Nah, I’m boycotting that garbage on principle. Let them play with themselves.
1 Like
alrg im gonna join you, so italy air tt here we go
You still haven’t explained the statistical math that shows that when a player base on average plays better than another their worse players are even worse than others.
Statistical analysis, math, normal distribution curves, p-values. At the least the logic behind the claim.
Edit:
How i see your claim:
I will explain using dice as an example.
You roll 10 dice rolls and get the average 3.5 .
Now you have better dice that rolls higher on average so those 10 rolls result in an average of 4.
If those better dice then also take all 2’s and turns them into 1’s (worse part being even worse) then how can the average still be 4?
There’s no evidence that premium vehicles perform worse than their TT counterparts.
Well, one tank is often bought by inexperienced players, the two others by grined by experienced players
Still parroting nonsense that premiums have worse stats because “muh inexperienced players” ?
Not the point i was trying to make
I think it was actually.
“Bro that vehicle is a premium so it can be played by every noob out there, thus ruining it’s stats.”
Tell me where i`m wrong with that?
I provided you with my personal example above. To provide examples of other players, one needs to delve into their statistics and consult with the player about when, on what, how, and in what order they played, and then take those figures and compare them. It’s not impossible, but it is somewhat difficult to implement.
But the logic of my statement is simple: all other things being equal, the same player, when using a more powerful and easy-to-use vehicle, will show a better result than on a more complex vehicle, simply because the vehicle allows it. This is obvious. And my gaming experience and statistics confirm this.
Now, regarding the question of why we observe that players on so-called minor nations have higher statistics, there are several factors:
- Due to their lower popularity in global media culture, fewer newcomers start with them.
- Due to the complexity of the gameplay, there is a player attrition. Most people don’t like being free frag dealers. This further reduces popularity.
- Those who remain play better because, as a result of playing on terrible vehicles in unequal conditions, they are forced to learn. And someone, as a result of playing on superior vehicles, when the vehicle is already playing for the player, does not strive to improve their skills and plays terribly (and then writes topics “T-34-85 is slop and belongs to 4.7”).
This creates a statistical skew that is especially noticeable in a mode where everyone is in equal environmental conditions, where opponents are statistically identical. For other modes, the skew is also present, but it is not as noticeable due to the absence of ideal conditions for collecting statistics.
However, the fact that there are fewer bad players there in the end does not make the overall performance of these bad players better than that of the more popular factions.
2 Likes
Stats paint a completely different picture.
I’m waiting for evidence that isn’t just trust me bro.
This makes more sense, thank you.
So in statistical and mathematical terms you’re arguing for a positively skewed bell curve with an extended tail due to the nature of players facing adversities and overcoming them combined with the newer players having a harder time and thus performing worse.
This would make sense if we were looking at nations as a whole.
It does not in my opinion explain how that would result in a differing average performance for individual vehicles as in your explanations case; the worse players and stats would be on the lower BR end and the better performing statistics would be on the higher end of the BR range due to the natural progression of the game. The players that overcome the adversity and learn and become on average better players would be so at the later stages of the game and not the early ones. Same with the worse players, they would show up in the statistics for the lower BR vehicles but not the higher BR ones as a at that point they would have overcome the adversities and become better players.
I do not think this is represented or can be gleaned from the statistics discussed here. I might be wrong. But that is my current stance.
Imo comparing statistics of the vehicle with 2,6 millions of battles to vehicles with just ≈150k will never show the real situation.
It shows the average experience you will have while facing the vehicles.
I mean there is too large gap between sample sizes of 2S38 and CV90, you can’t see how in reality CV90 performs compared to 2S38, because much smaller sample size gives you inaccurate information compared to a sample size of 2S38.
Spoiler
" Larger sample sizes generally lead to increased precision when estimating unknown parameters. For instance, to accurately determine the prevalence of pathogen infection in a specific species of fish, it is preferable to examine a sample of 200 fish rather than 100 fish."
That is not really how that works when the stats are based on all instances of usage of that vehicle, there is no larger sample size to pick when all possible instances have already been used to form the statistic.
My previous statement still holds true.
In theory, yes, but de facto, no.
You’re missing many factors. Firstly, the statistics provided to us by Statshark cover the entire pool of players, not just newcomers. Yes, it would be true in the context of only new players, but… Even then, there are nuances.
Firstly, even in the context of a new player, the game develops sequentially, and the lower the BR, the lower the difficulty of the game. For example, at 1.0-2.0, you don’t even need to aim properly, just aim and shoot at the enemy, and they’ll somehow die on their own. Higher up, you already have to aim, but the pace of the game is generally not very high. The timings aren’t as strict yet… There’s also no real development of mechanics yet… In general, the gameplay changes quite a bit with the BR, so at some BRs you almost have to learn to play again.
Also, unfortunately for newcomers, there’s the factor of marshal-pedobear players, players who have thousands of battles sitting at low BRs in comfortable setups and kicking poor newcomers in the sandbox with the best representatives of the technology of their BR, newcomers who don’t even know the maps, or where to shoot these dominating dads… Therefore, for example, the kill ratio on the Swedish and Soviet T-34 4.0 is somewhat larger than on the T-34-85. The former was simply more popular with such players.
There are also players who don’t grind a nation first, because, let’s be honest, not all players buy high-rank premium vehicles to boost their grind, many players grind sequentially. Even on the 4th and 5th nation. For example, out of the 10 nations that I have grinded in one way or another, I only did this on 2.5: the USA and China. Well, and half of Israel, half because I acquired the Mk.2D when I already had the 6th rank of ground vehicles unlocked. And considering that low tiers are researched much faster, there will be even more of those players compared to those who have progressed to high tiers.
Therefore, you shouldn’t expect a reasonable dependency line. The general trend exists and persists, but… you can’t describe it so simply with a formula. Statshark provides too little data for that.
Actually, the sample size after 100,000 experiments isn’t that critical anymore: the measurement error is unlikely to be more than one-hundredth of a percent.
The thing is, the initial conditions are slightly different. And the measurements were taken using different tools.
Probably because 2s38 can be obtained as soon as any player puts 30/60 dollars into their disk drive. And probably because by the time players get cv9040c they become more and more experienced. And probably because sweden isnt one of three nations unlike ussr which means sweden isnt infested by more and more grim players every sales.