Jump to content
You need to play a total of 5 battles to post in this section.
Kombat_W0MBAT

WTR = Skill

44 comments in this topic

Recommended Posts

3,322
[-K-]
Supertester
5,153 posts
8,946 battles

False.

Did that trigger you though? 

While there is a loose correlation between WTR and in-game ability, it's hardly a direct correlation. My guess is that it's accurate to within +/-200. Most people already know this, but it doesn't hurt to bring it up again. As I've stated before, the WarshipsToday leaderboard is entirely meaningless. Want to end up on there? Play destroyers well. Play a terrible ship that was recently buffed (i.e. USN CVs). Don't play the overperforming ships. Don't play good ships that were recently nerfed. And god forbid you should play the Flint or the Black! That'll send your WTR right to Davy Jones' Locker.

Why am I making this post? Because it's worth repeating. The catalyst was having someone call out in all-chat this morning "Ermahgerd, you're #5 in NA". Okay... I'm really not. I don't even know that I'm top 100. There are a TON of extremely talented players in this game that would open a can of woop@ss on me. I started out this game playing USN DDs a LOT. This artificially inflates that number.

Statistics aren't evil. As an engineer,  I quite like them. But the interpretation of said statistics are highly subjective... ESPECIALLY when viewed through the lense of third party rating systems.

Enjoy the game, play your best, and don't get too consumed in numbers that don't mean all that much.  

Edited by Kombat_W0MBAT
  • Cool 15

Share this post


Link to post
Share on other sites
175
[TF16]
Members
826 posts
5,078 battles

Wasn't triggered but came in looking for a fun topic.  

Then I read your post.

Well put together words you get here!

Share this post


Link to post
Share on other sites
21
[SPTR]
Members
124 posts
4,523 battles
8 minutes ago, Kombat_W0MBAT said:

Statistics aren't evil. As an engineer,  I quite like them. But the interpretation of said statistics are highly subjective... ESPECIALLY when viewed through the lense of third party rating systems.

But the Franklin - Covey method of science is soo much fun (determine the result you want then find the test/data that supports it).

Otherwise great post M. WOMBAT.

Share this post


Link to post
Share on other sites
774
[SF-A]
Members
2,893 posts
5,686 battles

It's really only useful for general trends with extreme differences. I'll take the teammate with a 1400 WTR over 600 WTR every time, but things like detonations, RNG, and team play can mean the 1400 guy loses. It's far from set in stone.

Share this post


Link to post
Share on other sites
474
[LUCK]
Members
1,300 posts
19,563 battles

Another EF Hutton post.

(for those who don't know, search EF Hutton commercials and tagline from the 80's)

 

 

Share this post


Link to post
Share on other sites
354
[D12]
[D12]
Beta Testers, In AlfaTesters
3,083 posts
8,964 battles
14 minutes ago, Kombat_W0MBAT said:

False.

Did that trigger you though? 

While there is a loose correlation between WTR and in-game ability, it's hardly a direct correlation. My guess is that it's accurate to within +/-200. Most people already know this, but it doesn't hurt to bring it up again. As I've stated before, the WarshipsToday leaderboard is entirely meaningless. Want to end up on there? Play destroyers well. Play a terrible ship that was recently buffed (i.e. USN CVs). Don't play the overperforming ships. Don't play good ships that were recently nerfed. And god forbid you should play the Flint or the Black! That'll send your WTR right to Davy Jones' Locker.

Why am I making this post? Because it's worth repeating. The catalyst was having someone call out in all-chat this morning "Ermahgerd, you're #5 in NA". Okay... I'm really not. I don't even know that I'm top 100. There are a TON of extremely talented players in this game that would open a can of woop@ss on me. I started out this game playing USN DDs a LOT. This artificially inflates that number.

Statistics aren't evil. As an engineer,  I quite like them. But the interpretation of said statistics are highly subjective... ESPECIALLY when viewed through the lense of third party rating systems.

