Android25

Setting PR to a scale

106 posts in this topic

I think it'd be a lot easier for everyone if you just accepted that what you do will never be generally appreciated and move on.

I'd be fine with it if people didn't use that as a straight-up dismissal of any argument I try to make, such as why things like PR are mathematically faulty.

Share this post


Link to post
Share on other sites

I know expressing your hate-on for what I do is a lot easier than actually trying to defend PR having any value to anyone.

Or how about the fact you're being stuck up. You claim it takes a fuck ton of skill to go down tiers and play shitters. You call high tier gameplay garbage despite not even reaching tier 10 because it's "longer and more pokey" All you care about is the metrics that make you look good. I'm sure if WN8 had a super penalty on people below tier 6 you'd be bitching about how WN7 was a better metric. Personal rating is recognized by everyone as a valid metric, maybe not the metirc to all metrics, but it is better than efficiency and WN7, It also seems to be the most accurate when comparing WR to it as well.

I'd be fine with it if people didn't use that as a straight-up dismissal of any argument I try to make about why things like PR are mathematically faulty.

SHOW US THE FAULTS THEN. You have yet to bring out a formula of it showing where it is wrong.

Share this post


Link to post
Share on other sites
 

I'd be fine with it if people didn't use that as a straight-up dismissal of any argument I try to make about why things like PR are mathematically faulty.

While I agree with almost every argument against it, the one thing it does provide is a metric based on data that we can't use historically (spotting/assisted damage), and technically a more accurate overall scale, since there's no way we can filter data based on nerfs/buffs to tanks in the past, where WG seems to do that. Another advantage is that it does scale will with other ratings from most comparison charts I've seen, and you can't really pad it because you don't know what the formula is. We know it takes total games into account, which is mostly a bad thing, because number of games doesn't really mean experience, considering you can play 10k games with horrible habits and never really learn anything. But it can also be a good thing it that you can't reroll an account to pad PR, at least not for the first 3k ish battles.

The other arguments for/against it are in the current WN9 topic on in the same board as this topic.

 Ignore everyone else.

Share this post


Link to post
Share on other sites

I'd be fine with it if people didn't use that as a straight-up dismissal of any argument I try to make about why things like PR are mathematically faulty.

I've literally spent the past 5 or so days attempting to write a program to pull data, and then after I got beat at that (not to sound unhappy, I'm glad it didn't take as long as I had planned to get the data), I've spent my entire day calculating server percentiles that I know full well aren't very useful to any sort of stat or metric creation. I've done it simply because it's interesting to post and look at. I find it interesting, other people seem find it interesting, and so I spend my time on it because I enjoy it... not necessarily because it is or isn't mathematically faulty.

I spent 2 or 3 hours crunching the numbers around PR simply because I thought it would be interesting to give it a scale, since I've never seen that done before. Well, here it is, and you seem more interested in telling me why it shouldn't be used (which I 1. don't care either way, and 2. don't have any control over).

Share this post


Link to post
Share on other sites

I've literally spent the past 5 or so days attempting to write a program to pull data, and then after I got beat at that (not to sound unhappy, I'm glad it didn't take as long as I had planned to get the data), I've spent my entire day calculating server percentiles that I know full well aren't very useful to any sort of stat or metric creation. I've done it simply because it's interesting to post and look at. I find it interesting, other people seem find it interesting, and so I spend my time on it because I enjoy it... not necessarily because it is or isn't mathematically faulty.

I spent 2 or 3 hours crunching the numbers around PR simply because I thought it would be interesting to give it a scale, since I've never seen that done before. Well, here it is, and you seem more interested in telling me why it shouldn't be used (which I 1. don't care either way, and 2. don't have any control over).

I'm sorry that you've wasted your time analyzing a metric that any theorist could have looked at way back when it was introduced and picked apart why it isn't worth looking at.  I hope other people in the mathematics corner inspire you to spend your time with data that has more practical use to the community.  I wish I could do that, but you're too sucked up by Zeph's bullshit for me to have any influence on you at this point.

Just going to take Cuni's advice and quit the rest of the thread.  Maybe I'll go re-enlist in PZMOE so that the next time somebody tries to start up shit about me in an unrelated thread, I can just take it easy and ignore them.

