Way I see it after 630 battles in the Chi-Nu II
Chi-Nu II to 4.0
Chi-To early to 4.3
and the Chi-To late should sit at 4.7 with Chi-Ri.
Simple as.
No you can’t change my mind, no I won’t elucidate and no I won’t reply.
3 Likes
How is it unfair to move a performing better vehicle up in BR? The whole point of the BR system is to allow vehicles to be raised or lowered based on their performance.
The Chi-Nu II has better velocity. When it used to be at 3.3 it used to seal club every it fought because nothing could get to its effective firing distance and even if it did the Type-5 75mm would have been considerably more effective as the tank was in its effective firing range the entire time. Meaning the closer you got, the more pen I had since I didn’t need to aim as much.
Chi-Nu II used to be 3.3, then 3.7 then 4.0, then 4.7, and now 4.3.
At 4.0 in a downtier=seal clubbing. At 3.3=seal clubbing.
It’s fine at 4.3.
Seriously don’t get why you people keep arguing why it needs to be lowered but the one who created this entire forum post refuses to listen to people and you’re in a similar case, stating that the vehicle in a down-tier was overwhelming. This is why Gaijin moved it to 4.3 in the first place. At 4.7 it got uptiers and struggled due to it being a Chi-Nu hull. Meaning in terms of performance was worse than the Chi-To. But by all means, let’s see how you respond to this I’m rather curious but probably have guessed what’s going to be posted. So enlighten me.
1 Like
Earliest BR I’ve found for the Chi-Nu II is 3.7, in realistic battles at least. It does not seem like it ever was 3.3.
Edit: after looking around some more, it seems that it started at 4.3, then went to 3.7, and has now made its way back up to 4.3.
I mean I’m going by when I first joined this game back in late spring of 2017. So its possible I’m just misremembering its been long ago so hard to remember that honestly. Perhaps you are right though.
Bro, you are not even playing war thunder
I assume this is Thunderskill data. In that case, it is difficult to see how this is diagnostic of what you’re saying. Because only users who look themselves up are counted in the statistics, you can’t account for selection bias at the source. There are prior filters that make it impossible to extrapolate hard data either way.
There is another reason why I sincerely doubt those numbers. Gaijin balances by performance, as you say, it’s the core of your complaint against them (and it’s one I can understand). The same vehicle can be at different BRs in different trees… so if the Jumbo really was that more effective in FR than it is in US, it would move up in BR. Player complaints be damned. You know this to be true, because… that’s the exact argument you’re making. That Gaijin overtiers minor nations’ equipment because of higher player skill.
You say they wouldn’t do it because people get mad, but people get mad for all sorts of stuff, and some of Gaijin’s actions are outrageous even when it comes to BR selection. They usually press on and don’t care. This is especially true if the minors’ population is very small. Same reason why they can get away with the map changes, no matter how much we complain, people who like CQC outnumber us, and so all we’re doing is shouting into the void. The same would be true here.
And they’re harder to play for other reasons, too, such as lack of armour etc. Not beginner friendly for sure.
As far as I’m aware, Gaijin does not publish the data set they use to make decisions upon. And because TS is very limited, we have no way of reconstructing it by working backwards, either. We also don’t know the methodology, i.e. how they weigh certain factors, more or less than other factors.
Data scraping of WT would be very interesting, also as a way for us players to have metrics to compare ourselves against, find out how we can improve even further. But even if all WT data was scraped, you still would be missing the human, non quantitative component of Gaijin’s decision making process. Sometimes (like with undertiered premiums that move up after a while) it’s obvious. Sometimes, not so much.
Agreed. A poor trade, and personally I have no interest in playing it anyway. If I wanted to play Pattons, I’d head over to the US tree.
It’s more that this entire discussion is based on conventional wisdom and there’s little way of actually verifying any of this stuff. “This vehicle was overtiered because the players are very good, how do I know? Because they’re still doing very well even when the vehicle is overtiered!”. That could be factually true, or it could be circular logic, and might indicate that in fact, the vehicle is competitive and not overtiered… could be either way. We have no hard data to support it.
