Jump to content

Leaderboards Top 100 - The revised statistics


Cretinus

Recommended Posts

People have posted the stats for the top 100, but without taking into account the overall frequency of the ACs in the leaderboards. This means that these statistics are not only reflecting whether a class is OP, but also how popular it is. In order to eliminate class popularity from the stats and get only the balance indicator, the figures of the top 100 need to be seen relatively to the overall figures of each class in the leaderboards. That's what the right column of my stats show, normed to sum up to 100:

 

Explanation of columns with numbers:

 

Left column: top 100, absolute numbers

Mid column: overall numbers in the leaderboards

Right column: relative numbers, left col. divided by right col. and multiplied with norm factor

so that right column sums up to 100 again.

 

 

-------------------------------------------------------------------------------------------------

 

Vang/PT..................35....................1004........................43

Shadow/Assa..........29....................1737........................21

Sage/Sorc...............22....................1867........................14

Com/Merc................ 5......................601........................10

Scound/Op................5......................817..........................8

Guard/Jugg...............4....................1311..........................4

Gunsl/Sniper.............0.......................366.........................0

Sent/Mara.................0.......................499.........................0

 

If the game was perfectly balanced, then each class should have a representation of 12-13 in the right column.

As you can see, in particular VG/PT is horribly overrepresented, even much more than the absolute numbers tell.

Shadow is overrepresented, but not as much as the absolute numbers tell.

Sorc, com/merc and scound/op are the closest to the target numbers of 12-13, whereby scound/ops are much too weak.

Guard/jugg is under represented and Gunsl/sniper + sents/maras ..... well, RIP.

Edited by Cretinus
Link to comment
Share on other sites

If the game was perfectly balanced, then each class should have a representation of 12-13 in the right column.

As you can see, in particular VG/PT is horribly overrepresented, even much more than the absolute numbers tell.

Shadow is overrepresented, but not as much as the absolute numbers tell.

Sorc, com/merc and scound/op are the closest to the target numbers of 12-13, whereby scound/ops are much too weak.

Guard/jugg is under represented and Gunsl/sniper + sents/maras ..... well, RIP.

 

Bear in mind that even the relative position in the leaderboards is not by any means an accurate number of relative strength. If there is only the slightest advantage for a single class, people will flock to it more then others, especially the better players who have stables of alts to chose from. This will exaggerate the gap between them to look more than it actually is.

Link to comment
Share on other sites

This also doesn't take into account disciplines. Vanguards are the best tank in the game, so both dps and tank are very good. Doesn't mean AP needs more nerfs than what is on its way.

 

p.s. nerf operatives

Edited by lpsmash
Link to comment
Share on other sites

I'm not sure i follow your normalization though... you cant take 5% and mercs and normalize it to 10% while taking others and cutting them in half... but meh, it doenst matter the message is clear either way, i get what you were going for.

 

Bottom line pending which column you use is that 75-85% of those are in the same 3 classes. These really are kinda staggering numbers, i dont know that ive ever seen pvp in any game so completely controlled by so few classes.

 

Further you can extrapolate from this that the gap between healer/tank options is pretty large too as i know most of the Mercs/Ops ive seen are healing more than DPSing but there was some device so Sorc healers are FAR preferred over other options and sins/PTs seem to leave Juggs lacking.

Link to comment
Share on other sites

People have posted the stats for the top 100, but without taking into account the overall frequency of the ACs in the leaderboards. This means that these statistics are not only reflecting whether a class is OP, but also how popular it is. In order to eliminate class popularity from the stats and get only the balance indicator, the figures of the top 100 need to be seen relatively to the overall figures of each class in the leaderboards. That's what the right column of my stats show, normed to sum up to 100:

 

Explanation of columns with numbers:

 

Left column: top 100, absolute numbers

Mid column: overall numbers in the leaderboards

Right column: relative numbers, left col. divided by right col. and multiplied with norm factor

so that right column sums up to 100 again.

