Monday, September 29, 2014

The Mid-Game Power Spike: How Important Is It?


                                
It’s a phrase you’ve heard the analysts say dozens of times before a game starts, when they predict what strategies we’re likely to see being used. For those who are new to the concept of champion power curves, a mid-game power spike is exactly what it sounds like: a large increase in the power of a given champion relative to the other champions in the game. Power spikes usually happen after one or two key items are bought, or after a certain level is reached (to unlock an ultimate or max an ability).

The Theory

It’s simple: these spikes are so important because in most professional games, the midgame is when one team really starts to develop a lead. It’s also when the real team fighting starts, so you can see why being powerful at this time is so vital.

The Practice

This all makes sense in theory, but how applicable is it to actual competitive League of Legends? Well, as it turns out, the answer is a little bit complicated. In terms of raw statistics, in Groups C and D, the team with more champions with mid-game power spikes won 52% of the time. That doesn't sound like it makes much of a difference, but the team with fewer mid-game spikes actually only won 20% of the time. These numbers may not seem to add up, but in 28% of the games, both teams had the same number of mid-game champions.

However, some games made it very clear that these champion picks are very important. In the game with the most mid-game champions, Samsung Blue’s crushing victory over LMQ, four out of Blue’s five champions all had considerable power spikes after they completed a core item or two. Sure, Blue is a better team in general, but in the other meeting of these two teams (where Blue didn't have four mid-game champions), the game was much closer.

You can actually also draw conclusions from the games in which both teams had the same number of mid-game champions. The first game between Fnatic and LMQ, for example, was decided largely because LMQ’s mid-game champions had a much greater impact than Fnatic’s did. Ackerman on Rumble went 4/0/7 and XiaoWeiXiao on Yasuo went 4/0/8, compared to sOAZ’s 0/1/3 Lulu and Cyanide’s 1/6/4 Jarvan. Fnatic’s victory over Samsung Blue happened in much the same way. Most notably, Rekkles had an impressive 8/1/5 score while Dade went 2/5/1 on Zed, one of his signature champions.

The Odd Case of KaBuM

Now, it is sometimes the case in any competition that some teams just massively outclass others. For almost the entire set of games in Group D, this was the case for KaBuM eSports. They would stand up to any team for the first few minutes of a game but fall behind before too long. This isn't to put down any team - it’s great that KaBuM made it to worlds and got to play against some of the top regions, but for the sake of analysis, let’s see what happens when we don’t include KaBuM’s losses in our sample. Not much really changes; the team with more mid-game champions still won 55% of their games.

But here’s the really fun part: KaBuM upset Alliance on the final day of the group stage. Each team had two strong mid-game power spikes: Fizz with a Lich Bane and Twitch with a Blade of the Ruined King for Fnatic, against Ahri with a Zhonya’s Hourglass and Ryze with a stacked Rod of Ages and stacking Tear of the Goddess for KaBuM. LEP was not inspiring on Ryze this game—1/3/1 at the 16-minute mark—but some unconventional itemization from Minerva’s Jinx made up for this: the second item he bought was a Hexdrinker. An item rarely seen in professional matches at all, the Hexdrinker has been almost exclusively purchased by top-lane bruisers. However, a 4/1/1 Fizz with a Lich Bane is a scary proposition for any AD carry, especially an immobile one such as Jinx. The usual response to this would be a late-game Banshee’s Veil, but Minerva wanted safety from Fizz right then—and rightly so. Hexdrinker is a much cheaper item, and gives very good protection against burst magic damage. In purchasing this item, Minerva created an artificial mid-game boost in power on a champion who traditionally has a milder version of Tristana’s U-shaped power curve: fairly strong early laning, a dip in power mid-game while farming up for items, and then an explosive late-game as she approaches a full build.

In short, the mid-game power spike is a very useful tool which can really impact the outcome of a game, but a team has to set it up starting from champion select and continuing with their itemization, and then know how to use it.

No comments:

Post a Comment