Which LCS players and teams play the most off-meta champions? – NA Edition (in-depth)

    Of the many great joys of competitive League of Legends, the off-meta pick may be the greatest. There’s nothing quite like the crowd gasping when a player locks in a long-forgotten signature champion, or when the casters get excited about a never-picked-before mid laner. What I set to find out is which players and teams have picked off the meta the most since Spring 2014.

    What defines the meta, though? A Gragas top pick would have been a surprise during the 2014 spring split but passé during the late summer split. I found that the meta shifted most when a patch occurred, each of which happened around every two weeks. So, to calculate the “off-meta”ness of a certain champion pick, I used the champion’s pick/ban rate during the patch it was played in.

    Here’s an example for how this works. During Patch 4.9, which was used on Weeks 4 and 5 of the 2014 Summer Split, Jinx was picked or banned in 4 out of a possible 16 games. Thus, if a player picked Jinx during this patch, they would get a score of 25%. All of a player’s scores are averaged to come up with their total score. (The summer split meta was really stale, so I slightly adjusted the numbers so that all splits would have the same average.)

    A few things to note: These numbers come from only regular-season games played during 2014-2015 North American LCS. I’ve added each player’s off-meta picks, and what I identify as trademark picks (champions that a pro has played at least 3 times and that other players rarely pick) will be bolded. GP simply means “Games Played”. The lists are ordered top-to-bottom, from most off-meta to most on-meta. In other words, the people at the top play weird stuff.

    Also, active players are italicized, and keep in mind that players with few games may have extreme scores (5-game minimum to be on listed).

Top

    There are four clear-cut tiers here. The players in the first tier, from Dyrus to Ackerman, are simply the meta top laners, seen as steady performers who play what the team needs. The only real difference in scores from these players comes from how long someone’s comfort picks are in the meta (for instance, Balls’ Rumble was in the meta less than Ackerman’s Gragas). Nothing too exciting here.

    The top laners in the second tier, from Quas to Westrice, are players who play mostly meta, but are well known for playing some cool picks now and then. It is surprising that Seraph ranks as more off-meta than Quas or ZionSpartan, especially when looking at each’s trademark champions. However, keep in mind that CLG had Seraph play Shyvana deep into the summer split, well after she was popular. Meanwhile, before his Swain game on Sunday, Curse/Liquid had Quas playing pretty much all meta for the 2014 summer split and beyond. Also, we didn’t include Zion’s legendary Nasus game against TSM, because it was during the playoffs.

    The third tier, consisting of Gamsu, Avalon and CaliTrlolz, contains players new to the LCS who have made a weird pick or two. Cali was actually dominating this category after 6 games, but he succumbed to the meta by playing Maokai twice during Week 4.

    And, finally, standing in a tier of his own, is InnoX. The reason the Canadian national is so far in front of his peers is twofold. First, he would very rarely play the top meta champion. Out of 56 picks, InnoX only played the most contested champion on his current patch 5 times. For reference, Balls did this 19 times out of 56, and he barely even played Lulu.

    Also, InnoX would seemingly pull champions out of a hat at times, as evidenced by his off-meta pool. The best example is this: Kha’Zix in the top lane has only been brought out three out of a possible 528 times in the past year for the NA LCS. Guess who played all three of these games in the same week? InnoX, of course. The man is truly an off-meta legend.

Jungle

    This is one of two uninteresting metas, with Elise and Lee Sin as consistent forces throughout. It was comical how some junglers would tunnel on certain champions; A shining example of this was Kez, who picked Elise in an unheard-of 70% of his last twenty games.

    This is one role in which the “old guard” has been more off-meta than the newer players (Rush is so high because Lee Sin has only been “popular” rather than “mind-achingly popular” recently). This may be because meta junglers like Amazing were forced onto strange champions after being banned out. This banning out was made easier because of the smaller jungling champion pool.

    The only three players of note here are Crumbzz, IWillDominate, and Snoopeh. Crumbzz could certainly play the popular champions, but was afraid to branch out. Dominate has also marched to the beat of his own drum, even bringing out Nunu during his darkest hour (this season). However, Snoopeh was the only jungler who consistently pushed the boundaries of the meta. He only played the ubiquitous Lee Sin once, and he was the only jungler to play Mundo and Nautilus (although some would argue he was simply a few seasons behind).

Mid

    Keep in mind, this is after adjusting for the different eras. The veterans Bjergsen, XiaoWeiXiao, Pobelter, Hai, and Shiphtur have been the five most “in-meta” players of Season 5, so the new players look innovative by comparison. We’ll see how off-meta they truly are as time goes on. Also, keep in mind that Jesiz’s 3 Ahri games were on Patch 5.1, before she became 95% pick/ban, so his score is less meta than expected.

