Jump to content
Sign in to follow this  
You need to play a total of 5 battles to post in this section.
BullpupWOT

What gets the weight in WTR calculations?

29 comments in this topic

Recommended Posts

Members
674 posts
1,254 battles

A one battle session this morning got me wondering what gets the most weight in a WTR calculation. I'm guessing a win is very big?  I only did 12k damage and sunk 2 BBs in a win with the Aigle. That translated to a very high one battle WTR that I was expecting to only be in the 900s based on a lack of damage. 

I know it's probably a complicated calculation, but what is getting the heaviest weight?

Share this post


Link to post
Share on other sites
Members
674 posts
1,254 battles
8 minutes ago, RipNuN2 said:

I'll read it, but I probably won't comprehend what I'm reading. Thank you though as I had trouble locating that. 

I was looking on the other site as well. 

Edited by BullpupWOT

Share this post


Link to post
Share on other sites
43
[SP3AR]
Supertester
215 posts
6,159 battles

Basically its based on the average damage, kills, and wr (50%, 30%, 20% respectively). If you do better than the server average, you have a higher wtr.

Edited by EpicLeveler

Share this post


Link to post
Share on other sites
723
[NEUTR]
Members
2,207 posts
8,700 battles

LOL "a win is very big"

 

on contraire mon capitan

 

 

 

You can get unicum rating with less than 45% win rate, damage done is everything. Technically you can get unicum rating with zero percent win rate. For an example, see this picture below:

 

yW9v76M.png

 

From what glanced over on their forum, they did a regression on the player data and found damage done is more likely change with win rates. Reasonable, based on the stats and result, however they might have forgotten the age old statistical lesson of Abraham Wald: https://www.jstor.org/stable/2288257?seq=1#page_scan_tab_contents

Operations research is a tricky matter.

 

The Warships rating is basically based on the same in WoT. A horrendous 3rd party metric and is a detriment to the game's meta. Poisons player bahaviors and making a game that's already not rewarding team play even more selfish.

Edited by NeutralState
  • Cool 2
  • Funny 1

Share this post


Link to post
Share on other sites
Members
674 posts
1,254 battles
24 minutes ago, NeutralState said:

LOL "a win is very big"

 

on contraire mon capitan

 

 

 

You can get unicum rating with less than 45% win rate, damage done is everything. Technically you can get unicum rating with zero percent win rate. For an example, see this picture below:

 

yW9v76M.png

 

From what glanced over on their forum, they did a regression on the player data and found damage done is more likely change with win rates. Reasonable, based on the stats and result, however they might have forgotten the age old statistical lesson of Abraham Wald: https://www.jstor.org/stable/2288257?seq=1#page_scan_tab_contents

Operations research is a tricky matter.

 

The Warships rating is basically based on the same in WoT. A horrendous 3rd party metric and is a detriment to the game's meta. Poisons player bahaviors and making a game that's already not rewarding team play even more selfish.

Yes, I'm not getting wrapped up in WTR. I was just floored that a 12k damage win resulted in a 1440 WTR in that Aigle battle. The most important was that I played smartly and did my job as a DD to help the team win. I've had a decent two days ever since my thread on radar/hydro slapped me into the right mindset for helping the team win. 

  • Cool 1

Share this post


Link to post
Share on other sites
568
[CVA16]
Members
2,785 posts
8,410 battles
1 hour ago, BullpupWOT said:

A one battle session this morning got me wondering what gets the most weight in a WTR calculation. I'm guessing a win is very big?  I only did 12k damage and sunk 2 BBs in a win with the Aigle. That translated to a very high one battle WTR that I was expecting to only be in the 900s based on a lack of damage. 

I know it's probably a complicated calculation, but what is getting the heaviest weight?

Bullpup, Looked at your stats there off the site Ripnu posted. Your'e doing good man. 900 battles and your numbers are all pretty much green since Jan. 8. Well done Captain, well done.

