World of Tanks

[Sir T]

Funny, that's what I was thinking.

*edit* Oh and as for how many games I've played in scouts, you "forgot" to mention that I've played almost twice as many games in the VK28.01 (553) as the number 3 tank, the really survivable AMX 50 100 (344). And that the VK 3601 (h) used to be a medium. I've played it a grand total of once as a heavy and then I sold it. And my no 5 tank is the Leopard (274) and I played that before it was buffed. And my top tank (643) is the Lowe, a premium that's not exactly seen as being very good and is dead last on the win rates for tier 8 heavies, but it makes scads of cash so who cares.

So yeah I've played a lot of freaking scout tanks. Jeez

I mean ffs, I'm not even claiming to be much good here, I'm not really, but come on, don't distort shit to prop up your argument.

[Abagadro]

Except I have things like "math" and "facts" and "evidence" while you have a feeling (and vague notions of "but, but platoons!") based upon not wanting to believe the reality that you are bad at tanks.

[Ginaz]

What would it have been if I had kept going on the tier 9 and working on raw damage because that includes the time I was platooning on lower tiers in tanks I don't like much or are stock as shit. That was my point, it it does throw your think out of whack. Hell my k/s ration is better than what you said generally in any case these days. And it's amazing how a simple change in what tank I'm driving will throw your example out of whack. If I could drive a bat chat what would my k/d be like? Or the faster firing highly armored American autoloaders?

You can poke flaws on any number, but simple win rate is the worst metric you can look at to see how good someone is.

But no one is doing that.  Win rate is not the be all and end all way to measure how good you are.  However, it very often correlates to many other factors that do.  BTW, playing tier 9 and 10 tanks isn't a sure way to increase your stats.  While you can do a lot more damage in an IS-7 than you can in a KV-1, everyone else can do more damage, too.  You're level of play doesn't automatically rise due to the tier of tank your driving.  I've seen plenty of people in tier 9 and 10 tanks that do little or no damage before being evaporated in 2 secs.

[Sir T]

But no one is doing that.  Win rate is not the be all and end all way to measure how good you are.

Read the thread dude. People are doing exactly that. (*edit* The argument seems to be "win rate correlates to the other states, so therefore its the best thing to use!!!")

However, it very often correlates to many other factors that do.

No argument there. My issue really is people pointing to one single stat as "This means your crap"

BTW, playing tier 9 and 10 tanks isn't a sure way to increase your stats.  While you can do a lot more damage in an IS-7 than you can in a KV-1, everyone else can do more damage, too.  You're level of play doesn't automatically rise due to the tier of tank your driving.  I've seen plenty of people in tier 9 and 10 tanks that do little or no damage before being evaporated in 2 secs.

Well I was running around in fully upgraded tank packing a crew with 3 skills a peace and my "K/D rate" suddenly shot right up. My other stats like win 7 etc are way up from normal for today too which was pretty nice. Since that was held up as the metric of success I thought it was funny and relevant to point it out as a counter argument. I didn't realize it would offend the faithful.

Anyway, yup, I'm bad at tanks. The humanity.

[Ginaz]

But no one is doing that.  Win rate is not the be all and end all way to measure how good you are.

Read the thread dude. People are doing exactly that. (*edit* The argument seems to be "win rate correlates to the other states, so therefore its the best thing to use!!!")

However, it very often correlates to many other factors that do.

No argument there. My issue really is people pointing to one single stat as "This means your crap"

BTW, playing tier 9 and 10 tanks isn't a sure way to increase your stats.  While you can do a lot more damage in an IS-7 than you can in a KV-1, everyone else can do more damage, too.  You're level of play doesn't automatically rise due to the tier of tank your driving.  I've seen plenty of people in tier 9 and 10 tanks that do little or no damage before being evaporated in 2 secs.

Well I was running around in fully upgraded tank packing a crew with 3 skills a peace and my "K/D rate" suddenly shot right up. My other stats like win 7 etc are way up from normal for today too which was pretty nice. Since that was held up as the metric of success I thought it was funny and relevant to point it out as a counter argument. I didn't realize it would offend the faithful.

