The Quest for S-Tier: How Good is SKT's Duke?

by theScore Staff May 4 2016
Thumbnail image courtesy of KeSPA / LCK Spring 2016 Final

Lee "Duke" Ho-Seong of SK Telecom T1 is the best top laner participating in the 2016 Mid-Season Invitational. In terms of player comparisons or rankings, this is about as close as you can get to an objective fact. In his first two games at MSI, Duke has earned a combined scoreline of 7/0/11 on Ekko and Poppy. Against international wildcard representatives SuperMassive, Duke easily snowballed out of the lane phase. In a stiffer test against G2 Esports, Duke played clean and controlled, filling his role while the team built a win around the middle and bottom lanes. SKT hasn't needed Duke to be a superstar — but is he one?

If we were to define the global "S tier" of top laners, the absolute best of the best, would Duke's name be included? Is Duke truly on par with players like LCK Spring MVP Song "Smeb" Kyung-ho of the ROX Tigers, or perennial superstar Kim "Ssumday" Chan-ho of KT Rolster? Is he really a step above the A-tier talent of a player like the Jin Air Green Wings' Yeo "TrAce" Chang-dong?

It pains me a little to leave Jang "MaRin" Gyeong-Hwan off the shortlist, but he's not playing like the 2015 World Championships MVP these days. And though there are strong top laners elsewhere, such as North America's Heo "Huni" Seung-hoon, Europe's Andrei "Odoamne" Pascu, the LMS's Chen "Ziv" Yi, and China's Li "Flandre" Xuan-jun, none of these players match up to Korea's best.

The suggestion that Duke might not be S-tier has drawn vocal reactions from some, but common opinion from other sources seems to see Duke as inferior to both Smeb, the current consensus as the world's best top laner, and Ssumday. Where opinions fail to bring clarity, a more detailed analysis might shed light.

Duke's 2016 LCK Spring performances show that while he may not match Smeb's all-around excellence, Duke is certainly in the discussion. A closer look at both statistics and in-game footage shows how favorably Duke compares to some of his strongest opposition.

Duke is more than just a humble, uninspiring A+. Duke is S-tier, one of the best top laners in the world.

Defining S-Tier

Before throwing around facts, figures, and opinions, we need a foundation for the discussion by defining what an S-tier top laner actually looks like. To me, an S-tier top laner excels in most or all of the following areas:

  • Laning: Winning 1v1 lane matchups as and against a wide variety of champions.

  • Playing carries: Doing damage, securing kills, and positioning wisely.

  • Playing tanks/utility: Initiating fights, peeling and protecting, and/or locking down the enemy backline.

  • Split pushing: Applying pressure in side lanes in coordination with the rest of the team, as a way of gaining map control, securing objectives, and/or setting up team fight initiations.

It's possible to be an S-tier top laner without achieving perfection in every single one of those functions, but if you're an average laner, tank player, or split pusher, you'd better make up for it by truly excelling elsewhere.

To use Smeb as an example, holding him as today's highest standard, we can point to him as a very strong 1v1 laner; a versatile player who can play a variety of roles, including tanks like Poppy, carries like Fiora, and utility champions like Lulu; and a great split pusher who effectively coordinates his pressure with his team.

Champions and roles

There's one more piece of crucial context — since the top lane is a pretty diverse place, and different champion types can produce different statistical profiles, we need to have some idea which champions each player tended to play this split. Here are the most common picks for each player in the Spring regular season, and the overall size of their champion pool.

Duke Smeb Ssumday TrAce
Poppy x8 Poppy x10 Poppy x12 Gangplank x7
Maokai x6 Fiora x5 Fiora x7 Poppy x7
Fiora x5 Nautilus x5 Maokai x5 Nautilus x6
Quinn x5 Gangplank x5  
11 unique picks 11 unique picks 14 unique picks 13 unique picks

While each of these players spent some time on carry champions, especially Fiora and Gangplank, Duke trended a little more towards tanks than his counterparts. Across 41 games, Duke played carries just 11 times and tanks 30 times (counting Gnar and Ekko among the tanks). Compare that to Smeb's 22-of-41 games on carries, Ssumday's 16-of-43, or TrAce's 17-of-41.