 

 

-------------------------------------------------------------------------------------------------

 

Vang/PT..................35....................1004........................43

Shadow/Assa..........29....................1737........................21

Sage/Sorc...............22....................1867........................14

Com/Merc................ 5......................601........................10

Scound/Op................5......................817..........................8

Guard/Jugg...............4....................1311..........................4

Gunsl/Sniper.............0.......................366.........................0

Sent/Mara.................0.......................499.........................0

 

If the game was perfectly balanced, then each class should have a representation of 12-13 in the right column.

As you can see, in particular VG/PT is horribly overrepresented, even much more than the absolute numbers tell.

Shadow is overrepresented, but not as much as the absolute numbers tell.

Sorc, com/merc and scound/op are the closest to the target numbers of 12-13, whereby scound/ops are much too weak.

Guard/jugg is under represented and Gunsl/sniper + sents/maras ..... well, RIP.

 

Yes but that has a lot to do with the FOTM and their shoulder cannon... What will happen after that Nerf?

Link to comment
Share on other sites

Vang/PT..................35....................1004........................43

Shadow/Assa..........29....................1737........................21

Sage/Sorc...............22....................1867........................14

Com/Merc................ 5......................601........................10

Scound/Op................5......................817..........................8

Guard/Jugg...............4....................1311..........................4

Gunsl/Sniper.............0.......................366.........................0

Sent/Mara.................0.......................499.........................0

 

If the game was perfectly balanced, then each class should have a representation of 12-13 in the right column.

As you can see, in particular VG/PT is horribly overrepresented, even much more than the absolute numbers tell.

Shadow is overrepresented, but not as much as the absolute numbers tell.

Sorc, com/merc and scound/op are the closest to the target numbers of 12-13, whereby scound/ops are much too weak.

Guard/jugg is under represented and Gunsl/sniper + sents/maras ..... well, RIP.

 

Do you want us to draw conclusions with this?

like.... VG/PT probably Qsynch more than any other advanced class? :rak_02:

 

 

If you are to talk about class balance, you should on the contrary EXCLUDE the top 100 (which is mostly biaised), and consider the numbers between rank 100 and rank 1000 (top 20%).

And even then, my question is ''what is balanced to you?''

Half of the population is consular/inquisitor (they are over-represented according to your sample). in a rock/paper/scissor game, a class strong against consular/inquisitor, but weak against smuggler/agent --in a word :balanced--, will be overperforming just because they will be mainly fighting inquisitors.

Link to comment
Share on other sites

I'm not sure i follow your normalization though... you cant take 5% and mercs and normalize it to 10% while taking others and cutting them in half...

 

No, that's not how it works. First you calculate the relationship between absolute numbers in the top 100 and overall numbers in the leaderboards for each class. This is where the 5% become 10%. Simply because mercs absolute 5% in the top 100 is much too high compard to their occurrence in the overall leaderboards which is lower then 5%. Assuming that every class had the same overall occurrence of 100% / 8 = 12,5%, then mercs relative occurrence in the top 100 would translate to 10%.

 

Or let me give an example (without me saying here which class is OP or not, it's only an example):

Assume that there are 50 juggs in the overall leaderboards and 30 are in the top 100, while there are only 5 maras in the overall leaderboards and 3 of them are in the top 100.

If you take the absolute numbers like people have been doing here, then you have 30 juggs and 3 maras in the top 100 and juggs look OP compared to maras. But that's not the truth. Because the truth is that 30/50 juggs is the same relationship as 3/5 maras. The fact that there are 10 times more juggs in the top 100 then maras in this example is only reflecting that there are 10 times more juggs then maras overall.

 

The normalization is a simple factor which I then calculated so that if I multiply every single class value with it, then the overall sum becomes 100. It's a purely cosmetic factor.

