This is my Results based tierlist based on only post final patch results using my new chaarcter dataset which was compiled by extracting character data from over 73,000 youtube titles/descriptions of set vods and matching them to matches on start.gg. This data was combined with the smashdata character data from start.gg to get accurate and up to date character info on nearly every active player. A lot of these metrics involve ratings which come from my ranking/seeding algorithm which has been used at a handful of majors and proven to be statistically more accurate than human seeders.

Methodology

This tierlist is an aggregate of 8 different lists which are each based on different metrics that indicate a character's viability in the meta and where they should be placed on the tierlist. Each character's final score is a weighted average that gives less weight to outliers, which is a comprimise between using the regular average and the median, so that outliers still have an effect but a small one. I went with this since there will nautrally be high volatility and I don't want 1 low placement to doom a character but I still want it to have an effect. Here are the weights:

A character's highest placement has a weight of 5%

A character's 2nd highest placement has a weight of 10%

A character's 3rd highest placement has a weight of 15%

A character's 4th highest placement has a weight of 20%

A character's 5th highest placement has a weight of 20%

A character's 6th highest placement has a weight of 15%

A character's 7th highest placement has a weight of 10%

A character's lowest placement has a weight of 5%

This tierlist is based only on results post-final-patch(12/1/2021), which is important because some characters had their best results before this date. I made this decision because otherwise characters who were nerfed (cough pichu cough) would be unfairly high and characters that were buffed or added as DLC would be unfairly low.

For the purposes of this list, "major" is defined as any tournament with 300+ entrants. I know the definition of "major" usually has more nuance than this but it's the simplest and most-straightforward way to define them and it works well enough.

Lists 6 and 7 involve "hardcore" players which I define as any player who has been to 50+ offline events post-quarentine. This is used to measure the performance of the players who have put the most time into the game and their characters in the current meta.

Every player's effect on a character's spot on the ranking is proportional to what percentage that player plays the character. Specifically, all of these lists are weighted by what percentage the player plays that character. So if someone plays mario and donkey kong 50/50, their results will half count for mario and half count for DK. Exact details vary from list to list

The 8 Individual Lists

1) Sum of wins at majors

The score is based on the sum of the ratings of the players defeated by players piloting that character, weighted by that character's share of the victor's character data. So if you beat someone with a rating of 10,000 but you only play X character 50% of the time, only 5,000 points would be added to X character's score

2) Rating of wins per loss

This score is almost the same as the previous one but the score is divided by the total number of Ls(weighted by that character's share of the loser's character data) taken by that character at majors. This punishes characters that lose a lot and rewards characters that do well despite few representatives. It's essentially a measure of the performance of the average pilot of each character.

3) Sum of their best players' ratings

This is probably the hardest one to explain the details of. It's the weighted rating of the top 10 weight, not the top 10 players. So if player X plays character A 80% of the time and they're the best, they will take up the 0 to 0.8 slot. If player Y plays character A 50% of the time and they're the second best, they'll take up the 0.8 to 1.3 slot and so on up to 10. Then the score is the weighted average of those player's ratings. This confusing system is so that if a character is a popular secondary, they are rewarded for that by the players who secondary them having a spot in the top 10 but those players' effects on the score is proportional to how much they use the character.

4) Peak Rating

This score is the highest weighted rating achieved by the character since the final patch. Weighted rating is the player's rating times the character's share of the player's character data. This particularly rewards sucessful solo mains since their rating is multiplied by 1(100%). That's why Steve, Sonic, GnW, Fox and Wario are particularly high.

5) Top 64 representation

This score is the percentage of top 64 that are from pilots of this character, weighted by the character's share of the player's character data

6) Sum of ratings of all hardcore players

See above for the definition of "hardcore" for the purpose of this list. Besides that, this score is pretty self explanatory: the sum of all ratings of hardcore players. This measures how good the most dedicated players are as well as the number of players who are willing to dedicate that much time to the character

7) The average rating of the top 10% of hardcore players

Again besides the definition of "hardcore," this one is also pretty self-explanatory. It measures how good you're expected to get if you're dedicated AND you're among the most competent with your character.

8) Rating of the average new player

This one is probably the most unique among these lists. It takes all the players who went to their first tournament(including online) under 2 years ago who've been to atleast 20 events(to filter out alts and players who quit) and calculates the amount of rating they gained per day since they went to their first tournament. The score is the average(weighted by the character's share of the player's character data) rating gained per day by new players that play that character