Mismatch between timeout rates in endpoints and webpage data

Sort:
Avatar of Tricky_Dicky

Odd. I checked a match in registration. Yesterday the API TO didn't match the web stats. Today it seemed to in the few I have checked. I am using each individuaal stats endpoint, not the match endpoint.

Avatar of Tricky_Dicky

Checked a couple in this endpoint and looks to be OK

https://api.chess.com/pub/match/1666403

Avatar of Martin_Stahl
Tricky_Dicky wrote:

Odd. I checked a match in registration. Yesterday the API TO didn't match the web stats. Today it seemed to in the few I have checked. I am using each individuaal stats endpoint, not the match endpoint.

Yeah, I know that endpoint has some code updates.

Avatar of Pawnlings

It has not been corrected yet. Seems worse than ever.

Avatar of stephen_33

Where are we now with this? I sense that some are correct but others not.

Avatar of Martin_Stahl

The main stats endpoint should be accurate based on the changes recently made. They may not exactly match what's showing in the profile but should be close.

From what I've seen on the match endpoint, they may not be correct.

Avatar of stephen_33

"From what I've seen on the match endpoint, they may not be correct"

That's a nuisance because they're the ones I use. It saves having to make hundreds more endpoint requests (for each player), placing even more demand on the site's servers, which can't be ideal.

Avatar of Martin_Stahl

I can't validate this easily, but my guess is that members with a 0 timeout value in a match are probably correct. If they have a value, it might be worth checking the stats endpoint instead, at least until the match endpoint is fixed