Edited by Cretinus
Link to comment
Share on other sites

This also doesn't take into account disciplines. Vanguards are the best tank in the game, so both dps and tank are very good. Doesn't mean AP needs more nerfs than what is on its way.

 

p.s. nerf operatives

 

This.

 

Also a larger pool of classes = More on the leader board

 

There could be 300 Vanguards active in Rank PVP whenre there are only 40 Sents.

Link to comment
Share on other sites

No, that's not how it works. First you calculate the relationship between absolute numbers in the top 100 and overall numbers in the leaderboards for each class. This is where the 5% become 10%. Simply because mercs absolute 5% in the top 100 is much too high compard to their occurrence in the overall leaderboards which is lower then 5%. Assuming that every class had the same overall occurrence of 100% / 8 = 12,5%, then mercs relative occurrence would translate to 10%.

The normalization is a simple factor which I then calculated so that if I multiply every single class value with it, then the overall sum becomes 100. It's a purely cosmetic factor.

 

I understand your intent and method, i simply dont agree with it. Doesnt matter really though we are splitting hairs, i dont disagree with your points as much as i do the suggestion that some are more or less represented than they actually are leading some to make invalid conclusions from the data.

 

Id like to see some discussion on what some are lesser played though. We all know why Mercs are where they are, but i find snipers perplexing. Are they not played due to weakness or just because others offer more? To me they are the most curious case... i do very well on mine, better than my Merc to be truthful... is it skipped due to not being good enough or just because more can be done with a Sorc which further feeds the FOTM concerns if one class becomes preferred over those of equal ability due to secondary skills.

Link to comment
Share on other sites

Do you want us to draw conclusions with this?

like.... VG/PT probably Qsynch more than any other advanced class? :rak_02:

 

 

If you are to talk about class balance, you should on the contrary EXCLUDE the top 100 (which is mostly biaised), and consider the numbers between rank 100 and rank 1000 (top 20%).

And even then, my question is ''what is balanced to you?''

Half of the population is consular/inquisitor (they are over-represented according to your sample). in a rock/paper/scissor game, a class strong against consular/inquisitor, but weak against smuggler/agent --in a word :balanced--, will be overperforming just because they will be mainly fighting inquisitors.

 

honestly, you do need to include the top 100 because they are part of the statistics.

 

The essence of the top 1000, though less than the top100 in terms of percentages, it still highly, highly favors the big 3

 

Sins - 262

PT - 230

Sorc - 232

jugg - 110

oper - 70

snip - 45

merc - 39

maur - 26

 

This with a total yolo population of 7807

 

in terms of class representation

 

sins - 1800 - 23%

PT - 1029 - 13.2%

sorc - 1918 - 24.5%

jugg - 1345 - 17.2%

oper - 855 - 10.9%

snip - 376 - 4.8%

merc - 625 - 8%

maur - 517 - 6.6%

 

Thus the big 3 does have a larger population in all of yolo, roughly 60%. But are 75% of the ranking in the 100 and a larger percentage the higher up you go.

Edited by Anyakaschala
Link to comment
Share on other sites

People have posted the stats for the top 100, but without taking into account the overall frequency of the ACs in the leaderboards. This means that these statistics are not only reflecting whether a class is OP, but also how popular it is. In order to eliminate class popularity from the stats and get only the balance indicator, the figures of the top 100 need to be seen relatively to the overall figures of each class in the leaderboards. That's what the right column of my stats show, normed to sum up to 100:

 

Explanation of columns with numbers:

 

Left column: top 100, absolute numbers

Mid column: overall numbers in the leaderboards

Right column: relative numbers, left col. divided by right col. and multiplied with norm factor

so that right column sums up to 100 again.

