SereneGrace

Rank Distributions (23rd Feb)

Recommended Posts

Data collected at 12:30 am. Data collected from top 950 ranks from each class at ranks 1, 25, 50, 75, 100, ..., 950. I'll probably do this again near the end of the season.

Interpret the data carefully!

 

pUAlwM0.png

 

If you want to take a closer look for each class' distribution, I've uploaded them individually here http://imgur.com/a/T5unI/all

Edited by SereneGrace

Share this post


Link to post
Share on other sites
elliv   
2 hours ago, SereneGrace said:

Interpret the data carefully!

  1. Summoner is broken
  2. BM need buff

Share this post


Link to post
Share on other sites
Jamein   

So if we ignore anything above 1950 as that data starts to become too sporadic you can see summoners ahead at literally every single data point.

I've always been one to say while summoners are strong, the biggest strength they have is at lower ELO with people not understanding the gameplay and that the class itsself isn't that complicated (which you can sort of see as the gap at the higher end is a lot smaller than at lower end.). 

 

However you can clearly see even at higher ELO, summoner is ahead. I refuse to believe that just every summoner player is a better player than every other player that plays a different class, so take from that what you will.

 

Pretty nice way to evaluate the data by the way, simply taking the 'how many summoners are in the top 500 doesn't show much, but this is actually some meaningful data based on class rankings and the overall ELO chart for the class.

 

Kudos.

Edited by Jamein

Share this post


Link to post
Share on other sites

It's a nice graph, but you can't interpret anything from it other than that class doesn't mean much after a certain point (although it still does because rock-paper-scissors). We lack participation numbers, skill book completion, time of play, ping, class walls, etc. 

Edited by OddFinrir

Share this post


Link to post
Share on other sites
Numie   

lately anywhere i look i see a summoner this is just sad

Edited by Numie

Share this post


Link to post
Share on other sites
Duniak   

Lol, worst chart I've ever seen. Doesn't show anything. Rank/MMR chart? MMR IS RANK. Interpret my ass. Other than showing that rank 950 Scummoners have better MMR it shows nothing. Numbers count. Instead of wasting your time for every class make number chart with MMR for all classes.

Share this post


Link to post
Share on other sites
TakunDes   

To those who thinks the game is balanced and say stuff like learn your class. please don't post if your a KFM/ Des since ofc you will be fine since they are one of the counters to a Sum if they play well. try to play a BM against a summoner you'll see how unbalanced it is.

 

 

Share this post


Link to post
Share on other sites
Enoll   

I take it this is only for NA ? And btw where do you get the numbers from I am very curious

Share this post


Link to post
Share on other sites
9 hours ago, Enoll said:

I take it this is only for NA ? And btw where do you get the numbers from I am very curious

Yes this is NA, you can see ranks and stuff in-game with F11.

I wish we can access the ladder outside the game as well.

Share this post


Link to post
Share on other sites
Gamemako   

Can we get the data directly? I'd be interested to see 1900+ and 2000+. I would also like to fiddle a bit and test a few hypotheses. For example, a steeper MMR drop-off might be expected for classes that are easy-to-play but don't grow in power as much with skill. Of course, this data may be well beyond that point by top 1000 of a class.

 

9 hours ago, Duniak said:

Lol, worst chart I've ever seen. Doesn't show anything. Rank/MMR chart? MMR IS RANK. Interpret my ass. Other than showing that rank 950 Scummoners have better MMR it shows nothing. Numbers count. Instead of wasting your time for every class make number chart with MMR for all classes.

 

No, rank is your MMR relative to other players of your class. Summoners ranked in the top 1000 summoner players have much higher MMR than BMs ranked in the top 1000 of their class. This can happen two ways: Summoners are vastly more played, resulting in tighter competition for top 1000, or Summoners are just plain doing better.

Edited by Gamemako

Share this post


Link to post
Share on other sites
Kutsuu   

More than anything, this chart shows what the NA player base has chosen to play. It says a lot for us that so many players have rolled or rerolled summoner because of their perceived strength. 

Share this post


Link to post
Share on other sites
Gamemako   
55 minutes ago, Kutsuu said:

More than anything, this chart shows what the NA player base has chosen to play. It says a lot for us that so many players have rolled or rerolled summoner because of their perceived strength. 