  • Cool 1

Share this post


Link to post
Share on other sites
405
[ICOP]
[ICOP]
Members
1,042 posts
15,426 battles
21 minutes ago, BullpupWOT said:

Yes, I'm not getting wrapped up in WTR. I was just floored that a 12k damage win resulted in a 1440 WTR in that Aigle battle. The most important was that I played smartly and did my job as a DD to help the team win. I've had a decent two days ever since my thread on radar/hydro slapped me into the right mindset for helping the team win. 

So many players (I struggle with it, myself) never learn that lesson, brother. That you have puts you miles ahead of many others, imo. Flamu just did a video about the T-22 (German DD) that really emphasizes the importance of a focus on a team winning, rather than a focus on kills or damage farming. Thought you might appreciate seeing it, if you have not already done so.

Respects,

Am

  • Cool 1

Share this post


Link to post
Share on other sites
245
[BNKR]
Members
544 posts
3,471 battles
1 hour ago, RipNuN2 said:

Thank you!

Question: Is there a site like Target Damage for tanks that tells you what to aim for damage-wise? I like to have an idea where I am at as a game is progressing so I know if I am actually sucking as much as I think I am, or am actually doing okay. :Smile_teethhappy:

Share this post


Link to post
Share on other sites
5,734
[OO7]
Members
6,005 posts
13,026 battles

WTR is not a great metric. Too much emphasis on damage and kills and not on wins and XP. XP is impossible to do, because as I understand it, the API doesn't differentiate between standard and premium accounts. Wins can be somewhat luck based with a low number of games, but over the long haul its the best.

 

Basically you will get lower WTR as a BB if you do 50k damage to 4 dd's and team mates finish them off leading to a win, than if you do 150k damage to enemy bb's, get 2 kills and lose.

Edited by Ducky_shot

Share this post


Link to post
Share on other sites
864
[MPIRE]
Beta Testers
3,791 posts
5,911 battles

Ignore WTR. The only thing that matters is W/R.

 

If you are a good player and contribute to your team, it'll show.

  • Cool 1

Share this post


Link to post
Share on other sites
5,734
[OO7]
Members
6,005 posts
13,026 battles
1 minute ago, MajorEnglush said:

Thank you!

Question: Is there a site like Target Damage for tanks that tells you what to aim for damage-wise? I like to have an idea where I am at as a game is progressing so I know if I am actually sucking as much as I think I am, or am actually doing okay. :Smile_teethhappy:

Warships today tells you the average damage in each ship.

Share this post


Link to post
Share on other sites
568
[CVA16]
Members
2,785 posts
8,410 battles
2 minutes ago, aether_tech said:

Ignore WTR. The only thing that matters is W/R.

 

If you are a good player and contribute to your team, it'll show.

Aether, I got to ask this. I've been seeing your posts since I started playing in March of 16. That is one wild looking outfit in your avatar! I think I finally saw it today with enough beer drinking to ask what it is.

Share this post


Link to post
Share on other sites
864
[MPIRE]
Beta Testers
3,791 posts
5,911 battles
Just now, Ghostdog1355 said:

Aether, I got to ask this. I've been seeing your posts since I started playing in March of 16. That is one wild looking outfit in your avatar! I think I finally saw it today with enough beer drinking to ask what it is.

It's a me`-a, mario.

Share this post


Link to post
Share on other sites
245
[BNKR]
Members
544 posts
3,471 battles
12 minutes ago, Ducky_shot said:

Warships today tells you the average damage in each ship.

I'll check it out. Would like some sort of chart that lists yellow, green, blue, and purple damage targets, but will work with what they got I guess.

 

Thanks! :Smile_great:

Share this post


Link to post
Share on other sites
331
[PEED2]
Beta Testers
1,718 posts
21,096 battles
15 minutes ago, aether_tech said:

Ignore WTR. The only thing that matters is W/R.

 

If you are a good player and contribute to your team, it'll show.