There is also a further issue here. Balancing BRs by vehicle performance makes intuitive sense, because if a vehicle is not competitive, nobody will play it, and if nobody will play it, why bother implementing it in the game? So far so good. But how do you separate the performance of the actual vehicle, from the performance of those who play it? It’s not that easy. I’m not saying this to unequivocally defend Gaijin, a lot of the BR decisions they make are clearly informed by priorities other than making the game balanced and fair. But it’s very hard to establish a normalised performance baseline for thousands of vehicles that take no account of the people using them. Possibly a fool’s errand.
Of course, that doesn’t apply to vehicles that are exactly identical. Those should, imho, be the same BR, player performance or not.
What Misery is saying, is that if that data were true, Gaijin would raise the BR. Which I agree with, it’s what they do. Irrespective of whether it’s fair or not.
So in the case of the two Jumbos, the fact that they are the same BR, suggests that they perform the same. We can’t tell based on Thunderskill because of its obvious limitations.
In other words, if you were to give it a BR purely based on its characteristics and not player performance, it would be 5.5. Halfway between H1 and E. Since that BR does not exist, you can make it go down or up. In a binary decision like this, the outcome is always likely to be suboptimal one way or the other (if it went to 5.3, then “why is the Tiger H1 at the same BR?”, and viceversa with the E). It’s possible Gaijin opted for the higher BR because they do think Japan players are better. It’s also possible that other priorities (some of which may have nothing to do with game balance) informed that decision.
2 Likes
I find that the Heavy Tank No.6 has the more important upgrades of the Tiger E over the H1, those being the cupola and the engine, so the BR of 5.7 is justified. I feel like if you were to get a BR between 5.3 and 5.7, the Heavy Tank No.6 would sit at 5.6 rather than 5.5.
1 Like
It’s a bit of a fine grain thing. The launchers mess up with your depression at the angle where your armour is most effective… though in my experience in this meta anyway, Tiger armour cannot be relied upon in any circumstance against the vast majority of what you face, so when I play it I treat it mentally like it’s a light vehicle.
Smoke can save your life.
APCR is useless, I do agree. If they ever un nerf it, then that might change.
Because only users who look themselves up are counted in the statistics, you can’t account for selection bias at the source. There are prior filters that make it impossible to extrapolate hard data either way.
Every person who looks themselves up, every person who looks up someone else, and every person in their squadrons, which means for ever sweaty person that gets added, often another 120+ randoms get added that entirely negate this argument.
Plus you can use that argument to invalidate every statistic ever that relies on participation.
so if the Jumbo really was that more effective in FR than it is in US, it would move up in BR
You say that as if Gaijin is some perfect and divine being that operates on pure logic and is without flaw so every BR in the game is fully justified and explained when they pretty blatantly move things around in BR for a whole host of reasons outside of performance, even if that’s just sheer incompetence.
That Gaijin overtiers minor nations’ equipment because of higher player skill.
Which is pretty clear to see, the argument stands even if it doesn’t include every single vehicle in the tree, a difference in BR for the Chi-Nu ll is easy to justify on saying they are different vehicles with a different performance, but identical Jumbo have a very clear point of reference.
It’s all just guessing in the end whatever the hell snail is doing as they share nothing.
And they’re harder to play for other reasons, too, such as lack of armour etc. Not beginner friendly for sure
Which gets amplified by the fact they have high BRs, a benchmark set by more experienced players that a new player will very much struggle to match, which further increases the problem of low player counts and inflated BRs.
As far as I’m aware, Gaijin does not publish the data set they use to make decisions upon. And because TS is very limited, we have no way of reconstructing it by working backwards, either. We also don’t know the methodology, i.e. how they weigh certain factors, more or less than other factors.
Because I’m positive it would expose a whole host of flaws in logic and execution, errors, faulty interpretation and whatnot, it’s not like they guard their information for our own benefit, it’s for theirs and to maintain their ability to muck about and do what they want and pretend it’s data driven.
This is not a point of overall numbers: the numbers themselves are more than enough in terms of being representative samples. The problem is establishing what the sample is representing, that is to say: who are the players most likely to look themselves up on Thunderskill? And which players is a Thunderskill user likelier to look up, compared to others?
This is the selection bias I was referring to, the filters. If you can successfully establish what they are, then you can have a much better idea of which population is being represented by Thunderskill.
It’s pure personal anecdote, but here is an example. I sometimes look up, and update the stats of, Herrman The German on Thunderskill. Why? Because he’s very good with German tanks, and I watch his videos to try and become better at playing German tanks, so this is a good way for me to track my progress against someone I would consider to be a great player.