Enjoy the game, play your best, and don't get too consumed in numbers that don't mean all that much.  

I did the exact opposite, the first line I ran?  IJN DDs.   No guns all torpedoes.  I hardly played Battleships for the first 3 months. Hell, I have as almost as many games in my Fujin than my combined games in Battleships.   Screw K/D ratios, I would like to go by Exp earned, but that is inflated with premium time.

Share this post


Link to post
Share on other sites
1,672
[OO7]
Members
2,229 posts
9,721 battles

Most of the requirements necessary to achieve a high WTR require skill.  You even contradicted yourself:

 

"Want to end up on there? Play destroyers well."

 

Yes you can manipulate WTR, but all that does is raise the standard of what a "good" WTR is.  That is separate from skill.  If you had Team A where all players had a WTR of 1300 or above play against Team B where all players had a WTR of 600 or below, team A is going to slaughter Team B every time.  It takes an incredibly irrational person to dispute that and if they do, you know they cannot be reasoned with.

  • Boring 3

Share this post


Link to post
Share on other sites
294
[O7]
[O7]
Beta Testers
1,252 posts
7,736 battles

Yep, makes sense, but do like seeing on the pretty purple on the sig. I do find that WTR is a rough guess on how good people are in certain ships but it isn't everything. Actually when I want to compare I generally look at WR and average damage to see where I stack up against another player.

Share this post


Link to post
Share on other sites
3,322
[-K-]
Supertester
5,153 posts
8,946 battles
1 minute ago, VGLance said:
Spoiler

 

Most of the requirements necessary to achieve a high WTR require skill.  You even contradicted yourself:

 

"Want to end up on there? Play destroyers well."

 

Yes you can manipulate WTR, but all that does is raise the standard of what a "good" WTR is.  That is separate from skill.  If you had Team A where all players had a WTR of 1300 or above play against Team B where all players had a WTR of 600 or below, team A is going to slaughter Team B every time.  It takes an incredibly irrational person to dispute that and if they do, you know they cannot be reasoned with.

 

 

 

Did you read the part where I said there is a loose correlation? Nobody is disputing that a 1300 player is better than a 600 player. 

Share this post


Link to post
Share on other sites
272
[PT8TO]
Members
1,410 posts
13,256 battles

Whats that saying?  You can not judge a book by its cover?

Lots of players in this game have what the rating scale would consider them to be average maybe below average.

Maybe they chose a path that was a hard road and had lots of bumps along the way and their WTR may never recover.

But those are the ones you have to watch out for and not just look at some number and think they are gonna roll over.

I see a player with a horrible WTR overall and then notice they carry Tier X matches almost every game.

This makes me realize that WTR and WR  can not be used to judge everyone.

You must dig deeper than that and really look at player to determine their status on this server.

If nothing else first hand experiences with or against a player usually decide if that number is worth the digital paper it is printed on.

 

Share this post


Link to post
Share on other sites
1,672
[OO7]
Members
2,229 posts
9,721 battles
Just now, Kombat_W0MBAT said:

 

Did you read the part where I said there is a loose correlation? Nobody is disputing that a 1300 player is better than a 600 player. 

Your argument is to claim that there is a binary result from the equation WTR = skill.  True or false.   It is not binary.  It's is one of many metrics used to show past performance and forecast future performance.  And skill is a major factor in performance.

  • Boring 2

Share this post


Link to post
Share on other sites
22,875
[HINON]
Supertester
19,207 posts
12,733 battles
24 minutes ago, Kombat_W0MBAT said:

ESPECIALLY when viewed through the lense of third party rating systems.

Especially especially when said third party rating system is as flawed and based on limited data as WTR is.

Good post. +1

  • Cool 1

Share this post


Link to post
Share on other sites
361
[5D]
Members
1,245 posts
7,213 battles
2 minutes ago, Lert said:

Especially especially when said third party rating system is as flawed and based on limited data as WTR is.