Anyway, yup, I'm bad at tanks. The humanity.

Whatever, dude.

[Sir T]

So, anyone want to talk about the war on Scouts? 

[Amarr HM]

See, THAT is legit criticism.  Folks do look at raw W/L ratio and weigh it too heavily.  Dam Taken/ Rec'd and Kill/Death are better stats to measure by.

This is true and it's perhaps a throwback to the days before platoons and companies (beta) when it was a good reflection of performance and the highest Win% was something like 60%.

There are cases though, if a player plays only using light/scouts with poor matchmaking and happens to be incredible at it. He can turn-over low on both the Damage and K/D and still record a high win rate %. A good spotter will win matches when paired with other good players who can take advantage of this. So suddenly you're looking at high Win % for this guy cause there is no public stat for players spotted / damage from spotting / distance travelled. He'll probably have a lot of funky medals though if he's that good.

No argument there. My issue really is people pointing to one single stat as "This means your crap"

I think the general consensus is that we all know Win % can be skewed a good player can have a lower % than a bad player, but it does as Abagrado has pointed out, it can paint a picture. It's your choice how you want to frame it.

I actually agree with your other point, playing T10 and maybe T9 tanks is the one way you can actually affect your Win%/efficiency more often. As T10 You are always the top tank on your team so your apportion of responsibilty towards a good performance increases and there will be some games where you have the lions share. If a team of T8s and T9s loses their only T10, it's a bit of a blow to their chances.

[Strazos]

Sometimes I think it's almost unfair to take your entire play history into account - I know I've gotten better over time:

Overall:

3,467 battles -
K/D 1.21
Win% 50.85

Last 1,000 battles:

917 battles -
K/D 1.52
Win% 51.80

I really wish I could pull these stats based on specific tanks - I know they'd look better in my T29 than my Tiger.

Also, because I don't want to double-post - two replays:

The first - I thought it was my last match of the night with Garga, but I refuse to let an arty match on Himmelsdorf be the last. And, I couldn't get any decent shots, so instead I decide to get a bit "creative" - https://dl.dropboxusercontent.com/u/2459685/WoT%20Replays/20130929_0308_usa-M40M43_04_himmelsdorf.wotreplay

So then we play one more match, in the hope that we get a decent map. Even though the RNG wasn't kind in the beginning, I think we made it work later in the match. Also, this is why I hate when people give up at the end of matches: https://dl.dropboxusercontent.com/u/2459685/WoT%20Replays/20130929_0317_usa-M40M43_36_fishing_bay.wotreplay

[Merusk]

See, THAT is legit criticism.  Folks do look at raw W/L ratio and weigh it too heavily.  Dam Taken/ Rec'd and Kill/Death are better stats to measure by.

This is true and it's perhaps a throwback to the days before platoons and companies (beta) when it was a good reflection of performance and the highest Win% was something like 60%.

There are cases though, if a player plays only using light/scouts with poor matchmaking and happens to be incredible at it. He can turn-over low on both the Damage and K/D and still record a high win rate %. A good spotter will win matches when paired with other good players who can take advantage of this. So suddenly you're looking at high Win % for this guy cause there is no public stat for players spotted / damage from spotting / distance travelled. He'll probably have a lot of funky medals though if he's that good.

Which is really just saying, "Using one set of stats for all tanks is stupid."  Which it is.  You don't track INT for linemen, and good TDs should have a much higher K/D than Lights, Meds or Heavies.  Spotting should be tracked for lights, accounting for the smaller % of damage they're awarded.  Etc.

Win will track with a lot of these, but platooning DOES affect it.  If you never, ever, ever platoon (me) you're going to wind-up with a lower WR because you're all random chance on groups all the time.  Platooners should have a higher % because that coordination does help and matters a lot, esp if they're quality players and not a troll platoon.

[angry.bob]

They are highly correlated though.  I don't know why this is so hard for people to understand.

