WN8 vs WR correlation

[sgtbarnes75 0]

Just wondering why my WR is so far apart from my WN8 at the moment last couple of months). I always thought there was always going to be a bit of a correlation in the gap between or could spotting not being taking into account or other non included metrics cause the gap ? I am a true pub scrub but do love my lights and would love to see my WN come up more. 

 

Cheers! 

[simba90 100]

Focus on winning rather than WN8.

WN8 is an almost meaningless broken stat now. It is just so padded these days. (The only real use I find from WN8 is tracking my own personal improvement)

If you must a number other than WR, then use the WG's PR for your overall and the recent tank rating.

My $0.02 but YMMV

[Archaic_One 1,853]

Because you are a seal clubber.  Your highest WR tanks are all low tier clubbing specialists and those tanks are really hard to pad WN8 in.  There are so many other clubbers playing the matilda, cruiser 2, and PZ 35t, that they have really high expected damage.  You are competent enough to carry tier 3 battles but you are not going to be able to do the insane damage numbers that guys like @Kuroialty are able to do.

[Cronk 694]

There's really not much of a gap at all, I'd disregard it and just focus on overall improvement at the game.

[MagicalFlyingFox 1,022]

I generally just focus on the winning bit, the WN8 just comes when you focus on winning games. 

 

If you prioritise WN8, you would be chai sniping more and not contributing to the pivotal moments of the game, so you will have a higher WN8 than your respective colour WR.

don't do that. 

[sohojacques 271]

My recent Wn8 is dark purple despite my recent WR only being blue because of my light tank fetish. You say you like the class. If you want to pad your Wn8 ( I would rather win more ) spamming Type 64, AMX 13 75, T71, AMX 13 90, etc will quickly get the job done.

[Sapros 348]

7 hours ago, MagicalFlyingFox said:

I generally just focus on the winning bit, the WN8 just comes when you focus on winning games. 

For me it's the exact opposite. 

I focus on farming as much damage/assisted as I can and wins happen as a result of that.

Focusing too much on something I have limited influence on (outcome of the battle), makes the game exceptionally frustrating imo.

[MagicalFlyingFox 1,022]

2 hours ago, Sapros said:

For me it's the exact opposite. 

I focus on farming as much damage/assisted as I can and wins happen as a result of that.

Focusing too much on something I have limited influence on (outcome of the battle), makes the game exceptionally frustrating imo.

Generally its just getting to the positions where you are going to influence the game the most, which is almost always the positions where you are going to do the most damage and assist, at least in the early game where the damage you put out is so much more important. 

[kolni 69,670,872]

what i do:

stomping nerds have two outcomes

stomp enough nerds to win game

stomping enough nerds but enemy team stomps our nerds harder so we lose

Edited March 20, 2018 by Kolni

[Tanager 150]

This thread is close enough to the question over which I've been mulling that I'll reuse it, rather than start a new thread.

(FWIW, I tend to focus on WR more than any other stat, b/c I enjoy winning)

Assuming a constant WN8 (to pick a single metric, albeit a much-disparaged one, as a proxy for player skill), is there any tracking between the choice of tank type/tier and the effect on outcome?  To cherry-pick two data points, my wn8 ratings in the T-10 and KV1S are nearly identical (1741 vs 1769), but my win rates are nearly 11 points apart (49.21% vs. 59.91%).  Is this merely indicative of the axiom that it takes greater player skill to affect WR positively at the higher tiers, due to the more skilled overall player base (among other possible factors)?  And would you draw any conclusion (beyond simple random chance, I which I readily acknowledge is a contributing factor) from e.g. my stats in the Jagdtiger (1764/55.65%).

So I know I'm cherry-picking from one player's profile (my own), but I was curious if there were stats that showed, for a given tier/wn8, what class of tanks produced the biggest impact on win rate (either positively or negatively)?  I think the CW is that, for a statistically close-to-average player like myself, heavies tend to produce the best results (this is borne out by my overall stats, in any case), but is that actually true?  And does that still hold true for unica (I keep reading that the better players leverage mobility more than armor to avoid taking damage, so I'd expect them possibly to do better in meds, assuming you could control for the meta)?

It's also curious to me that my highest win rate is actually at tier 9: 52.37% .  Does this say anything at all about choice of tank, or is it more that by the time I started playing 9s regularly, I'd improved enough that my numbers were better, or that the tier 9s I've chosen simply fit my playstyle?  Or maybe that having to grind more experience forces me to play them longer (and thereby improve)?  (The M46 is an obvious outlier if you look at my stats, ye gods.)

I don't really expect any grand conclusions from my personal stats - I'm a sw engineer and a nerd by nature, so I'm throwing stuff out to see if someone might have crunched the numbers to look at this before.

[kolni 69,670,872]

57 minutes ago, Tanager said:

This thread is close enough to the question over which I've been mulling that I'll reuse it, rather than start a new thread.