Good post. +1

 

Why is wtr data limited?

Share this post


Link to post
Share on other sites
1,672
[OO7]
Members
2,229 posts
9,721 battles
1 minute ago, Lert said:

Especially especially when said third party rating system is as flawed and based on limited data as WTR is.

Good post. +1

It depends on the third party program.  The one I use is extremely detailed.  Shows WR, WTR, Number of Battles, K/D, and Avg XP in BOTH the ship they brought to battle and overall.   Just one stat won't tell you much at all, but combining all ten (and yes avg xp is affected by premium time but the other stats make it obvious who's using premium and who isn't) paints a very high resolution image of what I call "expected output."  But people often use that interchangeably with skill because it's a simpler, shorter word.  Skill greatly impacts expected output, but all players, including unicums can have a derp match and a potato gets some wood on the ball every now and then.

Share this post


Link to post
Share on other sites
22,875
[HINON]
Supertester
19,207 posts
12,733 battles
Just now, Spartias said:

Why is wtr data limited?

It counts damage, kills and wins. It doesn't count defenses, captures, scouting, spotting damage, smoking your allies ...

Share this post


Link to post
Share on other sites
3,322
[-K-]
Supertester
5,153 posts
8,946 battles
Just now, VGLance said:

Your argument is to claim that there is a binary result from the equation WTR = skill.  True or false.   It is not binary.  It's is one of many metrics used to show past performance and forecast future performance.  And skill is a major factor in performance.

 

You're really an expert at splitting hairs. I never said that WTR means nothing. I simply said that there isn't a DIRECT correlation between WTR and skill. Yes/no questions are all binary. Is WTR a direct correlation of skill. No, it isn't. See? Binary. 

I already said that a correlation exists. WTR  skill. How closely it correlates is a different subject entirely. 

Share this post


Link to post
Share on other sites
1,672
[OO7]
Members
2,229 posts
9,721 battles
2 minutes ago, Lert said:

It counts damage, kills and wins. It doesn't count defenses, captures, scouting, spotting damage, smoking your allies ...

And defenses, captures, spotting damage, smoking your allies all increase your chances to win which affects WTR.  Can it be improved, yes.  But let's not kid ourselves.   Look at any player's WTR and WR graphs (over a large enough sample size) and they all move parallel for a reason.  And because they do, we know who produces results and who doesn't (aka who is skilled and who isn't).

Share this post


Link to post
Share on other sites
22,875
[HINON]
Supertester
19,207 posts
12,733 battles
1 minute ago, VGLance said:

And defenses, captures, spotting damage, smoking your allies all increase your chances to win which affects WTR.  Can it be improved, yes.  But let's not kid ourselves.   Look at any player's WTR and WR graphs (over a large enough sample size) and they all move parallel for a reason.  And because they do, we know who produces results and who doesn't (aka who is skilled and who isn't).

Not denying. Just saying that WTR uses a very limited set of data to come to a number, and that that's one of the reasons it's flawed.

Share this post


Link to post
Share on other sites
1,064
[OPG]
Members
3,970 posts
5,672 battles
20 minutes ago, VGLance said:

Most of the requirements necessary to achieve a high WTR require skill.  You even contradicted yourself:

 

"Want to end up on there? Play destroyers well."

Well to be fair, sometimes all it takes is playing a poorly performing ship half decently.  For example, take the Nurnberg.  Its server average damage is only 25k, but it also has decent ranged guns and a respectable DPM.  It's incredibly easy to just spam HE from max range and absolutely eclipse the server average in damage, which is a surefire way to net a super unicum WTR.

*edit

Sure, it does take some skill to play the Nurnberg with out being deleted in the first 5 minutes, but I'd argue the skill it takes to spam HE in the Nurnberg is not enough to warrant being the "super unicum" WTR suggests you are.  

Edited by yashma

Share this post


Link to post
Share on other sites
361
[5D]
Members
1,245 posts
7,213 battles
Just now, Lert said:

It counts damage, kills and wins. It doesn't count defenses, captures, scouting, spotting damage, smoking your allies ...