Because correlation does not imply causation? You've got a degree so I can't believe you haven't taken at least a 100 level stats course. Frankly I'm shocked my stats are even that good since I'v played mostly artillery since the patch in addition to playing through matches as fast as I can to get as many 2x wins and missions done for the last year. I also enjoy blowing up buildings in the cap on encounter maps so my hit rate is probably in the crapper too. After this semester I'll have to play seriously for stat maximizing for a week or whatever to establish a baseline for you since apparently you're some sort of crazy stat stalker.

The list of our stats you posted tells me more about how seriously, or too seriously we each take playing the game or our motivation for playing than anything else.

[Amarr HM]

Which is really just saying, "Using one set of stats for all tanks is stupid."  Which it is.  You don't track INT for linemen, and good TDs should have a much higher K/D than Lights, Meds or Heavies.  Spotting should be tracked for lights, accounting for the smaller % of damage they're awarded.  Etc.

Win will track with a lot of these, but platooning DOES affect it.  If you never, ever, ever platoon (me) you're going to wind-up with a lower WR because you're all random chance on groups all the time.  Platooners should have a higher % because that coordination does help and matters a lot, esp if they're quality players and not a troll platoon.

I totally agree with you (even though I have no idea what a lineman is), win ratio can be misrepresented, skewed etc. Generally using stats to compile a profile of a player is an amalgation of play style and what he has achieved through this.

 I'm basically arguing that it's possible influence your win% even solo, so you can use it even for yourself to see some improvement in your play. The amount of influence (playing solo) is obviously quite limited and only noticeable over large samples. I would guess the min-max is in the region of 45-55% from botters to good players. It can be boosted up by fitting out your tank with gold everything, max crews, best equipment. Also playing tier Xs or tanks with good matchmaking potential, puts a quality player in a better position to have a high sphere of influence in every game he plays. Theoretically you may be able to pump 60% solo if you work at it and get a bit lucky.

[Nebu]

Theoretically you may be able to pump 60% solo if you work at it and get a bit lucky.

Zakaladas probably represents the upper limit.  The guy has a 64% WR and one of the best WN7 values in the game and plays almost exclusively solo.  To achieve higher than he has, you really need to run in an optimized platoon with 3 incredibly good players.  

If I play tanks that I'm best in for tier 5-7 matches, I feel I can carry to near a 60% WR solo and I'm nothing more than above average (58.5% in T1HT, 59% 3601, 62% IS2).  Having a 4 skill crew and equipment helps a ton when playing against people with 75% crews.  My personal goal is to have a 55% WR and 2000 average damage per match solo in tier 8.   I have a lot of work to do.  

I agree that WR isn't the only metric, but I would argue that it's among the best.  Damage per game can be meaningless for those that camp as most of their damage comes at a point in the match where the outcome is largely decided.  Efficiency relies too much on cap points. WN7 counts wins twice (experience per match and win rate) and over values kills.  I also agree that scouts get screwed by most of these metrics... which sucks as scouting takes the most skill to pull off and often receives the smallest payback.

[Sir T]

But no one is doing that.  Win rate is not the be all and end all way to measure how good you are.

I agree that WR isn't the only metric, but I would argue that it's among the best.

 

That explains why every one of my stats is constantly going up apart from my win rate, eh, which was actually tracking DOWN the last 1000 battles or so. :D I'm sure the stat stalkers can confirm and handwave.

[apocrypha]

FFS Sir T do you even read the things you quote?

Nebu EXPLICITLY says there that it's not the only metric. There's a big difference between saying something is the "be all and end all" and "among the best". Over and over again in this thread people have said that WR shouldn't be taken alone and you consistently ignore them saying that and keep on trying to claim that that's exactly what people say.

You really do debate this issue like a creationist or a climate change denier. I haven't played WoT for *months* but you are shitting up this thread so badly I feel compelled to join in and call you a cunt.

[Sir T]

Do I have to quote the entirety of the last fucking page where people were whiteknighting winrate alone as the best indicator of player skill, and then shitting on me because I offered a counter opinion.

The top clans don't look at win rate (they use dmg/game, kill/death ratio and kills/game) to separate out the really good players, but those all correlate very highly with win rate.

Win rate is an accurate indicator of player skill.  

