People of all skill levels say a lot of things about tanks. They talk about if they are overpowered, underpowered, broken, or whatever. However, their views may not accurately reflect power levels. Here are some discrepancies I have noticed between common opinions and actual power levels, based on win rate curves (which tell you how players of various skill levels do in various tanks).
Firstly, the defender is not weak. Superunicums might do better in the 50tp prototype in the past month, but all in all, players of all calibers do ~2% better in it. The gun might be derpy, but it still has insane frontal armor (even LFP is 220 mm, which is a 50-50 for most standard ammo, and if you angle, everything becomes 260+, which is a 50-50 for most same tier gold) and lots of alpha.
Secondly, the Bourrasque does not require a high level of skill to play. Everyone with a 47%+ win rate overperforms in it, by ~1-4%. Playing in a Bourrasque isn't going to turn a tomato into a unicum, but a shitter in a Bourrasque is more dangerous than the same shitter in a Pershing or something. Even if someone just yolos about and shits out 2 clips before dying, that's already 1440 damage, more than they probably do in most tier 8s.
Thirdly, the tier 8 progetto is not that broken.
In the last month, players with 51-59% win rate overperformed slightly in it, while unicums underperformed and worse players played at their account win rate. It is a strong tank, but it is not gamebreaking.
Fourthly, the T110E3 is apparently really strong. No one really mentions it when discussing overpowered tanks. It is slow, inflexible, and unexciting. But when you look at win rate curves (both for last month and overall), you see that it is very strong.
Players of every skill level do at least as well in the E3 as they do in other tanks, and players with a 50%+ win rate overperform drastically. Sure, if your team rolls the enemies, then you can't chase damage, and if the enemies roll your team, you get gangbanged, but presumably, the incredible armor and big gun allow players to make a big difference in close games.
Fifthly, the VZ-55 isn't that broken. Players with a sub 50% win rate do slightly worse in it, players with a 50%+ win rate do slightly better in it.
Sixthly, the JPE100 is actually not that bad.
In the last month, for players with a win rate of ~56% or less, playing the JPE100 tends to actually raise winrate. It does underperform for players with 56% or more win rate. When looking at the overall curve (not just the last month's), it underperforms extremely slightly (by less than 1%) for everyone. My theory is that because it is slow and inflexible, it underperforms for the skilled players creating most WOT content (who therefore rate it poorly), but for most players, although it is slow, huge, lacks camo, is inflexible, and has armor that doesn't really work against gold spam, the massive alpha and penetration let them do better in it than in their other tanks.
Obviously, I have not examined every single tank in detail here. But this is just to start a discussion.
Also, I know some people will start talking about how they know XYZ tank is actually good or bad, no matter what the numbers say, but that could be explained by differences in personal playstyle or even just luck (over the long run, the only constant is oneself, but if one hasn't played a lot of games in some tank, one could easily get good or bad teams during those few games).
Source: https://www.reddit.com/r/WorldofTanks/comments/rhehlp/information_from_win_rate_curves/