 

We can test that theory fairly easily. Just view the chart upward from top 50 Blade Master. That is somewhere in the 1800-1825 range. From that spot on the chart, we can read the approximate number of players who have reached that MMR level for each class.

 

50 Blade Masters

~135 Force Masters

~185 Assassins

~245 Kung Fu Masters

~295 Destroyers

~340 Blade Dancers

~670 Summoners

 

Do you believe that there are 13.4 times as many Summoners as Blade Masters? Do you believe Lyn-exclusive classes are a combined 20 times more common than Blade Masters? Do Lyn-exclusive classes outnumber non-Lyn classes by 30%? If these are not the case -- and I do not believe they are -- then the popularity explanation alone is insufficient to explain the effect.

 

//EDIT: It also appears that the population gap grows with rank. Where there are 375 BMs, there are approximately 950 KFMs, 875 Assassins, and 775 Force Masters. These are population rations of 2.067x for FM, 2.33x for Assassin, and 2.53x somewhere in the range of 1720 or so MMR. By the time you advance by 100 MMR, the population ratios grow to 2.7x for FM, 3.7x for Assassin, and 4.9x for KFM. This again does not support the hypothesis that population alone can explain the class participation gap.

Edited by Gamemako

Share this post


Link to post
Share on other sites
Kutsuu   
12 minutes ago, Gamemako said:

 

We can test that theory fairly easily. Just view the chart upward from top 50 Blade Master. That is somewhere in the 1800-1825 range. From that spot on the chart, we can read the approximate number of players who have reached that MMR level for each class.

 

50 Blade Masters

~135 Force Masters

~185 Assassins

~245 Kung Fu Masters

~295 Destroyers

~340 Blade Dancers

~670 Summoners

 

Do you believe that there are 13.4 times as many Summoners as Blade Masters? Do you believe Lyn-exclusive classes are a combined 20 times more common than Blade Masters? Do Lyn-exclusive classes outnumber non-Lyn classes by 30%? If these are not the case -- and I do not believe they are -- then the popularity explanation alone is insufficient to explain the effect.

 

Yep. I sure do believe that. Many of the PVP players who initially rolled BMs have already rerolled Summoners or other classes. I did not say that population ALONE is the only reason for summoners being more successful, especially in the sub 1950 range. It's an easy class to become relatively proficient with, and noob vs noob is probably by far the strongest. 

Share this post


Link to post
Share on other sites

Id say it's hard to interpret destroyer since it's probably lower than it should be due to a billion bots playing the class. I think without the bots it would be equal or greater than blade dancer on this list.

Share this post


Link to post
Share on other sites
Gamemako   
21 minutes ago, Kutsuu said:

 

Yep. I sure do believe that. Many of the PVP players who initially rolled BMs have already rerolled Summoners or other classes. I did not say that population ALONE is the only reason for summoners being more successful, especially in the sub 1950 range. It's an easy class to become relatively proficient with, and noob vs noob is probably by far the strongest. 

 

The game is not designed for 1900 to be "noob vs. noob". There are fewer than 50 BMs who have even reached that point at all. If you're in the top 1000 of any class, you're probably not a novice player. It takes at least 14 matches at 100% win rate just to reach 1700, and the only class with fewer than 1000 players in that range is BM. There are nearly 1000 summoners above 1800. This chart doesn't describe novice play at all. If it did, then you would also expect a steeper drop-off for classes which are "easy" -- that is. once skill comes into play, other classes should be closing the gap. Up to Platinum-level play -- decidedly not noob-friendly -- the ratios are maintained quite well. There are still twice as many BDs as assassins, and twice as many Summoners and BDs, and 15% more BDs than Destroyers.

Edited by Gamemako

Share this post


Link to post
Share on other sites
Kutsuu   
4 minutes ago, Gamemako said:

 

The game is not designed for 1900 to be "noob vs. noob". There are fewer than 50 BMs who have even reached that point at all. If you're in the top 1000 of any class, you're probably not a novice player. It takes at least 14 matches at 100% win rate just to reach 1700, and the only class with fewer than 1000 players in that range is BM. There are nearly 1000 summoners above 1800. This chart doesn't describe novice play at all. If it did, then you would also expect a steeper drop-off for classes which are "easy" -- that is. once skill comes into play, other classes should be closing the gap. Up to Platinum-level play -- decidedly not noob-friendly at all -- the ratios are maintained quite well. There are still twice as many BDs as assassins, and twice as many Summoners and BDs, and 15% more BDs than Destroyers.