Share this post


Link to post
Share on other sites

I'm sorry that you've wasted your time analyzing a metric that any theorist could have looked at way back when it was introduced and picked apart why it isn't worth looking at.  I hope other people in the mathematics corner inspire you to spend your time with data that has more practical use to the community.  I wish I could do that, but you're too sucked up by Zeph's bullshit for me to have any influence on you at this point.

Just going to take Cuni's advice and quit the rest of the thread.  Maybe I'll go re-enlist in PZMOE so that the next time somebody tries to start up shit about me in an unrelated thread, I can just take it easy and ignore them.

My Bullshit? You little autist, you seriously keep bashing PR like it's fucking Hitler. He wanted to make a scale because he was bored. If PR is so bad, why does BULBA use it in their minimum requirements along with the other metrics? I honestly would be threatening to TK you in the next pub match i saw you in BUT OH WAIT YOU DON'T PLAY RELEVANT TIERS MEANING I'LL NEVER SEE YOU IN GAME. xD

Share this post


Link to post
Share on other sites

I hope other people in the mathematics corner inspire you to spend your time with data that has more practical use to the community. 

Hey Kuro, if you don't like something in a thread and don't have anything useful to add or say, then sincerely shut the fuck up. Go make a thread in the venting place if you want to hate on something that he just wanted to make a scale with for the heck of it. 

I hope some common sense inspires you to have more practical use to the community other than sperging in a thread. 

Share this post


Link to post
Share on other sites

Hey Kuro, if you don't like something in a thread and don't have anything useful to add or say, then sincerely shut the fuck up. Go make a thread in the venting place if you want to hate on something that he just wanted to make a scale with for the heck of it. 

I hope some common sense inspires you to have more practical use to the community other than sperging in a thread. 

Dude he's played 14k battles at an avg tier of 3. I don't think there is any common sense to appeal to.

Share this post


Link to post
Share on other sites

Consider me shouting it up. PR is more than enough for overall measures. If we dump development for over-all, we can ignore so much nonsense, and ultimately it doesn't matter right now how anyone played 1+ years ago. Leave PR and overalls to the people who have all the data and paid a consultant good money to implement it.  Focus on recents and recents only with a good formula. The only reason overalls were the basis for WN4-8 with Preator was because we wanted to get it into XVM to replace efficiencyv1 (which was innovative garbage) and XVM didn't have access to recent stats. XVM has switched to PR by default which is fine. Let them have PR/WN7/WN8/EFFv2 as settings and just focus on making something relevant and free of the noise of the past.

Most of my reasoning for the thread.

Share this post


Link to post
Share on other sites

Dude he's played 14k battles at an avg tier of 3. I don't think there is any common sense to appeal to.

Liking low tiers doesn't mean anything but even if it wasn't 3 his sperg is still enough to be called out. 

Share this post


Link to post
Share on other sites

The upper value for WN8 is 9999, anything between 2900 and 9999 is super uni. Kinda took a shot in the dark as to the highest personal rating and went beyond. 

I forgot it worked that way; sorry, yeah, you've got it fine :)

Share this post


Link to post
Share on other sites

I'd be fine with it if people didn't use that as a straight-up dismissal of any argument I try to make about why things like PR are mathematically faulty.

You haven't made any mathematical arguments. If you don't bring up mathematical arguments in a mathematics discussion of course anything you say is going to be dismissed. You've spent most of time just saying it is a terrible rating.

Share this post


Link to post
Share on other sites

So along this scale in around 200 more games or so, I will be a blunicum- pretty my exactly where my Wn8 is. I find that pretty interesting actually.

 

Share this post


Link to post
Share on other sites

You haven't made any mathematical arguments. If you don't bring up mathematical arguments in a mathematics discussion of course anything you say is going to be dismissed. You've spent most of time just saying it is a terrible rating.

I'll do the mathematical (or at least engineering) argument, because it's mildly interesting. WG-PR doesn't use an explicit tier adjustment like WN7: Instead, it attempts to trade off winrate vs damage such that a typical player would get a similar total regardless of which tier they played. The component formulas look like this:

Zklk1PZ.png

You'll have noticed that they're completely different shapes. Essentially WG-PR gives you very little reward for increasing your winrate beyond 60%. Because the damage components are tiny at tier 2 regardless of your performance, competent sealclubbing is under-rewarded compared to incompetent sealclubbing. This is probably intentional.

