19 replies to this topic

[_Weeping_Angel_]

The damage per tier mark from Blitzstars.com got me thinking about how to evaluate damage done in a game.  I set up a simple level system for damage that should be fair when comparing different tanks from the tiers that I play (tiers 7-10).  I just played games then marked down the damage dealt by each tank for both sides and if they won or lost.  I played premium tanks, tanks I'm grinding, tanks I like to play, and tanks I suck at but want to get better.  I just played my normal way and tracked the data as simply as possible.

 

The levels

I set up a simple ranking system based on tier x 100 to determine each level.  So tier 10 levels move up by 1,000 damage, tier 9 by 900, tier 8 by 800, and tier 7 by 700.  Here is tier 10 for example and how I labeled the levels of damage (I know I am a comedian, right).  

 

Tier X - damage needed for that level 

Suck - 0-1k dmg - what was I doing?

OK - 1k-2k dmg - decent but need to do better.

Expected - 2k-3k dmg - I did my part and nothing more.

Good - 3k-4k dmg - big contributor

Great - 4k-5k dmg - what are my teammates doing?

Amazing - 5k+ dmg - my back is sore from carrying so hard!

 

The Numbers

Here is the number of tanks that finished in each of the levels and their percentage for each tier and the totals along the bottoms.  I marked the two lowest levels to differentiate between what I consider a below average game.  The far right it the total tanks I tracked for each tier.

 

			Tanks
	Suck 0-1x dmg	OK 1x-2x dmg	Expected 2x-3x dmg	Good 3x-4x dmg	Great 4x-5x dmg	Amazing 5x+ dmg	Totals
Tier 10	132 (27%)	150 (31%)	131 (27%)	50 (10%)	18 (4%)	4 (0.8%)	485
Tier 9	160 (29%)	210 (39%)	119 (22%)	40 (7%)	14 (3%)	3 (0.5%)	546
Tier 8	240 (35%)	257 (37%)	148 (21%)	40 (6%)	8 (1%)	2 (0.3%)	695
Tier 7	129 (43%)	110 (36%)	44 (15%)	16 (5%)	3 (1%)	0 (0%)	302
Totals	661 (32.59%)	727 (35.85%)	442 (21.79%)	146 (7.20%)	43 (2.12%)	9 (0.44%)	2028

 

Total tanks below expected level damage - 1,388 (68.44%)

Total tanks at expected level damage and above - 640 - (31.56%)

 

Here is the breakdown for WR for each tier and level of damage for the games I tracked.

 

			WR
	Suck	OK	Expected	Good	Great	Amazing
Tier 10	28.03%	45.33%	63.365	76.00%	66.67%	50%
Tier 9	31.88%	50%	60.50%	75%	78.57%	100%
Tier 8	38.33%	49.81%	56.76%	80%	87.50%	50%
Tier 7	41.86%	57.27%	59.09%	81.25%	66.67%	---
Total	35.67%	50.07%	60.92%	77.40%	74.42%	66.67%

 

What I see

The Good

The skill to deal damage is almost linear in numbers as you go up the tiers.  There is a higher percentage of tanks that can deal more damage as you go up each tier for all levels above Expected (2x the tier of damage or more).  There are fewer tanks that do poorly at tier X compared to VII.  Each time you move up a level the players are slightly better than the one before.  I was just shocked at how clear and clean this was on the graph.

The Bad

Look at the Suck level.  Those poor tanks who only get off one or two shoots before dying.  As you go down in tiers, the percentage of players in that category rises at each tier.  I was surprised at the sheer number of players That don't do well in each game.  Over two thirds of tanks don't even to 2x their tier in damage each game. I knew tanks did bad when you look at the end game screen but I was surprised by the sheer numbers of players in the first two levels.

Conclusions

I think this flat out shows that there are more skilled players at each tier as you go up from 7 to 10.  The amount of players who do poorly goes down as you move up the tiers so the amount of players who do well goes up.  This is vindication for the players who say the best players are playing up at tier X.  You need to pull your weight or you lose.  Tier X has the lowest WR for both the games you SUCK or do OK damage.  Tier X is brutal and you need to play well to win.  You want to get better as a player make sure you are not finishing in those two levels for damage as often.

