onGamers News

Analyzing how gold is distributed for each EU LCS team

A look at how each team distributes their gold relatively compared by role and how this illustrates each team's strategy and star players.

With the success of the NA article, I wanted to look at gold distribution for EU LCS this time. The way this was calculated by taking the average gold distribution between roles (20.89% of gold to top laner, 18.35% to Jungler, 22.83% to Mid, 23.76% to AD Carry, and 14.17% to Support), then from here we looked at how each team distributed their gold, and then looked at how each team compared relative to the EU averages. The following represents how each team compares to the average role:

How to interpret these graphs

Inspired by a comment last article by Jatt, I wanted to put a few notes to help with interpreting what exactly these charts mean, as there end up with a bunch of misconceptions.

  1. These values are relative to each role’s average. When you see a team like Fnatic it doesn’t mean that Yellowstar gets more absolute gold than Rekkles. It just means that Yellowstar, relative to other supports, gets more gold than Rekkles gets relative to other AD carries.
  2. More gold =/= Better Performance. This metric is more used to measure style of play, rather than “who is the best”. Just because Diamond gets the most gold in the jungle does not absolutely make him best jungler, there are many other factors to conceptualize performance.
  3. This data was balanced to ignore the “Winning Bias” shown in NA version. Now the averages are weighted in a way that these values are adjusted so that there is no longer “the top teams get more objective gold which makes their graphs more consistent”. For more info on how this is done, see this comment.

How I interpret these graphs

One of the main goals of these graphics is to generate discussion, where you can take the information yourself and form whatever opinions you want based on it. However, I have a few of my own theories after being so intimately involved with the data, so I will go team by team and break it down:

Gambit Gaming:

I think it is no surprise that Diamond takes up the biggest proportion of his teams gold compared to other junglers, he is known for his farm and counter-jungle style. But it is interesting that he has the highest relative positive spike. In a team like Gambit, where every player has its “star quality”, its interesting to see such an extreme graph.

SK Gaming:

SK has one of the most balanced graphs of each team, which was a surprise to me. My feeling from watching SK is that they somewhat rode on freddy’s back, with him split pushing on a carry-style top laner like Shyvana. I expected a spike in top lane’s area, but SK ended up with a very balanced graphic, with exception of Jesiz. Something worth noting, is that this is a tied for first team in EU, the region of mid laners, but their mid laner gets significantly less proportional gold (the 2nd least in EU), which is an interesting contrast compared to the other top 5 teams.

Alliance:

Alliance holds the biggest peak for mid laners, which is of course Froggen. I guess despite the roster changes it has still stayed as the “Froggen is the star” show, but it was interesting that Froggen was so heavily ahead, with Wickd also being above-average.

Fnatic:

The biggest unique factor in Fnatic’s graph is Cyanide’s lack of gold, which we covered previously in the Fnatic article. Cyanide is by far the most selfless jungler, especially when it comes to getting minion/monster kills, he basically stops farming after 25 min. The other wrinkle here is the bot lane of Fnatic, which Rekkles is known for amassing gold, and is the current GPM leader overall, and Yellowstar currently leads in Assists for all players.

Roccat:

As the streak of volatile junglers continues, the next spike is for Roccat’s Jankos, the minion/monster kills leader among junglers. This graph lines up with our expectations, with Jankos and Overpow being the spikes for this team, while Celaver and Xaxus get below-average distribution.

Copenhagen Wolves:

This CW graph lines up a lot with my expectations, as Amazing is clearly the star of the team, but Forg1ven has been consistently putting up high GPM numbers despite losing, and Unlimited who has the 3rd highest kill participation in the league, as well as 2nd most total assists behind Yellowstar.

Super Hot Crew:

SHC has the most balanced graph, with the Mid, ADC, Jungle, and Support almost all hitting average, and top being the only spike. This is actually very rare in EU, as top laners very rarely get the highest relative gold, and Mimer doesn’t even put up exceptional stats. I think it is mostly this way because of how low total gold SHC as a team averages.

Millenium

Mil has the most crazy graph, which somewhat reflects their awkward performance so far this split. Kev1n and Creaton are seen as their "stars", but I think this is a not so good distribution and might be somewhat the cause of their lack of success. This low-farm role of Araneae does not fit with his champion picks, such as Lee Sin, Evelynn, and Vi, which are some of the more farm-intensive junglers.

Data collected with help by the onGamers Stats team: Derek 'Kathix' Adams, Steven 'whedgehead' Falgout, Kent 'Traepoint' Frasure, Jake Morales, and James 'PelkaSupaFresh' Pelkey. Design by Ben 'Sarcasmappreciated' Li.

6 Comments
Posted by realfakebrit

When you have to spend more than half the article explaining what the hell your 'Graphs' even mean, should tell you that they are incredibly poor. A Pie Chart would have made this very easy to read, and if you just have to make them 'look cool' make them 3D and then slice out the role with the most gold. So easy, actually readable, and you could spend that wasted space on investigating the numbers, not how the graphs work.

RFB

Posted by YoshixLoL

this graph is so easy to read, even a second grader can interpret it. and this graph isn't to make it to "look cool", this graph makes it so much easier to the eye the distribution of skills in an individual person or a team.

Posted by Osilath

@realfakebrit: Damn never met someone so hard for pie charts before

Edited by tjlep

I think that @realfakebrit is missing that these graphs are not meant to show the gold distribution among team members but how the gold distribution varies from each role's respective average (kind of necessitating the explanation @spellsy gave in the article?) A pie chart would not give a visual representation of this. Instead you would get a set of eight pie charts that look mostly identical. This is because the general hierarchy of gold distribution is unlikely to change dramatically in a well established meta. As an example: Despite the wide variation in Fnatic's graph, their support is still making less gold than their jungler. From their LCS stat pages, it can be seen that Cyanide still averages 30 more gpm than YellOwStaR. Going past that, even establishing a visual representation of the average based off of a pie chart scheme sounds like a real mess.

Putting all that aside, this was an enjoyable and insightful read. I think it would be very interesting to further explore the factors that correlate to and effect these changes in distribution.

Posted by sarcasmappreciated

Hey @realfakebrit, unfortunately, it's hard for a pie chart to represent the differences between players and then averages of every team. Anyhow, I think the comments above me addressed it better.

Cheers

Staff
Posted by hebble

Just really stands out to me how extreme almost every graph here is. EU teams must play very differently, which may explain the widely varying results week to week.

Staff