Share this post


Link to post
Share on other sites

RN's pro con list is basically my thinking for proposing to do no more development of overall account metrics. They just aren't useful anymore and PR is no worse. Especially like it with battle count flattened out.

Share this post


Link to post
Share on other sites

If PR scales well for overall... What the fuck is the problem? I like that it punishes low-tier play and re-rolling.

Fuck it. Scale PR to a GPA-type grading system and give us WNx recent with letter grade overalls. I think I'm a solid B.

Share this post


Link to post
Share on other sites

If PR scales well for overall... What the fuck is the problem? I like that it punishes low-tier play and re-rolling.

Fuck it. Scale PR to a GPA-type grading system and give us WNx recent with letter grade overalls. I think I'm a solid B.

"GPA" triggers me. pls no

Share this post


Link to post
Share on other sites

wow

PR hype...~ This is better than WN8, cuz NA server is broken server somehow good player at 2300 wn8 till scare against every tier. I could say they were not fully acknowledge about game mechanic such as Gun stats/Tank stats/controlling map.

 

Share this post


Link to post
Share on other sites

iirc the exact formula for PR was posted somewhere some time ago, can someone dig that out, I can't seem to find it?

Share this post


Link to post
Share on other sites

iirc the exact formula for PR was posted somewhere some time ago, can someone dig that out, I can't seem to find it?

http://s3.postimg.org/prljiad41/formula.png (I can't get this to show up for some reason)

win – winrate (from 0 to 1)
surv – survival rate (from 0 to 1)
dmg – average damage per battle
bc – total amount of battles played
bc8.8 – amount of battles played from the point of patch 8.8 release
xp8.8 – average XP per battle (without premium account, from the point of patch 8.8 release)
radio8.8 – average damage done after your spotting (from the point of patch 8.8 release)
track8.8 – average damage done after you detracking the opponent (from the point of patch 8.8 release)

 

I didn't realize this had been released... cool

Share this post


Link to post
Share on other sites

Ive enabled PR for my current config of XVM and used the scale from the colors.xc I created on the previous page. 

Its reasonably accurate for in-game use, IMO. Theres a LOT more "average" players in the game which is to be expected, with fewer blue/teals/unicums as I also expected. 

Share this post


Link to post
Share on other sites

I'm sorry that you've wasted your time analyzing a metric that any theorist could have looked at way back when it was introduced and picked apart why it isn't worth looking at.  I hope other people in the mathematics corner inspire you to spend your time with data that has more practical use to the community.  I wish I could do that, but you're too sucked up by Zeph's bullshit for me to have any influence on you at this point.

Just going to take Cuni's advice and quit the rest of the thread.  Maybe I'll go re-enlist in PZMOE so that the next time somebody tries to start up shit about me in an unrelated thread, I can just take it easy and ignore them.

Quite funny that you get to decide whether somebody wasted his time having fun with w/e and even did create smth others may benefit from in the future. All that for the expense of exactly nothing, but his free time.
That way to argue or rather shitpost reminds me of smby I've read about earlier, dunno if 
M1ster_R0gers rings a bell for you.

Share this post


Link to post
Share on other sites

I am all up for using PR over anything else. I found it a pretty good metric after the second iteration when they tweaked it and PR wasnt boosted like crazy based on battle count. 

But the best part is it takes into account stat padding, seal clubbing, includes average spotting dmg and tracking dmg and takes into account win rate.

I have used it in xvm before and its reliable enough. I might even switch back.

Share this post


Link to post
Share on other sites

Mind explaining why PR unicum @ NA isn't at ~9600?

It's placed to match the WN8 unicum mark by population percentage. The current WN8 unicum mark is nowhere near 99.9% for NA players with 1k+ battles: The previous mark was derived from something like RU players with 2k+ battles and an arbitrary filter at 146 WN8.

This adjustment hasn't been applied to the other boundaries, so the chart is internally inconsistent. Hence the blunicum range is only 500 points.

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.