 

How did I do?

I was on fire during this charting so I know my WRs for this will be way better than normal.  I could not lose when dealing just OK damage.  I also don't have a huge sample size for each tier as I played 4 different tiers and only tracked 151 games for myself (20 @ X, 67 @ IX, 45 @ VIII, 19 @ VII).  Not a huge sample size at any tier and even worse for each category to determine an accurate WR for myself.  Therefore only totals are really with looking at.

 

	SUCK	OK	EXPECTED	GOOD	GREAT	AMAZING
GAMES (151)	18 (12%)	42 (28%)	51 (34%)	28 (19%)	11 (7%)	1 (0.7%)
WR	38.89%	61.90%	60.78%	89.29%	81.82%	100%

 

I was happy to see that I was well below the average tank when it comes to playing badly.  I only had really bad games 12% of the time compared to 33% of the general population.  That also means that I flipped the scales and played at my expected level or better 60% of the time compared to only 32% of the time for the average player.

 

What did I learn

I also wrote down what happened each time I finished in the SUCK level of damage.  A TD ruined my game 11 of the 18 times.  The usual suspect was the 183 and waffle.  Even worse was 7 of those 11 times the TD was not spotted so I never saw it coming.  Conclusion... The TD class is doing it's job.  I need to work on my awareness of where they are and where they like to camp and don't give them an easy target.  I still had games where I survived a TD shot and did well but that shot early or the big finishing shot is what caused me to SUCK 61% of those games.  The second leading cause was sticking my nose where it doesn't belong.  I would convince myself to try to take a shot when I know I will get hit by at least 1 shot in return.  I get tracked and then take 2 or 3 shots.  It always went way better in my head before I did it.

Edited by _Weeping_Angel_, 16 August 2018 - 06:24 PM.

[_Weeping_Angel_]

saved for comments or additional info

Edited by _Weeping_Angel_, 16 August 2018 - 03:10 PM.

[SpartacusDiablo]

I suck

[krietenstein34]

Let me paraphrase...........   I looked a blitzstars, and in a shorter amount of time it's going to take me to type this sentence, I've concluded you don't suck

[Untrust______]

This dosent prove best players are at tier 10. It proves you can type a bunch of nonsense and say it proves somthing. 

[cheesehead247365]

You don't suck. Your stats are good.  

[CK_GoldenNuggets]

Untrust______, on 16 August 2018 - 03:17 PM, said:

This dosent prove best players are at tier 10. It proves you can type a bunch of nonsense and say it proves somthing. 

 

Don't see how his post is nonsense. It was small scale and would need a lot of continuous data to prove his point but he is right nonetheless. Better players tend to gravitate towards tier 10 for the challenge and worse players or seal clubbers stay at low tiers for the exact opposite reason.

There is little to no challenge at low tiers because players are worse at lower tiers, they have less game knowledge and experience.
Makes sense, since you can grind from tier 1 - 5 in less than a day, where as most players take well over a month of dedicated grinding to get to tier 10 depending on their skill level. Yes there are still plenty of crap players at tier 10 too, but most good players tend to just play tiers 8 and above because its the only tiers worth playing. Grinding credits at 8, and playing 10 for the challenge.

 

It takes way more skill to get a 50% winrate in any given tier 10 then it does to get a 50% winrate at tier 5 - 7. Its a whole different ballgame.

Edited by CK_GoldenNuggets, 16 August 2018 - 03:53 PM.

[_Crusader6_]

Untrust______, on 16 August 2018 - 10:17 AM, said:

This dosent prove best players are at tier 10. It proves you can type a bunch of nonsense and say it proves somthing. 

 

Tier X has always been the playground of fools and Unicums. 

  There is very little middle ground. 