 

 

-------------------------------------------------------------------------------------------------

 

Vang/PT..................35....................1004........................43

Shadow/Assa..........29....................1737........................21

Sage/Sorc...............22....................1867........................14

Com/Merc................ 5......................601........................10

Scound/Op................5......................817..........................8

Guard/Jugg...............4....................1311..........................4

Gunsl/Sniper.............0.......................366.........................0

Sent/Mara.................0.......................499.........................0

 

If the game was perfectly balanced, then each class should have a representation of 12-13 in the right column.

As you can see, in particular VG/PT is horribly overrepresented, even much more than the absolute numbers tell.

Shadow is overrepresented, but not as much as the absolute numbers tell.

Sorc, com/merc and scound/op are the closest to the target numbers of 12-13, whereby scound/ops are much too weak.

Guard/jugg is under represented and Gunsl/sniper + sents/maras ..... well, RIP.

 

I think you also have to consider the fact that AP/tactics is retardedly easy to play with some degree of competence while NO sin class is nearly as easy to play. so you have a lot more fodder in that large sin sample (aka: their right column is actually lower than it really plays, and the PT/VG is higher than it actually plays). mind you, this has been the case with these two classes since launch. so it's nothing new. everyone knows PTs are simple to do well with and sins require more finesse than most classes.

Link to comment
Share on other sites

Overall, I think these stats effectively show what the community believes in terms of what classes are viable and overpowered/underpowered. I don't know if we can determine whether that is actually true or not from any of the stats we can grab from the Leaderboards.

 

However, the players certainly favor the classes in this way. Frankly, I'm very surprised that Mara/Sents aren't represented more than they are. I definitely thought there would be more of those, especially given the frequency with which I saw them in the Preseason (actually saw many of them).

 

 

Again shows people are running their PT/VGs ahead of 3.1.1 and Jugg/Guard is in mothballs to favour the FOTMs tht are in danger of a nerf.

 

Agreed. I'm betting the end of season tally will look considerably different from Week One of S4, especially after the nerfs/buffs hit. I expect Mara/Sents to be much better represented by end of season. And yes, I expect we will see even more Sage/Sorcs.

 

 

in a rock/paper/scissor game, a class strong against consular/inquisitor, but weak against smuggler/agent --in a word :balanced--, will be overperforming just because they will be mainly fighting inquisitors.

 

This is an outstanding point. With all the Sorc/Sages rolling, it only makes sense that VG/PTs would do well since they are one of the most effective Sorc/Sage busters.

Edited by DarthOvertone
Link to comment
Share on other sites

The essence of the top 1000, though less than the top100 in terms of percentages, it still highly, highly favors the big 3

 

It would be 6 if you had included the Republic side as well, but you chose to ignore them - this way no-one learns how much faction inbalanced the game really is.

 

However, the players certainly favor the classes in this way. Frankly, I'm very surprised that Mara/Sents aren't represented more than they are. I definitely thought there would be more of those, especially given the frequency with which I saw them in the Preseason (actually saw many of them).

 

You seem to be a scrub regarding Psychology : So few Marauders/Sentinels just means that word of mouth spreads incredibly fast - and the word of mouth was : "Marauders/Sentinels are garbage in PvP".

 

You really didn't notice all of these complaints regarding Marauders being so weak in PvP ? Are you biased ? Or are you merely blind ?

Edited by AlrikFassbauer
Link to comment
Share on other sites

Everyone is aware that this is only day 5 of a 90 day season, right? There's still 84 days left for people to queue up and get rating, many of the top players haven't even spent that much time queuing yet and there's a subset of players who have yet to finish their 10 matches and won't for another month or so. It hasn't even been a full week yet but people are speculating this and that. Lmao.
Link to comment
Share on other sites

Snipers rely on CC and snare/roots as their defense. With BW handing out CC immunities and snare/root immunities out to every DPS spec. Snipers stand no chance in arenas.

 

^This.

 

It's also extremely rough for DPS operatives for the exact same reason. Their DCD's are meant to roll and evade. You can't evade **** when you're being rooted and rubberbanded by a bunch of retard jugs spamming ravage root.