 

Damage is central to making red ships go away.

 

Kills are very important. The amount of times I've slipped away because I wasn't spammed by everyone astounds me. Half of the time that I do really well it's because I wasn't focused... because people were politely not trying to kill steal. Kill securing can win games. It's extremely important.

 

Wins are the reason why a lot of players even hit the battle button.

 

Smoking your allies isn't readily quantifiable, so that's out of the window. Even if it was, it'd be too easily abused.

 

Defenses, captures, scouting, and scouting damage are all role specific. Everyone can do them all, but not equally. Not even close. They don't make good variables in a semi-universal rating scale.

 

The data in the available war gaming API is consistent and good. It is only limited by how wargaming only broadcasts the last 30 days. This means that 3rd party sites have to build internal databases. Since warships.today is the most popular, it is by far the least limited.

Share this post


Link to post
Share on other sites
1,672
[OO7]
Members
2,229 posts
9,721 battles
2 minutes ago, yashma said:

Well to be fair, sometimes all it takes is playing a poorly performing ship half decently.  For example, take the Nurnberg.  Its server average damage is only 25k, but it also has decent ranged guns and a respectable DPM.  It's incredibly easy to just spam HE from max range and absolutely eclipse the server average in damage, which is a surefire way to net a super unicum WTR.

I am totally in agreement over this, same with Tashkent and other "WTR farming ships."  But you're dealing with an extremely small percentage of players who WTR farm, so small in fact that it's moot to argue that it can't be used as valuable metric.  And besides, it takes all of two seconds to review the rest of the person's stats, including ships played recently, to tell if they have been WTR farming.

And all things considered, you still have to have a decent brain on your shoulders to farm WTR successfully, including in the Nurnberg.  Think about all the countless 45% and below win rate players with sub 700 WTR and a 15% hit ratio who suicide themselves in less than 10 min almost every match and do less than 20k damage on average.  They don't have the mental or physical skill or both to WTR farm successfully.

  • Bad 1

Share this post


Link to post
Share on other sites
272
[PT8TO]
Members
1,410 posts
13,256 battles
40 minutes ago, VGLance said:

Most of the requirements necessary to achieve a high WTR require skill.  You even contradicted yourself:

 

"Want to end up on there? Play destroyers well."

 

Yes you can manipulate WTR, but all that does is raise the standard of what a "good" WTR is.  That is separate from skill.  If you had Team A where all players had a WTR of 1300 or above play against Team B where all players had a WTR of 600 or below, team A is going to slaughter Team B every time.  It takes an incredibly irrational person to dispute that and if they do, you know they cannot be reasoned with.

But lance the problem with that is you will never see that scenario of 600 vs all 1300 or better in randoms.  But even games where one side is stacked more than the other the teams with the better WTR players do loose often.  Clan wars saw many teams with all purple WTR players loose to teams with no purple WTR players.  I should no I was on those teams that won.

Share this post


Link to post
Share on other sites
1,672
[OO7]
Members
2,229 posts
9,721 battles
1 minute ago, GUNSTAR_THE_LEGEND said:

But lance the problem with that is you will never see that scenario of 600 vs all 1300 or better in randoms.  But even games where one side is stacked more than the other the teams with the better WTR players do loose often.  Clan wars saw many teams with all purple WTR players loose to teams with no purple WTR players.  I should no I was on those teams that won.

There are no absolutes.  Even in a 5 or 10% chance to win scenario, you're still going to win 1 out of every 20 or 10 matches respectively.  But pointing to outliers to support an argument is not the best way to establish credibility.  It's like saying smoking is harmless by pointing to your 90 year old chain smoking grandma.  Even an 80% win rate player loses 20% of the time.  That doesn't mean their high 80% win rate isn't a reflection of their skill.  Same goes with someone with a very high WTR as compared to others.

