Boards›Esports

Fantasy LCS: Do Slow Losses Outscore Fast Wins?

Riot·7/19/2014, 1:38:31 AM·0 votes·1,245 views

Have you ever wondered: do players in lengthy losses score more points than players in quick wins? Losing players who play in lengthy matches typically participate in more team-fights. And longer matches also give more time to grab kills and assists; both stats are vital to a player’s fantasy point output (kills and assists contribute roughly 90% of a player’s total points). Meanwhile, players who win games in fast surrenders don't have the chance to rake in points so it's not even worth starting them despite winning. But do the facts back-up this argument?

TLDR: Nope. We took the fastest 25% of LCS games (an average of 28.9 minutes) and a winning player scored 19.9 pts/game on average. Then, we looked at the longest 25% of LCS games (an average of 51.25 minutes) and a losing player scored only 14.7 pts/game on average. When we further condense the sample size and compare the 10 fastest winners (17.7 pts/game) to the 10 slowest losers (15.9 pts/game), the trend holds true.

Sprinting to the Finish

Week 1's match between SK Gaming and Copenhagen Wolves is a great example of a team that can quickly close out games, yet still score a healthy amount of points. This game lasted just over 23.5 minutes, yet SK players averaged 19.5 points. There weren't even that many deaths between the teams with just 4 kills occurring before the 13 minute mark. However, SK was able to snowball their advantages quickly, culminating in a 21-3 stomp over the Wolves.

Similar games such as Counter Logic Gaming's Week 6 victory over Dignitas (23.5 pts/player in 29.5 mins) and Copenhagen Wolves' Week 7 victory over Alliance (17.6 pts/player in 29.6 mins) are other indicators that suggest rooting for a quick close-out is actually not so bad for your fantasy team (it's better than dragging out games, only to lose anyway).

The Slow and Painful Death

When we think about a team's mentality when they get just a little behind in the early game, things begin to make sense. They aren't getting blown away so they are able to survive, play defensively, and wait for any opportunity to capitalize on. But this can lead into a quiet, kill-less game where the team in the lead chooses to not take many risks because the trailing team turtles effectively.

Unfortunately, Fantasy LCS only cares about your players' points, and losing slowly (while better than losing quickly) still doesn't match the points earned from winning quickly. Stagnant games such as Alliance's Week 7 loss to Fnatic (12.4 pts/player in 55.3 mins) and Cloud9's Week 7 defeat at the hands of Curse (5.4 pts/player in 53.9 mins) support this claim.

Looking into Alliance's loss, they fell behind just 2k gold by the 22 minute mark, only a slight disadvantage. But the kill score stayed 5-6 (in favor of Fnatic) from 22:41 to 32:01. And there was another kill drought from 34:11 to 41:45. Alliance was strong enough to prevent Fnatic from snowballing the game, a common trait in many drawn-out games. Their late game threat of Kog'Maw protected by Orianna, Lulu, and Thresh allowed the team to turtle out as long as they wanted to.

Okay, cool. Now show me something practical.

Assume SK and SHC each play against GMB for the week. You must choose who to start between CandyPanda (SK Gaming owns some of the fastest victories in the LCS) and MrRalleZ (SHC averages the longest matches across both regions). Even though MrRalleZ has more fantasy points, you may want to take a second look and start CandyPanda. 2 of SK's win against GMB have been around the 30 minute mark while SHC's single loss to GMB was just over 50 minutes. So while both SHC and SK are favorable matchups against GMB (and you could start either AD Carry despite RalleZ owning more points), SK's early game domination trumps in this roster decision.

Hopefully this article has dispelled the belief that "I'd rather have a losing player as long as their game goes long, than a player that wins as part of a quick surrender." Any other Fantasy LCS questions you want answered? Leave a comment here or let me know on Twitter.

4 Comments

FrozenFireV7/19/2014, 6:51:01 PM3 votes

What's the label on the x-axis? The graph looks counter-intuitive.

N4ge7/21/2014, 10:25:02 AM2 votes

The data doesn't tell the story here. It's not a matter of quick wins vs slow losses. It's a matter of teams who prioritize fighting over teams who prioritize objectives.

That's why Millenium is such a poor team to own, and yet their players are each individually at or near the top of their respective roles.

This article has the right idea, but it doesn't differentiate between a short game where kills were prioritized and a long game where objectives are prioritized.

It goes in this order: Long games where kills is priority > short games where kill is priority > short games where objective is priority > long games where objective is priority.

See, long games where teams go for objectives rather than kills are the worst for a fantasy team, because if the game is going long in that case, it indicates one or both teams are unwilling to commit to a fight over an objective, thus there will be few kills and less points, while an objective game will be short if a team commits to engage and obtains a massive advantage which they can snowball.

And that long kill based games are the best for fantasy players is proven by the roaming kill squad that is Millenium, whose games pretty much always drag on, but just because they're not good at forcing objectives and group up and run around the map fighting anyone they catch.

It is this logic that demonstrates why the two best teams to have players from are Millenium and LMQ, who both heavily prioritize getting kills.