 

[DanAnsomeHansome]

Wouldn’t it be more simple to say that dead average is averaging your tank’s hp pool in average damage ?

[__Synx__]

__Crusader6__, on 16 August 2018 - 09:54 AM, said:

 

Tier X has always been the playground of fools and Unicums. 

  There is very little middle ground. 

 

 

Very true statement here. You either don't understand anything about game mechanics or strategies... or you understand them and is rising towards 60% or somewhere around that. However, there is some middle ground for players that are rising in their winrates, like 55%ers that are still rising. Then there are two kinds of fools: ones that don't know any better than to "play tenks" or the ones that do know but choose to remain fools and troll people because they are just twisted that way.

[LtRaseiniai]

You’re wasting your time.

You need to make four alternate accounts.

Get them all in the same battles.

And farm damage for your main account.

[LordManaz]

I am the type of player who believes that doing a lot of damage in a battle does not automatically mean that you will win or help the team. (Im always in the top 3 in damage at the end of the battle)

 

example: if i hit the tracks and the entire team vanish a 183 i only shot +300 damage, if i spot a grille and my tds tag him and i shot +100 just to finish him at the end of the battle i only shot 2.5k in damage but i help my team, i make them work

Edited by LordManaz, 16 August 2018 - 04:37 PM.

[_Weeping_Angel_]

Untrust______, on 16 August 2018 - 09:17 AM, said:

This dosent prove best players are at tier 10. It proves you can type a bunch of nonsense and say it proves somthing. 

 

I don't think it was nonsense.  I was trying to make myself better by comparing damage of a normal tank vs what I normally do.  Is damage everything that goes into winning? No.  But is it one of the easiest things to track? Yes.  WG even moved the finished battle screen to rank players by damage instead of XP so it was an easy thing to track.  I just decided to share at this point because I had enough to make it decent where the stats stopped fluctuating as much.  I is pretty obvious that if you do more damage per game your WR should go up.  I was curious how many people really do bad in a game.  It seems like there are so many tanks that don't even do 1k damage when I play upper tiers that I wanted to see exactly how many.

 

DanAnsomeHansome, on 16 August 2018 - 09:59 AM, said:

Wouldn’t it be more simple to say that dead average is averaging your tank’s hp pool in average damage ?

 

Yes, sort of.  That would have made tracking this a complete nightmare.  I would have to set up a system to track each tank individually as almost every tank has  different HPs and then setting up the levels for each tank.  I still want to play the game and not spend 3 days setting up spreadsheets or paper to track each tank that way.  I set it up this way after seeing the stat on Blitzstars where it shows your damage per level.  I was hovering around 200 so I figured that is 2x the level for a tank so I set that as my benchmark.  That is probably high compared to what the average tanker does in game per level for battle but I am not trying to prove WG is rigging anything.  So there is no stat analysis to really do.  Just comparing myself to others.

 

LordManaz, on 16 August 2018 - 10:30 AM, said:

I am the type of player who believes that doing a lot of damage in a battle does not automatically mean that you will win or help the team. (Im always in the top 3 in damage at the end of the battle)

 

example: if i hit the tracks and the entire team vanish a 183 i only shot +300 damage, if i spot a grille and my tds tag him and i shot +100 just to finish him at the end of the battle i only shot 2.5k in damage but i help my team, i make them work

 

There are a ton of way to help win a battle.  Damage is just one of them.  The list of little things to help is endless.... clearing out tanks by focus fire, spotting, track shots, just plain being a distraction, flanking, capping, turning guns, COD, etc. .  I was just picking one stat to track for awhile and see what happened.  But the point of doing all those little things is to help others deal damage.  That is why tanks that finished in the SUCK level can still win around 35% of the time.  Not all of them are AFK.  Sometimes those are tanks who sacrifice damage for other intangibles to help win.

Edited by _Weeping_Angel_, 16 August 2018 - 05:06 PM.

[krietenstein34]

If i do 3K damage anytime any tier, I've kicked mudda ducking ace

[ahredstealth]

Is it just me or does your tier 7 seem skewed?