Duke's higher-than-average rate of playing tanks means that we should expect to see somewhat lower damage output than the others.

The numbers

Let's get into the analysis with a statistical comparison. In the following tables, we'll focus on highlighting and explaining any meaningful differences between Duke and his fellow top laners, starting with some overall stats, then breaking things down further for some different elements of the game. All of the statistics below reflect the 2016 Spring regular season, to keep the playing field even.

Overall stats

Statistic  Duke  Smeb  Ssumday  TrAce
Win Percentage  63%  83%  67%  56%
KDA Ratio  4.2  5.4  3.9  4.1
Kill Participation  65.0%  67.0%  72.6%  70.4%
Death Share  20.3%  21.4%  22.8%  21.1%
Damage per Minute  388  507 487  431
Damage Share  19.1% 24.4%  24.9%  24.2%
Gold Share  22.1% 22.6%  23.2%  21.5%

Across the board, these stats paint a pretty positive picture of all four top laners we're looking at — there's a reason these are our four candidates. We've got four strong KDAs, with Smeb jumping out ahead of the rest. We also have the four highest DPMs of all LCK top laners, though Duke lags behind in damage share — as expected, based on the champion pool we discussed above. Gold shares are a bit spread out, but we'll get a better idea of why that's the case when we look at the farming stats later.

Kill participation and death share tell a more nuanced story. Duke has the lowest numbers among this group in both stats, which shows that he is generally less involved in his team's combat. That could point to a few different explanations, some positive and some less so. First, it could mean that SKT does a lot of skirmishing, with small fights elsewhere on the map, but that equally describes Smeb's ROX Tigers, and Smeb's individual numbers are higher.

It could also mean that Duke is a heavily dedicated split pusher, and that his team fights elsewhere on the map while he's pressuring Towers. That holds some truth for Duke, but it equally or better describes Ssumday, who also has higher kill participation. Third, and less flatteringly, Duke's low kill participation and death share could mean that he doesn't arrive at some fights, and simply isn't involved. That explanation did, unfortunately, apply to Duke in some of his games in the regular season, due to coordination and Teleport issues that we'll make note of later, but it's worth pointing out that in the playoffs, Duke's kill participation jumped up to 73.8 percent, highest on his team, a sign of the improvement SKT showed as a team in learning how to play with one another.

Overall, these first stats don't make Duke stand out too much, for either positive or negative reasons.

Lane phase and early game

Duke's 1v1 laning is one of the strongest aspects of his game, and the stats below show it.

Statistic  Duke  Smeb  Ssumday  TrAce
CSD@10 (standard lanes)  +9.5  +4.1  +2.2  -5.4
GD@10 (standard lanes)  +161  +205  +48  -182
First Blood Rate  17%  34% 26%  29%
Kills + Assists at 15 minutes  0.9  1.9  1.2  1.1
Deaths at 15 minutes  0.3  0.7  0.6  0.4

In games with 1v1 and 2v2 lane assignments (flagged by tracking games where no tower was killed before the 5-minute mark), Duke dominantly led all LCK top laners in CS Difference at 10 minutes with +9.5. Smeb and Ssumday trailed behine, with TrAce a distant fourth.

Smeb's advantage in gold difference at 10 minutes in standard lanes speaks to the ROX Tigers' faster-paced play, which produced more kills in the early going, allowing Smeb to pick up gold from kills and assists. And the players' First Blood rates reveal the gap in effectiveness and focus of Smeb's and Duke's junglers — Peanut is a much better ganker than Blank and Smeb is more likely to receive gank resources than Duke is, since Duke is sharing the Rift with two of the world's best carries.

Duke is, statistically, the best 1v1 top laner in the LCK, and probably in the world. The longer Duke is able to sit in lane and butt heads with the enemy top laner, the happier he seems to be. He's capable of using roams or Teleports to influence the other lanes in the early going, but he prefers to stay at home when he can, and leverage his individual skill.

Farming and vision

Based on CS per minute and post-15-minute CS share, Duke soaks up a decent proportion of his team's farm, and split pushes a fair, but not exceptional, amount.

