This conversation is quite silly.
What normally warrants a bad is if the character breaks the risk/reward system. They are broken if there is no risk to picking them. The risk to picking Meta-Knight is a 5:5 match-up verses a better match up with some characters (Meta-Knight has 5:5 on most of the top tier where other characters have better numbers). Meta-Knight also isn't guaranteed to win matches when fighting other characters. So, you don't have a big reward for picking Meta-Knight, just the same reward as in every other fighting game when you can pick a top tier character. Some characters are comparatively better at fighting others then Meta-Knight, so you play a risk when choosing him over another. Contrast this with Akuma from SF2, who has air fireballs, top priority and has invisibility frames.
If the top 10 in a tournament was completely Falcon mains it would be a good indication that Falcon was broken or close to it...
No, it doesn't. It can mean that Falcon players are better then other players. There is no evidence in this scenario to say he is broken.
Nobody has claimed that MK cannot be beaten. They claim that the advantage gained by playing MK is great enough to damage the metagame.
Hey, guess what? Every fighting games has this. The Metagame is not going to be perfect. It is what it is. If Meta-Knight goes, someone else will take his spot. There will just be one less viable character.
Actually yes that is essentially what an outlier is. In ANY serious statistical analysis if someone is more than three standard deviations (I think) from the norm you throw out their data. This is the way stats are done. If you're measuring the IQ score of 20 different people and you have someone with an IQ of 250 you throw them out of the data. If someone has an IQ of 20 you throw them out two. Otherwise your data is skewed and wrong.
First, yes, no one knows what an outlier is. But your not right sadly. It's 3 standard deviations from the
mean. In your scenario, if the mean is 40, and the standard deviation is 10, then 20 is within 3 Standard deviations from the center. The same is true for 250 is the mean is 200 and the Standard deviation is 20. Again, this is all dependent on your variable and experimental unit. Your test could have been for the IQ of a special ed. school. Or IQ for those who passed a genius test.
To summarize, an outlier is a number that is three standard deviations more/less thanthe mean.
Whether or not ADHD and ALly are outliers though is another question altogether.
Which requires
1)A Experimental unit
2)A variable
3)A mean
4)A standard deviation
You can not have any evidence that they are outliers unless you have these 4 things.
I don't think you understand what you are saying here. In this case you ARE an outlier. You are significantly better than everyone else so winning the tournament is a result of how good YOU are. Not how good FALCO is. Falco is not broken, you are. Overswarm is making the SAME EXACT ARGUMENT. That Ally, M2K, and ADHD are winning because they far surpass normal players, not because of their characters.
Again, you are missing the four requirements to prove that. The statistic was of 4 people, so the Standard deviation is probably though the roof. Not to mention you have no variable.
What I'm saying guys is stop talking about outliers.
Also, this isn't true. I've proven Ally and ADHD are, in fact, outliers... on several occasions. Here's the most recent:
http://www.smashboards.com/showpost.php?p=9544720&postcount=3652
You did not prove it because there was neither a mean nor a standard deviation. In fact, I ran one of the Snake scores. The mean was 5.1 and the SD was 3.8. A score would have had to been over 16.5, but the best score was 14. You need some kind of numbers to claim you have an outlier. In other words, you proved nothing.