Share this post


Link to post
Share on other sites
2,538
[OO7]
Alpha Tester
6,734 posts
3,490 battles

WTR Explained:

For those who don't know WTR is a formula used by Warships.Today to calculate how good a player is based on Win Rate, Average Damage, and Average Base XP. They each have a value, with  damage being the best, and win rate being the worst. The data spits out a number that is color coated to match a range that says if you're terrible, below average, average, good, great, etc.

 

It takes into consideration the Tier, Ship, and Ship Type when doing the overall WTR, but also has one for each ship tyoe and ship. You're ranked according to the server average and and you can see your rank for ships as well.

 

It can all be sorted by a time scale, but two weeks is the best indicator of current performance of yourseld or a ship, but as a warning ships data is going to be inaccurate until 2 weeks after a patch, and will probably average slightly lower every week or month if the ship was significantly changed or freshly introduced.

 

Backing up @Kombat_W0MBAT's@Kombat_W0MBAT's point:

 

The point my friend is trying to make here is that the separation between players in this system is much smaller than we believe, and should be mainly used as a general guide instead of an absolute unless you see extreme differences between two players.

 

For example, last night I wondered into 4 Random Battles solo in my Indianapolis to finish my final 60k's worth of damage to get Vampire.

 

The first 3 matches we won I did around 45k combined damage. Truly awful damage numbers which will crush my WTR for the day and in the Indianapolis.

That being said I warded off several critical bomber strikes in the 1st game, used my radar and fire power to make an enemy DD turn tail and run from our CAP in the 2nd match, and finished the final mission by saving our team from an embarressing loss by killing a DD that had snuck behind our entire fleet on Standard Battlw and threatened to win the game even though we dominated. The best part was, when I asked for assistance so I wouldn't die to the two BBs approaching as well I was laughed at and shamed. But did we win the game? Yes, and I did around 14k damage. Does anyone think that's good? But what happens if I farm damage instead of saving the game for my other 11 team mates? I look like one hell of an amazing loser!

Overall my Indianapolis numbers are pretty good though, because I'm usually not forced to make choices that net us the win at the cost of such poor damage chances. So results over time are very important.

 

The fourth battle I topped the team in a loss. I had over 50k damage, sunk a Neptune with repeated citadels, and racked up 8 defense ribbons on a Missouri before I finally sank as I watched my team incompetently avoid the only cap we could take with ease, and get out muscled by a vastly inferior force as the trio I was with sank and were ignored by the stragglers at the other flank who refused to help.

 

Being top of the team was my best scoreboard finish of the 4 games, and given my higher than average damage and good XP, WTR would say that match was better than most of my 3 losses. Honestly though, being the best loser never meant anything to me.

 

I'd rather do the little things that win, and don't inflate my WTR than farm  damage with more losses and a higher WTR. The difference between those styles can be small, but who would you rather have ob your team? That's really the point of this.

 

You can use the WTR system to self validate yourself via their system, or you can use it's tracking system as a broken down data sheet of where you can improve areas of your game without sacraficing the little areas that get you wins that it doesn't track or value as much.

 

When my damage wasn't climbing any longer I reached out on Discord and the Forums to correct any misconceptions I had on how HE and Fire functioned, as well as different areas I could use AP where I hadn't thought of before. When I was having a regression in survival I went back and looked at my positioning and how I could improve but also learned more about the mini-map and what I could do with it. I've reached out to top players in our clan on certain ships I was struggling with and went to other top players in them on the forum and reddit as well. I looked up YouTube videos and read guides. I looked at Captain builds, and learned how I could adjust a few talents and what I was under or over valuing.

 

And all of that helped my WTR climb, but it was more about improving aspects of my game than it was simply trying to manipulate the system. Learning how to shoot down more aircraft for instance doesn't mean jack for WTR, but shooting them down and panicing them over a powerful ship like the Scharnhorst which has bad AA instead of an Iowa which can handle itself? Game winner.

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×