(FWIW, I tend to focus on WR more than any other stat, b/c I enjoy winning)

Assuming a constant WN8 (to pick a single metric, albeit a much-disparaged one, as a proxy for player skill), is there any tracking between the choice of tank type/tier and the effect on outcome?  To cherry-pick two data points, my wn8 ratings in the T-10 and KV1S are nearly identical (1741 vs 1769), but my win rates are nearly 11 points apart (49.21% vs. 59.91%).  Is this merely indicative of the axiom that it takes greater player skill to affect WR positively at the higher tiers, due to the more skilled overall player base (among other possible factors)?  And would you draw any conclusion (beyond simple random chance, I which I readily acknowledge is a contributing factor) from e.g. my stats in the Jagdtiger (1764/55.65%).

So I know I'm cherry-picking from one player's profile (my own), but I was curious if there were stats that showed, for a given tier/wn8, what class of tanks produced the biggest impact on win rate (either positively or negatively)?  I think the CW is that, for a statistically close-to-average player like myself, heavies tend to produce the best results (this is borne out by my overall stats, in any case), but is that actually true?  And does that still hold true for unica (I keep reading that the better players leverage mobility more than armor to avoid taking damage, so I'd expect them possibly to do better in meds, assuming you could control for the meta)?

It's also curious to me that my highest win rate is actually at tier 9: 52.37% .  Does this say anything at all about choice of tank, or is it more that by the time I started playing 9s regularly, I'd improved enough that my numbers were better, or that the tier 9s I've chosen simply fit my playstyle?  Or maybe that having to grind more experience forces me to play them longer (and thereby improve)?  (The M46 is an obvious outlier if you look at my stats, ye gods.)

I don't really expect any grand conclusions from my personal stats - I'm a sw engineer and a nerd by nature, so I'm throwing stuff out to see if someone might have crunched the numbers to look at this before.

I'm not a nerd so I don't crunch numbers but

Some things:

• Battle count. I have sub 50% WR sessions in T10s with 4,5K DPG and above. Until you're hitting 100+ it's honestly not very accurate, and by that point and I'd still say it can easily differ +/- 10 simply because of maps/teams regardless of how well you play. It obviously gets more accurate the more you play, a 1K games tank is about as precise as it will be. My STRV S1 has played more Stalingrad than every single other tank combined has in 9.21 which is just an example of how you just don't fit and do worse because of something completely out of your control. The T-10 fares decently on almost every map though, so less of an issue but it still is one. 
• Breaking point: Until you really understand how to actually win games it's going to vary because different tanks and different playstyles excel at winning in different ways. A T10 wants to push, and KV-1S wants to trade which are two different things. This means that these tanks shouldn't be played in the same way and while you might get similar results in damage the difference is that you're doing better damage at the right places in one of them. 
• Higher tiers are obviously harder to do well in, T-10 has always been one of the most popular T9s so the values are absolute dogshit so it's not very hard to pad in so this might give you a false sense of improvement etc. A KV-1S absolutely shits on other T5s and that's a place where people are generally worse players, and it's fairly new so the values are higher. 
• Very few people actually know how to win. Most unicums don't either. Picking a favourable engagement, stomp it and going to the next one is all they do which is not all that there's to it. Map control, vision, team comp and positioning are all equally important (vision is still vision, corridor meta but still matters because you want to see what's happening) and knowing them well can get you further than any micromanagement skill. Understand what leads to what to start getting the hang of gameflow, and timing. Knowing this well makes gameplay much easier to predict, and if you can predict it you can counter it without needing a triple platoon to help you. Until I started playing meme tanks I was soloing over 70% (playing 907, E50, M46 and other strong aggressors) and 100 game streak in the M46 landed me at 84% WR there. Understanding how the macro is played is far more important to making any tank work. I can hop into any tank at this point and perform very well, while you have the other type of unicums who literally only know how to play russian mediums and nothing else and so on, and while they might outperform me in that end I'll outperform in every other one because rather than knowing the tanks limits/strengths very well I know the game itself very well which is ultimately a bigger thing to have. Knowing the game well enough to never have to end up in a 1v3 is a better skill to have then being able to win the 1v3 in the first place. 
• Ultimately knowing what your tank is good at will have to take place in the equation as well, just by looking at your stats on other T9s I'm fairly confident in drawing some conclusions to it. Your E75 does well in winrate compared to your damage output, the tank is honestly not very good but it's a tank with traits that are simple to understand and easy to utilize which is why you're likely doing well in it (Slow as shit so overextending is hard, sidescrapes well and easy angles to manage and the gun hits fairly hard so trading is viable) while your Jagdtiger is putting up the same damage output but not doing as well in WR simply because it requires different ways of winning. You can't hold angles because you can't shoot and keep your armour strong at the same time, and as a result you have to use your HP for favourable trades and basically treating your armour as an unreliable extension of your HP whereas the E 75 does it the other way around since the armour is workable and the HP can be treated as an extension of your armour. An E75 can trade while low HP but a Jagdtiger cannot, while a JT is much more dangerous with HP than an E 75 is. For the T-10 it gets trickier to be honest since it's a tank that does everything fairly well, and you can play it to the strengths of your gameplay rather than the gameplay that the tank might be best suited for, at the drawback of not being as good as other tanks in that situation but still not bad and this is exactly why the T-10 works well for many players. Regardless of how you like to play (bunch up with mediums and take map control or trading in CQC with heavies or simply full on aggression) it works. The ideal T-10 gameplay is honestly crossfire creation and using the armour to mitigate the risks of taking those cross positions, but you can't really do that most games so it often falls back into whatever you feel comfortable doing since the T-10 can do it even though there's other tanks that are better for it. 