Statistic  Duke  Smeb  Ssumday  TrAce
CS per Minute  8.3  8.0  8.1  7.5
CS Share Post-15  26.6%  26.7%  28.1%  24.8%
Wards per Minute  .44 .49 .56  .51
Wards Cleared per Minute  .14  .14  .20  .17

Duke averaged the highest CS per minute of all LCK top laners, but his proportion of post-15-minute farm, which is usually the best barometer for split pushing, was middle of the pack. Ssumday is the standout split pusher, taking in over 28% of his team's post-15-minute farm, the most of any starting top in the LCK. That's a big part of the explanation for Ssumday's huge gold share, which we saw earlier.

To aid his split pushing, Ssumday places and clears more wards than most top laners, while Duke's ward output was the lowest of all LCK tops. The actual raw difference in warding numbers between Duke and the others isn't that big, though.

Both through these statistics and through watching Duke play, it's clear that he's a powerful split pusher, even if his team doesn't employ that tactic as often as Ssumday's.

Duke's Strengths and Weaknesses

Duke's primary strength lies in his 1v1 laning and, more generally, his powerful early game play. If left unchecked, Duke is capable of building himself up into an unstoppable monster. For examples, look to game 2 of SKT's week 5 series against KT Rolster, a big carry performance as Quinn that featured a solo kill at 14:00, or this pair of strong Trundle showings, which showed how Duke can use his early-game leads both to become a powerful split pushing threat and to influence team fights.

Over the course of the Spring season, Duke's play style shifted somewhat. Early on, he defaulted to split pushing more often, both because there were more top lane carry champions in the meta (such as Fiora and Quinn), and because it was the way Duke was used to playing with his previous team. As time went on, Duke incorporated more and more tanks into his game, to fit the meta and to accommodate Blank's greater effectiveness on carry junglers than on gank-focused champions. While he continued to split push, he also became a more dominant team fighter, soaking up enemy damage and cooldowns and camping the enemy backline. For an example of Duke's team fighting, check out the final fight from game 3 of the LCK Finals, which featured a game-winning Poppy ultimate and high-value frontlining. Duke's team fighting has fewer flashy moments than Smeb's or Ssumday's, but it is still a high point of his game.

Duke has weaknesses, as well. His split pushing has showed some vulnerability, allowing opponents to pick him off when he extends too deep. That's always a risk to split pushing, and certainly Ssumday gives up plenty of split pushing deaths, but Ssumday's deaths seem to more often result in cross-map trades that put KT Rolster ahead. Ssumday's diligent vision game may have something to do with that.

Another of Duke's weaknesses has been the timing and placement of his Teleports, which has sometimes limited his ability to effectively initiate or join into team fights. Here are some examples. But calling Duke a poor Teleport user is an exaggeration — for every poor Teleport, he lands multiple good ones. And sometimes, the issue with Duke's Teleports is that his team doesn't receive them properly, as in this example. That points to an issue with overall team communication, not an issue with Duke as an individual.

The Verdict

Duke's game carries imperfections, but his strengths are undeniable. Duke is arguably the best 1v1 top laner in the world, his teamfighting is very good, and he's a great split pusher, especially when the meta allows him to pick up duelist champions like Fiora. Meanwhile, his much-discussed Teleport game is less a weakness than you may have heard.

In asking whether Duke is truly an "S-tier" top laner, we're really asking whether Duke matches up to the elite standard set by Smeb and Ssumday. Based on his strengths and weaknesses, Duke is as good as or better than Ssumday — a better laner and equally effective team fighter, though slightly behind in his split pushing and less creative in his champion pool — but Duke isn't quite on Smeb's level, because of how well-rounded Smeb is.

If S-tier is truly a tier, not just an award that goes to the individual best player, then yes, Duke is an S-tier top laner, one of the world's very best.

In today's matches against Royal Never Give Up and the Flash Wolves, Duke will look to cement his status even further.

Tim "Magic" Sevenhuysen runs, the premier source for League of Legends esports statistics. You can find him on Twitter, unless he’s busy giving one of his three sons a shoulder ride.