You realize that, over a large sample size, we all get the same amount of crappy teams.  The statistics are pretty easy to explain.

Yup:  "The only constant is you"

Win rates are a good indicator of skill.  You can make the case that a 60 percent player is not necessarily better than a 55 percent player as there are factors that can affect that other than skill, but a 55 percent player will universally be better than a 47 percent player.  Guaran-damed-teed.  

I'm sorry that screaming bullshit is shitting up the thread but hey. At least I'm not fucking lying about what tanks I've driven like the fucking super-statistician did when he tried to claim that 8 of my top 10 tanks driven are heavies while forgetting to mention the numbers and that one used to be a medium. And then he shoved aside my excellent hit ratio as "not an indicator of player skill". And then had the GALL to say he had evidence and maths on his side. Yeah, because only the numbers he likes are valid evidence and only he is the judge of what is valid evidence and what isn't. Who do you think that sounds like? I'll give you a clue they are denying climate change and evolution.

But since he likes maths here's some evidence

My win rate <last 7 days> 48.41%  <Last 1000 battles> 49.09%
Average damage    563    515
Efficiency      910   907
K/d   0.62   0.64

So yeah my win rate has gone down but my stats have gone up. That's better than him trying to explain why a guy whose recent win rate has been above average (over 48%) is still a "statistical liability to the team" by his own metric, without giving any examples at all other than stat stalking and then lying about my stats.

Which I'll admit was a really clever way to call me a retard based on my stats, but it was still enough to get me angry. Oh and yeah this afternoons K/d? 0.94. Win rate 52.17. Efficiency 1,093. Am I a Tank god whose opinion has value yet?

What a bunch of stat chasers. Have faith!

[Abagadro]

They are highly correlated though.  I don't know why this is so hard for people to understand.

Because correlation does not imply causation?

It isn't about causation as that isn't what is being tested. Its about using one variable as a substitute for another.  If they are correlated highly enough, it can be considered a substitute or aggregate variable.  I have NEVER said it is the 'best" indicator for this very reason. It is an approximation that reflects lots of other variables that can be used if you have the time. But it is a VERY GOOD indicator of skill because of the correlation between those multiple variables and win rate.

As for "seriousness" I've always wondered what my stats would be like if a good 1/4 of my games weren't played while significantly intoxicated.

[Abagadro]

I'm sorry that screaming bullshit is shitting up the thread but hey. At least I'm not fucking lying about what tanks I've driven like the fucking super-statistician did when he tried to claim that 8 of my top 10 tanks driven are heavies while forgetting to mention the numbers and that one used to be a medium. And then he shoved aside my excellent hit ratio as "not an indicator of player skill". And then had the GALL to say he had evidence and maths on his side. Yeah, because only the numbers he likes are valid evidence and only he is the judge of what is valid evidence and what isn't. Who do you think that sounds like? I'll give you a clue they are denying climate change and evolution.

Dude, your point was that your stats were shit because you played so many lights. That's clearly false.  So the 3601 was a med, that doesn't change the point that all but TWO of your top tanks were NOT lights (plus the 3601 is a rape machine so that doesn't exactly help your point).

And your numbers basically prove my point. As your k/d ratio was higher, your win rate was higher. You played better in those games, so your win rate went up. SHOCK!  It is also still a slight negative below neutral, and your win rate is slightly below 50%. STOP THE PRESSES!  Efficiency is actually not a very good indicator in my mind as it overweights certain things.

Seriously dude, you need to take a deep breath and think about this a little bit more.


EDIT: Plus seven days is too small of a sample size to mean much of anything.  Total damage also isn't a very good stat as it is tier dependent, you need to look at dmg dealt/dmg taken ratio.

[Amarr HM]

Sir T those figures are basically the same, your k/d got transposed into slightly more damage. Efficiency probably went up cause it weights damage more than k/d. Come back with a k/d of at least 1.00 over 300 matches and then we'll talk.

[Abagadro]

I'm just going to ignore whatever tripe he posts in response to me as he is clearly deluded.  He can just keep on thinking every POS red pub with a 46 percent win rate is just some poor unlucky dude that the game conspires to keep down.