I'd suggest you grind and try the WZ-111-14 for some more insight on it, similar gameplay but different traits that make it a bit more defined in what it's good at. The WZ-111-14 is by miles a better tank for me in terms of winning, but for most people I think it's actually the T-10. Versatility is a very good trait to have when maps can shit on you so bad for playing anything more defined but it's also at the cost of performance in general to me. 

To me the T-54 is the end-all tier 9 tank right now. A lot of people don't agree with me on it but here you have versatility but much stronger presence at the cost of being more vulnerable. A T-54 can shit on any tank just with outplays and good positioning while it gets harder for the T-10 to do so even though it certainly still can. 

 

Basically winning is a concept that is difficult to figure out, and you never really hit a breaking point where the penny drops but just keep collecting small bits of information that gradually improves your understanding

[Tanager 150]

@KolniThat was one of the best posts I've ever read.  Thanks.

[Folterknecht 2,257]

@TanagerComparing KV1S (with 22K battles you probably played while it was still the t6 rofl stomp tank) to a T10 makes absolutly no sense (apples and oranges). And that is totally beside the fact that WN8 was never intended as a tank or balance comparison tool. It was designed to assess player skill compared to other players, especially with the 1250WN8 player with a few thousand battles (aka "Mr. Somewhat Usefull"). WN8 or more accuratly it's development process only highlighted the shitshow that is the WG balance department and the nonsense they create on a daily basis, because it was the first serious outside investigation (by mathematical + gameplay gifted people) into player and tank results outside of Minsk.

 

Ofc there is a correlation between the overall skill of a player and the performance of said player in different tanks no question. But that doesn't change the fact that WNx is the wrong tool for the job of direct tank comparision especially between different tiers and tanks that are at different points on the balance scale.

 

On 20.3.2018 at 6:57 PM, Kolni said:

Some things:

• Battle count. I have sub 50% WR sessions in T10s with 4,5K DPG and above. Until you're hitting 100+ it's honestly not very accurate, and by that point and I'd still say it can easily differ +/- 10 simply because of maps/teams regardless of how well you play. It obviously gets more accurate the more you play, a 1K games tank is about as precise as it will be. My STRV S1 has played more Stalingrad than every single other tank combined has in 9.21 which is just an example of how you just don't fit and do worse because of something completely out of your control. The T-10 fares decently on almost every map though, so less of an issue but it still is one. 

addendum

Until you get to at least ~400 battles the error is too big to compare directly bewtween players that are of a similar "caliber", though 100 battles are good enough to usally distinguish between a potato, Joe Avg and so on.

[Saffee 432]

On 3/19/2018 at 10:29 PM, Sapros said:

For me it's the exact opposite. 

I focus on farming as much damage/assisted as I can and wins happen as a result of that.

Focusing too much on something I have limited influence on (outcome of the battle), makes the game exceptionally frustrating imo.

No offence but if you really believe that you have such a limited influence on winning im not sure you really understand how to win? 

Damage is damage, but map control is everything.

On 3/20/2018 at 8:59 AM, Kolni said:

what i do:

stomping nerds have two outcomes

stomp enough nerds to win game

stomping enough nerds but enemy team stomps our nerds harder so we lose

^^

[kolni 69,670,872]

16 hours ago, Saffee said:

No offence but if you really believe that you have such a limited influence on winning im not sure you really understand how to win? 

Damage is damage, but map control is everything.

Comes from the guy who plays beach on El Halluf more often than not  

[Sapros 348]

17 hours ago, Saffee said:

No offence but if you really believe that you have such a limited influence on winning im not sure you really understand how to win? 

Pretty sure you win by killing, capping, or occasionally just surviving. 

Since nobody solopubs at even close to 100% w/r I think it's safe to say one's influence over battles' outcome is in fact limited. 

[nabucodonsor 461]

19 hours ago, Saffee said:

Damage is damage

Someone knows the memes

[Saffee 432]

On 24/03/2018 at 5:20 AM, Kolni said:

Comes from the guy who plays beach on El Halluf more often than not  

Dont be a hater yung! 

You underestimate the power of APAC 

[Strigonx 442]

play the WZ-111 1-4

Win

EZPZ

[Proloser 29]

Play the 268v4.

Win more.

EZEZEZPZ

[Lindy 41]

@Kolni, very nice post, thank you