CBARoughneck 22 Posted October 2, 2016 I have selected update stats in my profile on xvm a few times now, i have it picked to show wn8 and in game it shows 2198 as my wn8. on wotlabs it shows my overall 2213 they use to match why dont they match anymore does xvm use its own wn8 scale now or what? 1 1 Quote Share this post Link to post Short link Share on other sites
Yupi 505 #355789 Posted October 2, 2016 @CBARoughneck, xvm excludes vehicles without data. Some other sites can use wrong data for vehicles without data. Quote Share this post Link to post Short link Share on other sites
Andrew05 76 #355863 Posted October 3, 2016 Some other sites can use wrong data for vehicles without data. The values assigned by some sites can't be considered 'wrong' as there are no official values for these vehicles anyway, and as WN8 doesn't regulate such cases, independently calculating values is a valid method. I have selected update stats in my profile on xvm a few times now, i have it picked to show wn8 and in game it shows 2198 as my wn8. on wotlabs it shows my overall 2213 they use to match why dont they match anymore does xvm use its own wn8 scale now or what? Read this Depends on how wotlabs handles the issue in the link. Quote Share this post Link to post Short link Share on other sites
MATLOCK 5 #355881 Posted October 3, 2016 (edited) That does not explain why the numbers are increasing drastically for some players, and for some it does not. As I said; I dropped 45, teammate gained 30. Similar stats in the tank. That's around 3K WN8 in Grille 15 during those battles before they got implemented. The expected values can't be individual for each player, can they? I mean, you ought to have the same expected values for each and every tank and each player, on for example EU-servers? Furthermore, since that micropatch last week or the week before, my stats are stuck, completely stuck. Pushed Update Statistics, two days later it says that last update was five days ago. Kind of feel that ever since Grille 15 and Scorpion G popped up in the game, XVM and the expected values had a meltdown. I mean we've had new tanks/trees coming in to the game before (Skodas, TVPs, M41 GFs etc.) without breaking the calculation/formula. Edited October 3, 2016 by MATLOCK Quote Share this post Link to post Short link Share on other sites
Andrew05 76 #355952 Posted October 3, 2016 That does not explain why the numbers are increasing drastically for some players, and for some it does not. As I said; I dropped 45, teammate gained 30. Similar stats in the tank. That's around 3K WN8 in Grille 15 during those battles before they got implemented. I'm sure that is no magic or failure and all that can be explained mathematically. Please, provide proof of what you said, mentioning numbers, so that we could look into that and explain. And you don't quite understand the WN8 calculation, that's why these questions arise for you, in the first place. Kind of feel that ever since Grille 15 and Scorpion G popped up in the game, XVM and the expected values had a meltdown. I mean we've had new tanks/trees coming in to the game before (Skodas, TVPs, M41 GFs etc.) without breaking the calculation/formula. On the contrary. The way the WN8 was calculated in XVM before the fix was incorrect, even though for some players it could seem normal. 1 Quote Share this post Link to post Short link Share on other sites
CBARoughneck 22 #356499 Posted October 5, 2016 i thought xvm uses wotlabs numbers and this is the #1 place to track wn8 Quote Share this post Link to post Short link Share on other sites
Andrew05 76 #356715 Posted October 5, 2016 (edited) i thought xvm uses wotlabs numbers and this is the #1 place to track wn8 Wotlabs is #1 in NA. Not many people in RU use it or even know about it. And just a hint. Wotlabs color scale is incorrect. The "%SUP" numbers don't even closely match the ratings' values. Plus their scale doesn't get updated at all, unlike XVM, which gets updated regularly and for which different colors match the corresponding player percentage. Edited October 5, 2016 by Andrew05 Quote Share this post Link to post Short link Share on other sites
Tey 34 #356740 Posted October 5, 2016 And just a hint. Wotlabs color scale is incorrect. The "%SUP" numbers don't even closely match the ratings' values. Plus their scale doesn't get updated at all, unlike XVM, which gets updated regularly and for which different colors match the corresponding player percentage. Where can we get an up-to-date scale please? I mean a table which shows the WN8 rate for each player percentile (with a small step, like 1%). I didn't find anything like that on noobmeter or wnefficiency (or what I found was pretty old). I like the wotlabs color scale because it contains more levels than the XVM one which jumps from 20% to 60% and then to 90%. But if the wotlabs levels do not match the reality, it's useless indeed :( Quote Share this post Link to post Short link Share on other sites
Andrew05 76 #356772 Posted October 5, 2016 Where can we get an up-to-date scale please? I mean a table which shows the WN8 rate for each player percentile (with a small step, like 1%). I didn't find anything like that on noobmeter or wnefficiency (or what I found was pretty old). I like the wotlabs color scale because it contains more levels than the XVM one which jumps from 20% to 60% and then to 90%. But if the wotlabs levels do not match the reality, it's useless indeed :( There is no such table. But in this post you can find approximate correspondence of XVM scale and different ratings (it's the text in the spoiler under the "Примерное соответствие баллов по двузначной шкале XVM и исходных рейтингов (0.9.15.1):" line, third spoiler in the post). And the first spoiler in the post contains the formula to calculate %SUP for any rating in XVM scale. 1 Quote Share this post Link to post Short link Share on other sites
Tey 34 #356784 Posted October 6, 2016 (edited) Thanks, I was expecting experimental data but this is good enough for my needs :) Edited October 6, 2016 by Tey Quote Share this post Link to post Short link Share on other sites
Andrew05 76 #356793 Posted October 6, 2016 Thanks, I was expecting experimental data but this is good enough for my needs :) No problem :) Just don't confuse those numbers from 0 to 100 (XVM scale) with %SUP, these are different things. 1 Quote Share this post Link to post Short link Share on other sites
CBARoughneck 22 #356799 Posted October 6, 2016 yeah NA is big on wotlabs ratings Quote Share this post Link to post Short link Share on other sites