 

What conclusions are you trying to state about the TDs that took you out at tier 10?  I'm not quite following.

[_Crusader6_]

_Weeping_Angel_, on 16 August 2018 - 11:46 AM, said:

 

 

There are a ton of way to help win a battle.  Damage is just one of them.  The list of little things to help is endless.... clearing out tanks by focus fire, spotting, track shots, just plain being a distraction, flanking, capping, turning guns, COD, etc. .  I was just picking one stat to track for awhile and see what happened.  But the point of doing all those little things is to help others deal damage.  That is why tanks that finished in the SUCK level can still win around 35% of the time.  Not all of them are AFK.  Sometimes those are tanks who sacrifice damage for other intangibles to help win.

 

I like dmg -- but dmg in and of itself is not a great tool.

  Folks who play a lot of TD's can rack up significantly high damage results and if they camp it can be rather empty damage.

 

For X in a TD I want at least 2,700 dmg in a TD, while for a Med I will take 2,300 and a bunch of assist damage.

   

[_Weeping_Angel_]

ahredstealth, on 16 August 2018 - 12:03 PM, said:

Is it just me or does your tier 7 seem skewed?

 

What conclusions are you trying to state about the TDs that took you out at tier 10?  I'm not quite following.

 

skewed?  I am assuming the higher number of tanks in the SUCK level vs OK level.  I have a feeling it is due to the tier 8 effect.  So many players grind credits and have tier 8 premiums that tier 7 tends to get bottom tier more often than other tiers.  That is my guess.

 

I will go back and edit it to make it more clear.  I think TDs are doing their job in the game.  Get smacked early by a TD and it makes it just that much harder to for a tank to do  large amounts of damage back before they die.   Get sniped late in a game to take a weak HP tank off the map from dealing more damage.  Most of my bad performances were from TDs doing their job.  I was just saying that getting ripped for huge chuck of my HP at any point was the most effective way for me to have a bad damage game.  I realize I need to be more aware of where they are or where they are going to snipe from to reduce their ability to make me have a bad game.

[ahredstealth]

_Weeping_Angel_, on 16 August 2018 - 12:22 PM, said:

 

skewed?  I am assuming the higher number of tanks in the SUCK level vs OK level.  I have a feeling it is due to the tier 8 effect.  So many players grind credits and have tier 8 premiums that tier 7 tends to get bottom tier more often than other tiers.  That is my guess.

 

I will go back and edit it to make it more clear.  I think TDs are doing their job in the game.  Get smacked early by a TD and it makes it just that much harder to for a tank to do  large amounts of damage back before they die.   Get sniped late in a game to take a weak HP tank off the map from dealing more damage.  Most of my bad performances were from TDs doing their job.  I was just saying that getting ripped for huge chuck of my HP at any point was the most effective way for me to have a bad damage game.  I realize I need to be more aware of where they are or where they are going to snipe from to reduce their ability to make me have a bad game.

 

Skewed may not be the correct word...

 

Yes, I was talking about the general slide downward.  More bad/ok games.  No amazing games.

 

Perhaps 700 per level at tier 7 doesn't work?  Or did you have a rough time?  Just curious.

[minitel_NA]

I found the article well worth a read. 

People who want to improve should be analytical whether it is through personal stats, strategies, tank specs or watching a 1000 replays.

 

what I didn’t get is why such a small sample of battle, and also the overall reasoning is quite original — doesn’t remove merit.

 

 

overall a good read. 

[_Weeping_Angel_]

Here a game I just finished and is a perfect sample of tier 7 tanks as bottom tier. This type of game is is exactly the reason I started collecting data. This type of results is common.

 

As far as stopping at only 150 games, I was switching to help Mpastor2000 track more WR for teams. I thought 2000 was enough for totals tanks to be accurate but a little light in individual tiers and myself. I am not sure if doing this will change the stats significantly since I was not planning to do any mathematical analysis. 

Edited by _Weeping_Angel_, 16 August 2018 - 08:02 PM.