    A lesson one could take from this is that to stick around as a mid laner, you have to keep up with the meta. The best example of this is Shiphtur, who starting out as an incredibly off-meta player during the spring split, pioneering Katarina, Ahri and Syndra while finishing with a score of 33.1%. However, he seemed to have a change of heart during the summer, clinging to the popular picks while gaining 9 percentage points. A similar thing happened to Bjergsen from 2014 spring to summer, although this might have had more to do with Syndra becoming extremely popular.

    It is a little shocking that Pr0lly, Hai, and XiaoWeiXiao show up at the bottom, considering how deep the champion pools are known to be. (Who doesn’t remember Pr0lly’s Annie?) However, pretty much every mid laner in North America had an off-meta pick or two, so these three only look conformist by comparison.

    The highlights of the mid laners, of course, are the very strange champion pools of Scarra and HotshotGG. Scarra, like InnoX, seemed to roll a die each game to choose his champion. This die would have been heavily weighted toward Karthus, a champion I’m pretty sure half of NA didn’t know existed. HotshotGG, on the other hand, was pretty prescient with his Xerath and Zilean picks, playing them before the became popular in his region. Part of his extremely low score can also be explained by his stint on CLG White, where there weren’t exactly high stakes.

AD Carry

    First off, let us take a moment to honor Piglet for his refusal to play the North American meta. Half of his games were on champions no one else played (Vayne and post-rework Tristana). He might be staying at the top of this list for a while.

    The pocket picks for the AD carries were quite different from other roles. At mid, for example, Scarra and HotshotGG had four off-meta picks each. Each of the off-meta-leaning ADC’s, meanwhile, would generally have just one strange pick taken to different degrees.

    Some of these picks were a one-time thing, like imaqtpie’s Quinn or Cop’s MF. Others were comfort champions from a long-lost era, like yellowpete’s Urgot or WizFujiiN’s signature Varus. One stands above the rest, though; Doublelift’s Vayne.

    Vayne was not played by 12 of the 13 North American AD Carries in Season 4. The one out of 13 played her 9 times, something that is completely unparalleled in the league (WizFujiiN playing Varus 4 times was the closest I could find to something like this). That, folks, is dedication to a champion. Doublelift also played enough Sivir while she was unpopular to edge out the Caitlyn-leaning Cop among long-standing carries. imaqtpie would have been a contender, but he faded to meta picks during the late summer split.

    Unfortunately, the North American support meta wasn’t very diverse during Season 4. When the two contenders for “most off-meta support player” have “guy who played Nami a few times” and “the only person who played Braum during Season 5” as their headlines, you know there were very few surprise picks.

    The “top” supports on this list got there by mostly staying off the Annie, Braum or Morgana trains, or at least by playing Nami a lot. Also, Aphromoo and Krepo had a few interesting picks, but like Quas, they were stuck on meta champs for too long to get a low score.

    All this is interesting, but you can’t tune into the LCS to watch one player something off-meta. You have to watch the entire team, so here are the extreme teams from each season (sorted by average score):

Spring ‘14
Most Off-Meta: Dignitas (48.9%), Evil Geniuses (49.1%), Curse (49.7%)
Most On-Meta: Cloud 9 (59.2%), TSM (58.3%)

    Shiphtur and imaqtpie did just enough to edge out the InnoXes, while no one from Cloud 9 was particularly close to average.

Summer ‘14
Most Off-Meta: Curse (50.4%)
Most On-Meta: compLexity (55.4%), Cloud 9 (55.4%), LMQ (55%)

    Curse won by a landslide, as IWillDominate and Cop proved to be much more innovative than their counterparts. On the other hand, Vasilii was the saving grace for LMQ, while Cloud 9 continued the C9 way and compLexity kept playing Elise.

Spring ‘15
Most Off-Meta: Liquid (49.1%), Team 8 (49.1%), CLG (50.0%)
Most On-Meta: Cloud 9 (59.5%)

    Curse’s AD carry position continues to deliver low scores, but two off-meta top laner-led teams are breathing down their necks. And at least Meteos is starting to branch out.

    Here is the chart for full franchises:

    The past two champions are either in or near the basement of this statistic. Maybe there is something to the meta thing after all. Also, it’s pretty hard to believe that InnoX and Altec played on the same team.

    Keep your eyes peeled for the EU LCS edition coming soon, which at least should be more exciting in the support department.