Losing With Grace: Fantasy LCS

On the other hand, some teams are remarkably resilient in losses. While they do perform worse in losses than wins, they do so to a lesser degree.

As you can see, there is actually a significant correlation between a team’s record and their resilience in losses. The top teams of NA and EU, like CLG, Dignitas, Cloud9, SK, and Alliance all are towards the lower end (a much smaller Win-Loss gap), while the lower record teams, like Evil Geniuses, compLexity, and Copenhagen Wolves are all towards the higher end (a much larger Win-Loss gap).

Why exactly is this?

The first suggestion that comes to mind is this: game length. SK Gaming, for example, has become quite known for closing their games out quickly when ahead, with the shortest average wins across both LCS regions (32:51). On the other hand, when losing, SK Gaming manages to cling onto games for an incredibly long time, with the 4th-longest losses (42:16). Perhaps their short wins prevent them from racking up the points, while their losses stretch on long enough to earn them a bunch of points? Turns out that theory doesn’t hold water, once you look at the data:

Yes, there is a slight correlation. Teams who average longer wins compared to their losses also tend to have a slightly larger point gap between their wins and losses. But the correlation is so slight it’s probably just due to random variation.

Snatching Defeat from the Jaws of Victory

Another theory is a little bit closer. It says that the best teams simply win and lose differently than the other teams.

CLG in particular, is very unique; the team has been playing extremely well, with most of their losses coming to the top teams in North America. But several of their losses could easily have been victories for CLG. We can see this especially in their first 3 losses to LMQ and Cloud9.

CLG vs LMQ - W1D1

In their first game vs LMQ, CLG seemed slightly in control of the game, wavering between a 2k gold lead and an even game. But at the 19 minute mark, after they took down NoName and Mor, they rushed onto Baron because XiaoWeiXiao was all the way in bot lane. However, LMQ had a strong team-fight composition, with Twitch, Yasuo, and Shyvana while CLG had a poke/kite comp, with Nidalee, Lulu, Sivir, Lee Sin, and a yet-to-be-farmed Jax. Most importantly, CLG had no tank, and took an obscene amount of damage from Baron before LMQ arrived, losing the resulting team-fight and ending up behind by 2.5k gold.

Even from there, CLG stayed in the game for another 20 minutes - eventually losing at the 40-minute mark - picking up towers and an inhibitor but losing in gold and in team-fights.

C9 vs CLG - W1D3

CLG’s first game vs Cloud 9 was extremely hard for them, and they fell behind by 4k gold by the 13-minute mark. However, they had an extremely strong late game composition with Vayne, Orianna, and Irelia, and managed to hold on fight after fight despite a huge gold deficit.

C9 vs CLG - W3D2

In their second loss with Cloud 9, CLG had such a dominant foothold in the game that they ended the game ahead in gold, 57.4k to 54.2k, despite being down 4 towers to 8. At the 22 minute mark, they took the game’s first Baron while up 5-0, ending up 6-2 with Baron buff and a 7k gold advantage:

From there, they maintained a 5-7k gold lead for the next 12 minutes, but Cloud 9 managed to claw back, take some objectives, and ultimately win a decisive team-fight and push for the win.

Tenacity

Basically, CLG shows extreme tenacity in their losses. It’s not that their losses are long by any means. It’s that the majority of their losses this split have come in two specific ways. Either they started ahead - and had stats befitting a victory - before throwing the game, or they were behind but refused to be shut out of the game, taking uncontested objectives and winning skirmishes they had no right to win.

We can see this particularly clearly by looking at the rate which teams accrue points, or the Points Per Minute Per Game. PPMPW and PPMPL are the Points Per Minute Per Win and Points Per Minute Per Loss. Essentially, when you’re winning, you should get more points quicker than when you are losing, because - well, that’s what winning is.

So it’s really a combination of two factors that makes a team safe to play when they might lose. Some teams cede objectives they know they can’t contest, preventing team-fight losses. Other teams simply drag games out as long as possible, waiting for the opposing team to make a mistake. In both cases, it is the better teams who are much more likely to do so.

Here’s a weird fact. The winningest teams don’t actually score the most fantasy points in their wins. In fact, they seem to score significantly fewer points in their wins.

On the other hand, they seem to score about the same points in their losses.

Ultimately, what this means is clear. What makes a team good is the ability to win the game, not to rack up fantasy points.

When behind, better teams know how to claw their way back into a game, turning losses into wins. But winning a game in this fashion doesn’t actually yield all that many fantasy points, as these teams will often only one win one decisive team-fight before pushing to win.

When ahead, they also know how to prevent their opponents from clawing their way back into a game: win the game as quickly as possible, at the expense of a growing fantasy score.

Essentially, the teams at the bottom of the standings only win games when they are stomping. This means that their wins are way higher fantasy lodes. On the other hand, higher level teams win a lot of games in which they were initially behind, leading to a lower average number of fantasy points in their wins.

What does this mean for your fantasy team?

If you think your Copenhagen Wolves or compLexity players are going to lose in the upcoming week, you should probably bench them. On the other hand, if they have an easy schedule, they have the potential to outscore your “star” players from the top teams. However, if you have a spotty player and a star player both facing a hard schedule, with both likely to win one game and lose the other, you may just be better off with the spotty player.

The players you should start most of the time are those who outscore the average for their position most:

While most players score the average score for their position in their wins, the following players score the highest in their wins:

Finally, if all of your teams have bad matchups, the following players perform the best in losses for their positions. Yes, all these players still score less than the overall average (they're losing of course), but these players will only be losing you 3-5 points on average:

Now that you've gained some additional insight into the LCS pros' performances, you should be a bit more prepared to dominate your fantasy leagues. Good luck out there!