Introduction
New Year's traditional masquerade of colors and ranks is going to end pretty soon, and before it goes away, I decided to download a snapshot of everyone's true ratings and current handle colors to see how everyone's using the feature. I regret that I only have one snapshot; I only managed to push myself to finish this today. It would have been really cool to see how people's usage of the feature has evolved — perhaps an initial surge of masquerading as a nutella, then slowly coming to one's senses and opting for something milder?
Data was collected on January 9th, at about 01:00 UTC. I collected the data by scraping the pages of the "rating" tab (sorry for making a lot of requests!) — the table contains people's actual ratings, and by looking at the title-text of people's handles, I could also get their possibly-magicked ranks. I only included the active people because it was already taking a lot of time, and I believe we got most of the interesting parts anyway. Sadly this excludes magic-using unrated and inactive users, but I think they won't affect the statistics much.
The data
First, let's look at the basic data: how many people of a given rank are masquerading as another given rank?
Down: true rank, right: magic rank | N | P | S | E | CM | M | IM | GM | IGM | LGM |
---|---|---|---|---|---|---|---|---|---|---|
Newbie | 70554 | 209 | 253 | 407 | 394 | 245 | 125 | 136 | 136 | 1565 |
Pupil | 162 | 12087 | 93 | 146 | 195 | 101 | 62 | 43 | 45 | 430 |
Specialist | 213 | 27 | 7467 | 129 | 161 | 80 | 65 | 57 | 38 | 329 |
Expert | 260 | 71 | 60 | 6202 | 161 | 111 | 62 | 47 | 40 | 308 |
Candidate Master | 99 | 48 | 44 | 35 | 1886 | 43 | 17 | 27 | 16 | 90 |
Master | 76 | 34 | 66 | 40 | 23 | 1311 | 7 | 31 | 12 | 61 |
International Master | 18 | 9 | 12 | 9 | 8 | 1 | 249 | 8 | 3 | 13 |
Grandmaster | 18 | 22 | 18 | 14 | 12 | 3 | 4 | 326 | 6 | 20 |
International Grandmaster | 10 | 5 | 16 | 4 | 4 | 4 | 2 | 0 | 182 | 7 |
Legendary Grandmaster | 4 | 2 | 2 | 3 | 4 | 3 | 2 | 0 | 0 | 41 |
Down: true rank, right: magic rank | N | P | S | E | CM | M | IM | GM | IGM | LGM |
---|---|---|---|---|---|---|---|---|---|---|
Newbie | 6% | 7.3% | 12% | 11% | 7.1% | 3.6% | 3.9% | 3.9% | 45% | |
Pupil | 13% | 7.3% | 11% | 15% | 7.9% | 4.9% | 3.4% | 3.5% | 34% | |
Specialist | 19% | 2.5% | 12% | 15% | 7.3% | 5.9% | 5.2% | 3.5% | 30% | |
Expert | 23% | 6.3% | 5.4% | 14% | 9.9% | 5.5% | 4.2% | 3.6% | 28% | |
Candidate Master | 24% | 11% | 11% | 8.4% | 10% | 4.1% | 6.4% | 3.8% | 21% | |
Master | 22% | 9.7% | 19% | 11% | 6.6% | 2% | 8.9% | 3.4% | 17% | |
International Master | 22% | 11% | 15% | 11% | 9.9% | 1.2% | 9.9% | 3.7% | 16% | |
Grandmaster | 15% | 19% | 15% | 12% | 10% | 2.6% | 3.4% | 5.1% | 17% | |
International Grandmaster | 19% | 9.6% | 31% | 7.7% | 7.7% | 7.7% | 3.8% | 0% | 13% | |
Legendary Grandmaster | 20% | 10% | 10% | 15% | 20% | 15% | 10% | 0% | 0% |
The percentage of fake nutellas goes down so smoothly as the true rank increases, it's very satisfying to watch. Meanwhile, the popularity of grey is almost constantly 20% among all participants. Cyan is more popular than its neighbors (I noticed this last year too); in particular there is a spike among IGMs. Masquerading as one color below your own is universally unpopular.
Since the two yellow ranks are largely interchangeable, and so are the two red ranks, I decided to make a variant of this table that consolidates these same-color ranks. Note that in this table we don't consider masters masquerading as IMs and other such combinations as magic participants.
Down: true rank, right: magic rank | N | P | S | E | CM | (I)M | (I)GM | LGM |
---|---|---|---|---|---|---|---|---|
Newbie | 6% | 7.3% | 12% | 11% | 11% | 7.8% | 45% | |
Pupil | 13% | 7.3% | 11% | 15% | 13% | 6.9% | 34% | |
Specialist | 19% | 2.5% | 12% | 15% | 13% | 8.6% | 30% | |
Expert | 23% | 6.3% | 5.4% | 14% | 15% | 7.8% | 28% | |
Candidate Master | 24% | 11% | 11% | 8.4% | 14% | 10% | 21% | |
Master | 22% | 9.9% | 19% | 12% | 6.7% | 13% | 18% | |
International Master | 22% | 11% | 15% | 11% | 10% | 14% | 16% | |
Grandmaster | 16% | 20% | 16% | 13% | 11% | 6.3% | 18% | |
International Grandmaster | 19% | 9.6% | 31% | 7.7% | 7.7% | 12% | 13% | |
Legendary Grandmaster | 20% | 10% | 10% | 15% | 20% | 25% | 0% |
Percentage of participants
Finally, it's interesting to see which ranks feel the largest need to pretend to be someone else.
True rank | Percentage of magic users |
---|---|
Newbie | 4.69% |
Pupil | 9.56% |
Specialist | 12.83% |
Expert | 15.30% |
Candidate Master | 18.18% |
Master | 21.07% |
International Master | 24.55% |
Grandmaster | 26.41% |
International Grandmaster | 22.22% |
Legendary Grandmaster | 32.79% |
The popularity of magic increases as rating increases. This is perhaps not so surprising, considering that Div. 1 are the more active part of the community (although Div. 2 people are dominant in discussions, there is an even larger silent majority).
Thanks for reading! I still have the raw data, so what else would be interesting to see?
Thanks for your fun statistics.
Some thing fun: Why don't you get the data of unrated users to your blog?
Unrated users are not so simple to get. I don't know of any page or API method that would return a list of ALL users. We can get some unrated users top contributors list, but then I'd need to also make some API calls to verify that they are unrated. However this would mean that the inclusion criteria for unrated users would be completely different from the one for rated users. Besides that, I suspect the number of unrated users who have used magic is pretty small.
Oh, I know.
What an intersting blog!
By the way, have you noticed that I am an unrated user?
Yes, because I remember your comments from before magic.
Specialist masquerading supremacy
How do I change my color though?
I also wanna be part of the magic
Profile page -> Magic
Masquerading as one color below your own is universally unpopular.
Yeah, i don't want to get jinxed either