I think the structure is similar, but not identical (it doesn’t look like the termination api has pages)
Hmmm… I must discover if it’s a plus or a minus…
My new piece of abstract art!
It compares rankings taken on feb 1st from V1 API and on feb 18th (today) from termination-api.
I wanted to visualize how rating system change affected rankings.
First thing I notice is that there are no longer negative rankings. New ranks span from 0 (30k) to 37 (8d).
But I see something strange…
Someone went too high!
Here are two examples:
I double checked: these values come straight from the JSON.
{"id":731399,"username":"Cool Person",
"ratings":{"overall":
{"rating":581.6778514283598,"deviation":77.3788121959115,"volatility":0.059888360720368516}},
"ranking":30}
{"id":748237,"username":"PAPA LOUIE",
"ratings":{"overall":
{"rating":547.3072455519224,"deviation":78.58011345147807,"volatility":0.059984484173782465}},
"ranking":30}
Rating looks good, but ranking is wrong (ranking 30 means 1d, right?).
Here are some other usernames, if someone wants to check them.
I picked some and they definitely aren’t dans, but I can’t figure what the issue could be.
Looks like ranking 30 is used in case of something gone wrong… maybe a default… I don’t know…
That last one is really cool.
How to solve the OGS dan problem
Just make tonnes of people dans!
Dang you are just tearing this API apart! Saying that in a good way, I don’t think anyone would have known about these issues otherwise