 

Again, I'm saying there are FAR less BMs trying to succeed in the arena than Summoners. Can you seriously argue otherwise? Can you understand that, not knowing the actual number of BMs competing or the number of summoners competing, that we can't make a direct relation between the number in the top 1000 and balance? 

 

It would be neat if NC would release some statistics on the numbers playing arena. I have a feeling the population numbers would have a pretty close correlation with your chart. 

 

edit: Because I may not have been clear enough in my previous post - I am NOT saying that there is perfect balance or that BMs are equal to summoners at a high level of play. However, I do think balance between proficient players is much closer than population numbers show due to the fact that NA player gravitate to easy classes. 

Edited by Kutsuu

Share this post


Link to post
Share on other sites
Zooks   

Oh look Summoners are ahead of everyone at pretty much every rating. How shocking. Queue people saying it's meaningless data because statistics and math. It doesn't take a degree in statistics to see a very obvious trend. How can you tell us, with a straight face, that one of the explanations is that there's a lot more Summoners playing than every other class? No, the explanation is simple; Summoners are god-tier this patch. Watching a Summoner match vs pretty much anything is so frustrating. To see the amount of work a class has to put in to kill a Summoner versus what the Summoner has to do to kill another class is disgusting. 

Share this post


Link to post
Share on other sites
Gamemako   
2 hours ago, Kutsuu said:

 

Again, I'm saying there are FAR less BMs trying to succeed in the arena than Summoners. Can you seriously argue otherwise? Can you understand that, not knowing the actual number of BMs competing or the number of summoners competing, that we can't make a direct relation between the number in the top 1000 and balance? 

 

It would be neat if NC would release some statistics on the numbers playing arena. I have a feeling the population numbers would have a pretty close correlation with your chart. 

 

edit: Because I may not have been clear enough in my previous post - I am NOT saying that there is perfect balance or that BMs are equal to summoners at a high level of play. However, I do think balance between proficient players is much closer than population numbers show due to the fact that NA player gravitate to easy classes. 

 

I'm not arguing the cause of the discrepancy there. I'm just disputing a given hypothesis for the observed data by applying a simple test. There has to be more to the issue than merely "there are too many summoners" or "summoners are rated highly because they're easy". That's all I'm getting at.

Edited by Gamemako

Share this post


Link to post
Share on other sites
beowolf   

If anything, I find the "summoner threshold" of plat as the most interesting (sort of like the "destroyer bot threshold" is around 1700).  It either means that the game is balanced around plat, or that there is too little evidence of higher ELO play.  Judging from another graph that summoners still have 1.5x the number of diamonds vs the nearest competitor (35 to 21 KFM was it?), I'd guess both.  

 

That means that yeah, summs are bitches and they need tuning, stealth taking away power that pros would notice and lower players would accidentally discover while button mashing - my example suggestions would be to prevent cats from interrupting airbornes (physics don't allow that meow!), or remove the unblockable 1 attack after the 2 evade escape (few lowers would know of that, but they'd stumble on that power while panicking at first CC).  But yeah, the proficiency at plat is much less difference than the witch hunts on the forums, and the likes of "omg cut summ hp in half or delete them" are just gonna get a "fcuk you and have a nice day."

Share this post


Link to post
Share on other sites
Gamemako   
4 hours ago, SereneGrace said:

It's also interesting to note the gap in 1900, showing people stopped playing once they reached platinum.

 

Yup. If you continue to play after 1900, you get to be the butt of a joke if you derank. Kind of a problem for me because I close the client when I face too many assassins so I can have a breather and calm the *cricket* down, so I derank regularly and have to work my way back up to platinum.

 

Yes, I close the client vs. Assassins. They aren't so much as hard to beat as they are absolutely *cricket*ing infuriating rubbish design whose *cricket*ing designer should be forced to spend the rest of his life on a *cricket*ing rock pile lest he ruin anything else he *cricket*ing touches, the stupid *cricket*ing *cricket*.

Share this post


Link to post
Share on other sites
fhoeng   

I was trying to read something into the data and draw interesting fact, but the more I look at it the more I feel this sample size and variables are simply too... lacking to be used to draw any kind of conclusion.

 

Only thing I can say is,  below 1899, class not balanced.

Edited by fhoeng

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now