I’m back, and have I got some numbers for you! Week 9 saw a fitting finale to the LCS NA Summer Season, with some high points (and devastatingly low points). Scroll on down for final season leaderboards, then some special Week 9 analysis:
Regular Season Team Leaderboards
Regular Season Player Leaderboards
Regular Season Role Ratings
Week 9 Role Ratings
Full Regular Season in Numbers
Team K/D and KDA
Team Kills, Deaths, and ***its Per Minute
Team Gold and GPM
Team CS and CSPM
In general, trends that were noticed last week have continued. CLG spent hours on the battlefield after the other teams went to bed. C9 and Vulcun thoroughly slaughtered teams with panache. And Velocity were stuck googling what that means.
At some point, I (or a welcome volunteer!) will have to go back and record previous LCS NA seasons, along with LCS EU and perhaps other leagues, to build a solid benchmark. The per-minute totals will be especially revealing, especially across scenes. Of course, then you invite opponent-adjusted numbers like DVOA, and the statistics scene starts to resemble several dozens of individuals cranking these numbers out on a regular basis.
Player K/D and KDA
Player Kills, Deaths, and Assists Per Minute
Player Gold and GPM
Player CS and CSPM
Mancloud tops the kills chart with a dominant lead over the next-closest player - we should all chip in and get him a belt or something. Cloud 9 and Vulcun team members line the top of most of the important leaderboards. CLG make appearances in all the flat rankings due to their eternity spent waiting for the game to end.
I’m beginning to think a very telling analysis might lie in GPM versus CSPM. High CSPM is great, but as we see, objective control brought in a larger percentage of gold for teams with higher win percentages. Moreover, taking objectives may be a great way to indirectly measure a team’s performance throughout a match. Will definitely start exploring it next week once playoffs get rolling.
I redid all the formulas, feel free to check them out at the end of the article. The most noteworthy change was removing any CS penalty from Support. I also lowered the death coefficient for all “initiation” roles: Jungle, Top, and Support.
Sneaky pulled ahead of the crowd with a fantastic Week 9 performance (more on that later) while the rest of the pack fell in line behind him. Doublelift managed to rank third despite an absolutely awful final week. Maplestreet plummeted, in no small part, due to the wonky tactics used by Velocity in super week. Note: this is sadly a common theme this week for our 8th place team. At least they had a good time!
Jintae again sits at the top, though unless he plays in another LCS match, this will be the last time you see him. The skew is thrown off due to his performance, but you can see there is quite a lot of parity in the Top ratings over the season. Balls takes a solid lead like Sneaky. Cris improved a lot over the last two weeks, so it’s not all doom and gloom for VES.
What else is there really to say that hasn’t already been said? Meteos is far ahead of the pack, and many are excited to see him play against the best junglers of the world in the grand finals if Cloud 9 is to qualify. Nk Inc, Nintendudex, and Bigfatlp all grouped up. If you get the three confused, just remember that Nk Inc has glorious metal hair, and Bigfatlp has the best alternate nickname. Actually, some of that has got to be worth ratings points… right?
Mancloud reigned over the lot, as he sat happily on his throne made out of gold from champion kills. Ecco took a beating this week, falling from Rank 5 to Rank 8. Shiphtur also fell a bit on the board and is now tied with Reginald, who had a better than average week to finish just below the average.
Lemonnation dominated his role, but the real telling thing is the spread here. Lemonnation, Bloodwater, and Patoy all sit well above the average. Edward and Xpecial straddle the average, while Daydreamin, Chauster, and Evaniskus have some work to do to improve. If this were the NFL, we would separate them into tiers and argue what it means to be “elite.”
Wrapping it all up, we see that C9 and Vulcun are clearly the superior teams this season. The rest of the team ratings fall in fairly linear order, with only TSM as an outlier. A close look at all their performances reveal the only logical outcome - when they win, they do it by scraping by, and when they lose, they lose hard. Despite finishing in third, they seem to be underperforming, and may have troubles in playoffs.