Oh, and on the "stats suck because of lights", I just checked my 2801 stats (Sir T's highest played light) and my k/d ratio is 1.75X higher than his entire aggregate k/d ratio, with a 56% win rate (my luchs is even crazier with a 2.04 k/d ratio and 62% win rate).

So he should just shut the hell up and stop wasting everyone's time.

[WayAbvPar]

The only thing in common in ALL your matches? You.  (Most) everyone plays lights,  (most) everyone plays arty, everyone gets great teams, everyone gets bad teams. What YOU do with them is the single greatest factor in determining your win rate. It can be gamed by doing clan wars and companies in super clans, and/or by only driving the most overpowered tanks (ever noticed the super platoons you run into are never 3 purple guys driving ARL V39s or Lorraines? They are always in T-54s or similarly great tanks), but it is a good indicator of relative skill level and experience (along with other things mentioned earlier).


That being said- don't forget to stock up in consumables before the end of the weekend (50% off). There is no excuse to ever not have a fire extinguisher, and I will mock you incessantly if I ever see you burn to death because you were too cheap to buy one. I blew my whole bankroll (a pittance compared to some of you Warbucks types at 1.9M) on them before I logged off for the night. It will save me tons over the coming weeks.

[WayAbvPar]

Just got one-shotted by a JagTiger in my T44. 1300 damage. So no more crying about arty.

[Furiously]

The only thing in common in ALL your matches? You.  (Most) everyone plays lights,  (most) everyone plays arty, everyone gets great teams, everyone gets bad teams. What YOU do with them is the single greatest factor in determining your win rate. It can be gamed by doing clan wars and companies in super clans, and/or by only driving the most overpowered tanks (ever noticed the super platoons you run into are never 3 purple guys driving ARL V39s or Lorraines? They are always in T-54s or similarly great tanks), but it is a good indicator of relative skill level and experience (along with other things mentioned earlier).


That being said- don't forget to stock up in consumables before the end of the weekend (50% off). There is no excuse to ever not have a fire extinguisher, and I will mock you incessantly if I ever see you burn to death because you were too cheap to buy one. I blew my whole bankroll (a pittance compared to some of you Warbucks types at 1.9M) on them before I logged off for the night. It will save me tons over the coming weeks.

I disagee, if I'm playing arty I might have a case of cola instead of a fire extinguisher.

[Merusk]

Just got one-shotted by a JagTiger in my T44. 1300 damage. So no more crying about arty.

I re-bought my SU-152 and am grinding to the top gun (3k more exp) and then the ISU-152.  With my T-50-2 crew retrained (1 perk and 1 skill at 85%!) it is a big FUCK YOU beast even if it is only using the that lower gun, and it stays invisible on most maps.  I'm learning the lines of fire for the new maps and all I can say is ha ha ha.

Ed: Sentences, complete them.

[WayAbvPar]

The only thing in common in ALL your matches? You.  (Most) everyone plays lights,  (most) everyone plays arty, everyone gets great teams, everyone gets bad teams. What YOU do with them is the single greatest factor in determining your win rate. It can be gamed by doing clan wars and companies in super clans, and/or by only driving the most overpowered tanks (ever noticed the super platoons you run into are never 3 purple guys driving ARL V39s or Lorraines? They are always in T-54s or similarly great tanks), but it is a good indicator of relative skill level and experience (along with other things mentioned earlier).


That being said- don't forget to stock up in consumables before the end of the weekend (50% off). There is no excuse to ever not have a fire extinguisher, and I will mock you incessantly if I ever see you burn to death because you were too cheap to buy one. I blew my whole bankroll (a pittance compared to some of you Warbucks types at 1.9M) on them before I logged off for the night. It will save me tons over the coming weeks.

I disagee, if I'm playing arty I might have a case of cola instead of a fire extinguisher.

It doesn't really matter for arty, since if you are getting shot you are fucked anyway.