You can see the selection bias at work here. I want to learn, so I look myself up to see how I’m doing this month compared to last month, so I look up a content creator as an additional vehicle-specific benchmark for me to try and match.
This means that we are part of a sample that is representative of something, but it’s not necessarily representative of the average WT player.
Punchline is that this isn’t about invalidating the statistics. Just questioning whether they can be trusted to draw the specific conclusion you’re drawing, and that we’d need to know more about the TS “population” in order to interpret its data correctly.
You mistake me. Perhaps my argument was not clear enough, so I will try and explain myself better, apologies for that.
Gaijin decides BRs based on a number of motivations. The condensed version is “they decide based on what suits them”. But this can manifest in different ways. Some of these motivations are agreeable, others are not. They include, but are not limited to:
-
Player performance in a given vehicle
-
Greed (such as with undertiered premiums)
-
Lineup composition (inconsistent in its implementation but clearly taken into account some of the time)
-
Hurry to ship out new vehicles (which can make their BR placements an afterthought for them)
-
Feature change or powercreep (for example vehicle being uptiered because of feature X that is later nullified by update Y, and remaining stuck at the higher BR)
And more.
Some of these make sense from a holistic POV, others do not. The reason why I said I sympathise with the “performance dilemma” is that it’s the one I can see the sense of. BR placement based on performance is a sound idea, but evaluating performance independently of the player is… quite complicated. Obviously that doesn’t apply to identical vehicles.
That does not mean I think their overall BR system placement is good. It is not. The entire point of what I was saying with regards to the Jumbo is that Gaijin’s adherence to evaluating vehicles based exclusively on how well players do in them, means that they would absolutely uptier the French Jumbo if it was doing better, even if that in my view would be completely unfair. And I don’t think any amount of backlash would stop them, unless it got to an improbable magnitude.
Yes, it’s guessing. But some things are easier to guess than others. It’s clear for all to see why the Vidar is at 7.7. Or, to make a less self-serving example, the first time they introduced the Raketenautomat I touched it and thought to myself, lol, there’s no way this BS is staying at 7.3. And indeed, it was correctly moved up later on.
When it comes to player performance, here is the thing. Yes, evaluating vehicles based on performance is good because you want them all to be competitive. But there is one reason, beyond the difficulty of abstracting the vehicles from the players themselves, that Gaijin decides to focus on player stats exclusively over vehicle performance. And the reason is…
That this is an F2P MMO. The grind is everything. They will use every tool they can to keep players progressing, to keep the right balance between fun and frustration. If the French Jumbo was dominating, they would move it up, which would be competitively unfair, but it would make sense for their business case, because if you have too easy a time with the French Jumbo, you might be less incentivised to progress through the tree. To research and unlock the next vehicle, under the illusion that it will lessen your frustrations.
Just to be clear, I’m not saying this logic is sound. I’m saying this is the logic Gaijin uses. In their ideal world, everyone with any vehicle at any BR will hover around 50% win rate, and they’ll move vehicles up and down to make that happen.
That makes me suspect that the French Jumbo is not really outperforming its American counterpart.
Oh, it is data driven, just not in the sense of using data to determine a fair outcome. We agree on this. That’s precisely why I’m saying they would uptier the French Jumbo if it was too strong, as this would suit their business interests.
1 Like
The problem with Thunderskills is that we don’t see the entire picture.
We only see the number of battles and average stats but not the normal distribution or how the experience of the players affect it.
Gaijin is going to have a lot more accurate data and can even filter it, like only data for spaded vehicles.
1 Like
I wish we had access to a bit more information, and I’m not even necessarily saying this in the context of “transparency” and how they decide BRs… it’s just astonishing how little info you have to go on, for a competitive game.
Who’s the player with the most games in vehicle X? The one with the best K/d, WR in X vehicle? What’s the normal distribution of performance for players that have played more than X number of hours? Who wins more on what individual maps?
I’d love to know stuff like this. It would make improving yourself so much more accessible.
1 Like
This is honestly why I have issues with people who try to use Thunderskill or guess a person’s performance as a player in WT from a statistical standpoint. It never actually determines their actual skill level but people argue about it, another thing is they often will also use a WT level for a player so if you have a level let’s ay 48 but you spaded 3 tree’s out at this point then you’re not “good” at the game. It’s pretty much turning a blind eye and that’s it.