A closer look at Week 9 (Games 24-28)
Average ratings over a season are great in some respects, but they tend to reveal very little about specific performances. As I hinted at last week, evaluation over a weekend, or even a single game, can be very revealing. So let’s see what the numbers say about Week 9. By the way, go check out the full recaps on lolesports.com if you want to know more (or watch those brilliant LCSRecap videos).
Sneaky of course blew up the house with a 28/4/29 outing, but the big story here is Zuna with a remarkable 36 kills over the 5 game period. That’s the most this week, even one more than his fellow teammate and renowned champion destroyer, Mancloud. Maplestreet, on the other hand threw in the towel with one of the lowest GPM scores I’ve ever seen on a marksman.
Without Jintae, the performances of these Top players becomes a lot clearer. While most of the players do well, Zionspartan, Kiwikid, and Cris all could use some more work. Also notable was a very good week for Dyrus which propelled him above Nien in the season ratings.
Meteos and Xmithie rise above their competition again, with stat-lines of 17/6/38 and 17/12/55. Nk Inc, while still below average, seems to be faring better than some of the other junglers, in no small part due to him having the third highest kill rate this week. Unfortunately, it was paired with the highest death rate, and should be his main target for improvement (and for the rest of Velocity).
Ecco and Shiphtur both had ratings greater than 1 prior to this week. That is no longer the case. Respectively, they fell from #5 and #4 to #6 and #8. With those poor performances, it’s no surprise to see the rest having a good week - Mancloud actually had the highest GPM with 435.278. That’s over 74 gold per minute more than the average marksman.
It was bad this week for Supports, well, only for some of them. Edward came out with an 11/10/64 stat-line, supremely impressive for a support. Velocity’s difficulties with securing kills is apparent here, with Evaniskus only getting 25 through 5 games. At least he got a dragon kill?
Here’s the summary from Week 9. Vulcun put up some impressive numbers against Cloud 9 to take the win, but as you can see, they did poorer against other teams in terms of performance. Hard to argue with results, though, and Curse is clearly on the rebound after a mediocre regular season. Edward seems to be clicking now and they have a good shot at advancing to the semis.
- Velocity numbers came out low, but at least they entertained the crowd with some unusual strategies. They aren’t as bad as they look on paper, but they clearly have room for improvement. We’ll see how they fare in the relegation matches.
- Mancloud’s 167 kills are a record and should be a great target for seasons to come.
- Cloud 9 has a solid chance to win the summer playoffs, but showed some cracks this week with a loss against Vulcun. A lot will be riding on these players, as everybody expects them to win, and the stress could be too much for them to handle.
- Curse is clearly on the rebound and Edward’s performance has increased dramatically.
- Quarterfinal 1: Curse 2:0 Dignitas - Cop and Edward put on a show, shaming Patoy and Imaqtpie in 2v2 lanes, and thrashing poor Kiwikid in any 2v1 lane situations.
- Quarterfinal 2: TSM 2:1 CLG - TSM’s luck holds, as a solid performance by Dyrus help them scrape past CLG to advance to the semis.
- Semifinal 1: Cloud 9 2:1 Curse - These two powerhouses play in top form and have the best games of the weekend, with Cloud 9 coming on top with excellent play from Sneaky and Meteos.
- Semifinal 2: Vulcun 2:0 TSM - Vulcun sends TSM packing in a one-sided match featuring Mancloud setting another high KPM rate.
- Third Place: Curse 2:1 TSM - Curse brings home a spot at Worlds with a victory over TSM. Edward and Cop are again heavily featured and dominate their lane.
- Finals: Cloud 9 3:2 Vulcun - Vulcun try their best and take an early 2:1 lead in the series, but Cloud 9’s impressive win rate finally kicks in and Cloud 9 seal the deal with a dominant final 2 games. Meteos will die less that 10 times throughout the series.
Until next time, folks. I’ve added a tool to mess around with the ratings formulas. Just change the weights and it’ll update all the coefficients and ratings accordingly.
Full Season - https://docs.google.com/file/d/0B0GZN5dyH_BbRzRwQlVEanZNZHc/edit