Soviet T10 Cruiser Moskva - First Impressions [183k damage]

[22cm]

 

I didnt expect or fear anything. But now at least your numbers make sense. .)

 

 

Are not my numbers, but von Boeg s, I just verified them, he is right. RU ships are on average better than the others.

[mtm78]

 

Are not my numbers, but von Boeg s, I just verified them, he is right. RU ships are on average better than the others.

 

No, they appear to be because on incomplete dataset and incorrect assumptions because of failing to correlate win rate in RU ships vs win rate of individual players in other lines.

 

The outcome might be right, the method of getting there is not. 

[von_Boeg]

 

Those are still limited to nation stats. Show me the win rate of those players in other nations. Maybe GE is attractive to more 'lesser skilled players'..... 

 

edit: and I thought you would leave out the premium ships... ;)

 

This is an interesting point, but I would say it is not that relevant when we look at this over a larger time period; enough players have played both lines enough to make any significant difference in skill irrelevant at this point. However, this argument is much more relevant when we talk about shorter time periods with smaller amount of players.

 

So, naturally this is one of the major weak points of my argument: Longer timeframe is less dependent on differences in player stats, but longer time perods also say less about the current balance due to changes. Shorter timeframes tells us more about the current balance status, but also come with higher uncertenty. We will never have stats that are perfect (and I doubt even WG have such stats), but as long as we are aware of these shortcomings we might still get useful information out of the stats.

 

Your problem, however, is that you dismiss this information out of hand based on these shortcomings. Even if we could correlate general player stats with ship stats there are other data that still will be relevant, such as class based player stats, the impact of MM on the different ships, and so on forever. Not to be able to correlate all this data with national WR does not make the measurement national WR useless, it just makes the measurement more blunt.

[mtm78]

I don't dismiss it completely, but I do consider conclusions based upon it as not scientifically proven. The uncertainty because of missing correlation is too big.

 

Also, I am pretty sure WG does have these statistics, even 3rd party trackers would have access to them ( excluding those who opted out of publicly available statistics ), it's the matter of actually analyzing the data accordingly which isn't done ( afaik, might be some 3rd party tracker already doing it but I don't know about it ).