Note that both Xenics Storm and KT Rolster Arrows have undergone roster changes since the 2013-14 Champions Winter season, but the analysis provided here is still relevant.

Every aspect of SK Telecom T1 K’s play was phenomenal against Samsung Ozone in the finals of OGN Winter, but I want to highlight one thing in particular that they did exceptionally well and that other teams have started to do over the course of OGN and NLB Winter—roaming their support away from the duo lane to assist the rest of the team, particularly midlane. It’s something that harkens back to season 1 play, is extremely exciting to watch and could solidify the dominance of Thresh, Leona and Alistar - but primarily Alistar - as the best play-making support champions in the game.

The two teams other than SKT that stand out especially as sending their supports to roam are Xenics Storm and KT Rolster Arrows. In the group stage of Champions Winter, Storm would attempt to 2v1 lane-swap and then send Piccaboo on Alistar or Thresh to roam and make plays with their jungler Swift. Even though Piccaboo would end up significantly behind on experience, he and Swift were frequently able to snowball their teammates enough to make up for the experience gap between the two supports. NLB team KT Arrows employed the same strategy in the NLB with their support Hachani, who played similar play-making supports like Thresh and Annie - and even roamed mid to tower dive at level 2 in their second game against Samsung Blue (which they won, though Blue won the overall bo3).

Both of these teams would force counter-roams from the other support but would still dominate the “laning” phase - as long as they were able to have their support roam. Other OGN and NLB teams initiated a roaming support strategy from time to time as well, but these two were among the most notable, especially due to their strong performances against top teams when able to make plays with a roaming support but weaker performances otherwise. The only problem seemed to be that without the 2v1 lane-swap, a roaming support couldn’t be made to work.

Let’s look at the specifics of SKT’s support roams, though. Game 1 of the finals saw SKT initiating a 2v1. Disregarding roaming supports, the call seemed a bit surprising, because even though Sivir can outpush Vayne, in general a Leona lane will be slower at pushing towers than an Annie lane, and Sivir/Leona should be able to shut down Vayne/Annie decently well. However, around five minutes, we see the following rotation down to midlane:

SKT actually decided to send ManDu mid prior even to seeing DanDy (Olaf is still in Fog of War in this screenshot). Ozone responded by sending Mata up from botlane after ManDu passed the river ward, but the fact that ManDu arrived significantly earlier saved Faker from death. Because of Sivir/Leona’s waveclear, and the timing of Bengi’s gank, it was almost not a sacrifice at all to send ManDu midlane, and once again we are reminded of the importance of a team being the first one to have their support arrive to a fight.

In game 2, we see a similar rotation out of SKT, sending both ManDu and Piglet mid to assist Faker - this time, though, the lane matchup was 2v2 bot, and SKT just took advantage of recall timings to find this window. Ozone responded to the pressure midlane aswell, though they only sent Mata and left Imp botlane to collect farm under tower. Ozone’s response was correct here given that no one died in mid, Imp’s presence in mid couldn’t have resulted in a kill in their favor, and Imp was able to soak up the experience that he needed.

Note that if either Mata or ManDu and Piglet had walked straight down midlane and arrived earlier, the respective team would have picked up a kill here.

The important thing to get out of these two games’ rotations is that SKT considers their support as a global resource, just like the jungler, and proactively has ManDu roam, even if the lane matchup is 2v2. It’s been common for a while to have supports assist in jungle fights, but SKT anticipates that a fight starting in midlane will escalate before it even starts and so they decide to devote every resource possible to that fight, which includes the support.

The support roams out of SKT in game 3 were perhaps the most interesting, as it is an evolution of the continual lane-swap meta that was seen in Season 3 in every region. Ozone wanted a 2v1 against Faker’s Riven more than anything else, and their Sivir/Leona lane could outpush Alistar/Lucian. The season-3 response to an unfavorable 2v1 matchup was for a team to continually shove out lanes, lane-swap to regain 2v2, shove again after the first team went back to 2v1, lane-swap again, etc. This would result in a lane-phase victory for either the team with faster pushing or superior ability to make pick-offs during a rotation.

SKT’s response here to an unfavorable 2v1 that they couldn’t easily swap to avoid (unfavorable because Ozone was getting more damage on turret than they were and difficult to swap out of because Sivir/Leona outpushes) was instead to simply send ManDu mid to back up Faker, while Piglet stayed bot to farm against Looper. This minimap screenshot shows one of the several times ManDu left Piglet to go defend mid.

These SKT games show us that even if the 4.1 patch does kill 2v1 lanes completely in Korea (though unlikely given the number of 2v1 swaps that have happened already in LCS) and the Xenics Storm/KT Arrows style of having a permanently roaming support is no longer viable, support players will probably be active on the map anyway. This shift in the metagame should mean much more complex and interesting early-game play, even if fast-push rotations are no longer available, and overall is a positive change that we can be excited to watch develop.