6756." Fold vs CB - 3bet pot" is wrong

[vyvojer]

It's higher than in reports.

Morny, I have a question.
You wrote in another thread:

however keeping in mind there are over 1000 stats its not possible to check every stat on every update

I don't believe you was serious.
Why don't you use some automatic test tool after each build? It's so obvious and standard.

[morny]

Its something were in the process of implementing not only for stats but automated testing for other areas of the app too but there is a lot of work involved, there are so many different scenarios you need to account for each and every stat, since we've been continually adding stats over the past couple of months and considering it would take up extra development time we have been doing it manually up until now.

btw this stat is already fixed, were just doing some internal testing before releasing the next build

[vohulg]

btw this stat is already fixed, were just doing some internal testing before releasing the next build

When we can see new release ?

[morny]

Should be a couple of hours, were doing some testing now which is time consuming but well get it out asap

[Stasek]

Its something were in the process of implementing not only for stats but automated testing for other areas of the app too but there is a lot of work involved, there are so many different scenarios you need to account for each and every stat, since we've been continually adding stats over the past couple of months and considering it would take up extra development time we have been doing it manually up until now.

btw this stat is already fixed, were just doing some internal testing before releasing the next build

He is right. Why don't you simply write a validation tool that checks if stat value is the same as in HEM1?
Won't work for stats that HEM1 doesnt have, but it still has tons of useful stats.
It would prevent fuckups like recent "fold vs 3bet" stat for example

[morny]

Agree completely and as i mentioned were in the process on implementing it, it will take a bit of time but i agree its worth it and now is probably a good time since were coming to a stage where adding more stats is slowing down and we should have probably done it a lot sooner.

[Shloogy]

do not compare to HM1. Just come up with something generic that will work for every stat you make. Maybe, compare it to reports in HM2.

Too many errors in hud stats unfortunately. I stopped reporting at one stage, and switched everything to NC. it should be a HUGE concern by you guys.

[random]

Thank you for your comments alish.

Comparing it to reports was our initial check and is now automated. Comparing it to HM1 and other software that provides stats are additional measures we included to increase the certainty of a stat being correct.