[Ginaz]

The only thing in common in ALL your matches? You.  (Most) everyone plays lights,  (most) everyone plays arty, everyone gets great teams, everyone gets bad teams. What YOU do with them is the single greatest factor in determining your win rate. It can be gamed by doing clan wars and companies in super clans, and/or by only driving the most overpowered tanks (ever noticed the super platoons you run into are never 3 purple guys driving ARL V39s or Lorraines? They are always in T-54s or similarly great tanks), but it is a good indicator of relative skill level and experience (along with other things mentioned earlier).


That being said- don't forget to stock up in consumables before the end of the weekend (50% off). There is no excuse to ever not have a fire extinguisher, and I will mock you incessantly if I ever see you burn to death because you were too cheap to buy one. I blew my whole bankroll (a pittance compared to some of you Warbucks types at 1.9M) on them before I logged off for the night. It will save me tons over the coming weeks.

It's been ages since I ran across someone from f13 in battle.  It's too bad we ended up in a pretty lopsided battle.  I think it was about 5 or 6 of you against me at the end.  That seems to be how most have for me.  A total blowout, one way or the other.

[Merusk]

Fuck.. I just re-read Way's post and goddamnit I was at work most of the weekend so I didn't pay attention to the specials.  I had saved up millions JUST to buy consumables at 50%. 

Fuck.

[WayAbvPar]

The only thing in common in ALL your matches? You.  (Most) everyone plays lights,  (most) everyone plays arty, everyone gets great teams, everyone gets bad teams. What YOU do with them is the single greatest factor in determining your win rate. It can be gamed by doing clan wars and companies in super clans, and/or by only driving the most overpowered tanks (ever noticed the super platoons you run into are never 3 purple guys driving ARL V39s or Lorraines? They are always in T-54s or similarly great tanks), but it is a good indicator of relative skill level and experience (along with other things mentioned earlier).


That being said- don't forget to stock up in consumables before the end of the weekend (50% off). There is no excuse to ever not have a fire extinguisher, and I will mock you incessantly if I ever see you burn to death because you were too cheap to buy one. I blew my whole bankroll (a pittance compared to some of you Warbucks types at 1.9M) on them before I logged off for the night. It will save me tons over the coming weeks.

It's been ages since I ran across someone from f13 in battle.  It's too bad we ended up in a pretty lopsided battle.  I think it was about 5 or 6 of you against me at the end.  That seems to be how most have for me.  A total blowout, one way or the other.

LOL yeah I didn't even realize it was you until after I was dead. You should come find us on TS and get in some games with us. It is a lot of fun.

[Threash]

Everything i hear about this game is completely fucking awful and yet people seem to be having fun with it.

[Nebu]

Everything i hear about this game is completely fucking awful and yet people seem to be having fun with it.

Maybe you should try it and make up your own mind.  It's free.

[disKret]

Maybe you should try it and make up your own mind.  It's free.

And addictive as hell.

[Furiously]

It's a lot like league of legends. It should be simple. It's not.

[Threash]

Everything i hear about this game is completely fucking awful and yet people seem to be having fun with it.

Shit sorry, this actually belonged in the Final Fantasy thread.

[angry.bob]

Everything i hear about this game is completely fucking awful and yet people seem to be having fun with it.

Shit sorry, this actually belonged in the Final Fantasy thread.

Yeah, that makes a ton of sense. It does sound completely godawful. Everything about it seems designed to be as big a pointless pain in the ass as possible, and proves that many of the things that people here who are now playing and defending have said in the past are complete bullshit. It's a social game that may or may not let you playon the same server as any of your friends AND makes you play on Japanese servers? What the fuck.

[WayAbvPar]

Used 160 free XP to unlock the M53/55 this morning. No chance in hell I can actually purchase it, but it is unlocked!

[CaptainNapkin]

I've been waiting for a garage slot sale and figure the 50% off going now is probably about as good as it gets, so it's time to expand. I'm close to getting through the M3 Lee and think I've decided on the T1 Heavy next. Lee wasn't as painful as I was expecting, though I'm a fairly casual player and maybe that helps. Also started moving up the US SPG line, which is my first foray into arty and so far a nice diversion.

Has anyone given that Russian TD line a try? The lure of the current 50% credit discount and bonus has me curious.