 

I've said it before and i'll say it again... jugs NEVER ever needed a ravage *********** root. Sins never needed 17 seconds of stun immunity. Powertechs never needed 10 seconds of zomg rotfl sonic the hedgehog, etc, etc

Edited by JackNader
Link to comment
Share on other sites

To be fair, there are 13 competitive specs right now:

 

AP and tank PT

Hatred and darkness sin

All sorc specs

Veng Jugg

MM sniper

Healing ops

 

That is 13 out of 24. And some marauder specs, ops dps and other jugg specs are not so bad if you have a competent team. Problem is these can not compete with the above. Merc did not get anything so far aside from 1 sec snare, which is useless. Marauders got some love but does not solve the core problems the AC has, which is staying on target. Ops dps has to choose between dps and survivability, which does not work in competitive environment.

 

The following spec should under no condition be played in a WZ not to mention ranked:

 

Pyro PT

IO merc

 

Sadly, 2.10 was way more balanced than now..

Link to comment
Share on other sites

Snipers rely on CC and snare/roots as their defense. With BW handing out CC immunities and snare/root immunities out to every DPS spec. Snipers stand no chance in arenas.

 

This isnt 100% true.. snipers have a variety of shields and others things aside from that to assist under direct pressure such as cover, evasion, and shield probe... in addition to the AoE shield, and diversion at range. It isnt just about roots and or CC.

Link to comment
Share on other sites

Made a chart that kind of simplifies some things.

 

Top 50 Comparison of rank by each class

 

In it I took the ratings of the top 50 of each class.

 

In it you will find that the top 5 to 7 players of each class perform outside any trend that the rest of the group follows. I imagine rating inflation do to favorable circumstances, or some edge has been found for these few players, but none the less the top 5 to 7 players of each class should be ignored when making claims of classes being OP, or UP. Im not accusing anybody of foul play, but just looking at the numbers these players are not part of the trend.

 

They seem to perform outside of the capabilities of other players, and their numbers should not reflect on the capabilities of the class as a whole.

 

 

From the information I have taken this.

 

Their are clear indicators that Sins and PTs are clearly above the rest, followed closely by sorcs, in terms of success. PTs, and Sins perform so closely in terms of rating number that their lines of trend are practically on top on each other. Sorcs are a bit lower, but in terms of rating they perform just under PTs and Sins.

 

Juggs and ops fall into the middle area of the chart, juggs actually seem to fair a bit better as we go down the list. As Ops fall to number 5 when comparing the 40 to 50th of Ops to the 40 to 50th of Jugs

 

Mercs, Snipers, and Marauders make up the bottom. mercs actually fall down to rank 7 in the 50th merc vs the 50th of everything else. This actually came as a surprise considering the number of high performing mercs, and their presence in the top 100.

 

Overall though my window is too small. I would need to have the top 300 of each class in this chart before I could expect to have a serious trend. Cutting it off at the top 50 might be as bad as cutting off at the top 7 of each class. As I dont plan on spending all day on this chart I am only going to give you what I have so far, as its a snow day and im bored as hell right now.

Edited by Haystak
Link to comment
Share on other sites

Mercs, Snipers, and Marauders make up the bottom. mercs actually fall down to rank 7 in the 50th merc vs the 50th of everything else. This actually came as a surprise considering the number of high performing mercs, and their presence in the top 100.

.

 

I have a thought regarding the Merc surprise, i think that thing to be taken from that is heals versus dps. i suspect and know that many of the better performing mercs are healers, not dps.

Thats not to say that some dps can do well and sa lot of that is also random. Some people i know have been far luckier than i. I tend to get the other team with 3 Sins/PTs and im teamed with the Maras and other Mercs, its actually been someone unreal the team comps ive seen in my matches but its random so it is what it is. I can only hope it balances out at some point, lol.

Edited by Floplag
Link to comment
Share on other sites

×
×
  • Create New...