" and "Not Uber" could possibly fine that Latias is on one side, and Latios the other? That's like saying Skymin's Air Slash is broken, but Togekiss's (and Jirachi's Iron Head) isn't. Oh wait...
Care to really give a reason behind that? I mean, they really differ on how they're used rather than their actual simularities. Latios can't be made to be "bulky" enough to survive most OU contenders while Latias can. Latios can Dragon Dance and make an attempt to sweep better than Salamence (CBTrick/LO DD, etc.) while Latias can't. The main difference is how Latios can put fear into Pursuit users while Latias can bearly stand up against one :/.
Yes, yes I am.
On a side note, neither of those are broken in any way >_>.