This

Share this post


Link to post
Share on other sites
5,734
[OO7]
Members
6,005 posts
13,026 battles
1 minute ago, MajorEnglush said:

I'll check it out. Would like some sort of chart that lists yellow, green, blue, and purple damage targets, but will work with what they got I guess.

 

Thanks! :Smile_great:

Oh, you are talking about where to aim on certain ships? You can look at the armor for any ship in your port. But angles and such aren't going to be easy to tell. Just remember that your HE penetration is 1/6 of your shell size on everything but German ships and British Battleships, on them its 1/4. AP will overmatch any armor that is 1/14.3. But as far as figuring out penetration, most of those stats are hidden 

Share this post


Link to post
Share on other sites
8,127
[WOLF3]
Members
22,816 posts
20,723 battles

WR% is the main defining thing, IMO.  You can start getting into the weeds with other stats and get a feel about how that player is, especially ship specific stats.  But WR% is your cover sheet, your executive summary.

 

WTR... Has some uses but it kind of screws over some ships and playstyles.  DDs in particular.  You can have DDs doing well but their damage output, kills can be pretty inconsistent, that alone hurts WTR.  There can be games where the DD is lucky to get even ONE torpedo hit, much less rack up damage and kills.  Yet they may be playing the cap game well and spotting for the team, spotting torps, etc, and WTR doesn't truly reward stuff like that.

Share this post


Link to post
Share on other sites
1,110
Members
2,847 posts
6,470 battles

My WTR is yellow or orange, less than 1,000 in any case. 

My XP overall though is 1200, and In DDs my WR is 54% and I think my DD XP is over 1400.

my average damage overall is only like 24k. I smoke, I cap, I keep stuff spotted. I play solo, and all that matters is winning. Unless you’re trying to impress a clan, what else matters?

Share this post


Link to post
Share on other sites
Members
5,830 posts
6,711 battles
1 hour ago, BullpupWOT said:

Yes, I'm not getting wrapped up in WTR. I was just floored that a 12k damage win resulted in a 1440 WTR in that Aigle battle. The most important was that I played smartly and did my job as a DD to help the team win. I've had a decent two days ever since my thread on radar/hydro slapped me into the right mindset for helping the team win. 

Part of WTR is based on a comparison of others' performance in a given ship as well. Take Furutaka for example. Her overall stats are forever plagued by her existance pre-C hull. That drags her overall down quite a bit as Furutaka used to be a radical change from both Kuma and Aoba. Playing a relatively average to slightly above average match in Furutaka will net you deep purple and a high WTR, and this is pretty easy given that the C hull is essentially a down tiered Aoba.

Playing a ship with a low battle count, or that's not widely owned can sway overall and personal WTR wildly. Aigle probably isn't widely owned nor being played a whole lot yet, so low damage games can result in really high WTR if the overall stats are actually below average(what it would/should be if it were being played a lot more).

Share this post


Link to post
Share on other sites
Members
1,819 posts
1 hour ago, aether_tech said:

Ignore WTR. The only thing that matters is W/R.

 

If you are a good player and contribute to your team, it'll show.

W/L is up to 4% + or - your actual skill based on nothing but luck, how do you factor that?

If you too 10k bots and had them play 1000k games, some would have W/L of 46 and some would have 54 for no other reason than random chance. 

Share this post


Link to post
Share on other sites
864
[MPIRE]
Beta Testers
3,791 posts
5,911 battles
48 minutes ago, Kconvey_1 said:

W/L is up to 4% + or - your actual skill based on nothing but luck, how do you factor that?

If you too 10k bots and had them play 1000k games, some would have W/L of 46 and some would have 54 for no other reason than random chance. 

That is the dumbest thing ive ever read about w/r.

Share this post


Link to post
Share on other sites
864
[MPIRE]
Beta Testers
3,791 posts
5,911 battles
51 minutes ago, Kconvey_1 said:

W/L is up to 4% + or - your actual skill based on nothing but luck, how do you factor that?

If you too 10k bots and had them play 1000k games, some would have W/L of 46 and some would have 54 for no other reason than random chance. 

Ever.

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×