Over the last few years, eSports has grown in market share more than any other sport, and betting and daily fantasy markets have emerged with it. Since these markets are new, they tend to be extremely exploitable, so it's worth our effort to learn as much as we can about the eSports ecosystem before the markets become more efficient. If you’re new to eSports, that’s not a problem: Check out my League of Legends introduction.
In this piece, I outline the three biggest eSports mistakes a DFS player can make and (more importantly) how to avoid them.
Mistake No. 1: Prioritizing Individual Skill Over the Probability of a Team Win
In NBA DFS, even if you expect the Cavaliers to lose you would definitely consider rostering LeBron James. Whether his team wins or loses, LeBron always has a chance to put up a lot of fantasy points. His talent matters more than his team's circumstances. In League DFS, the dynamic is different: Winning players are more important than good players. Because of the way games play out for victorious teams and how League DFS is scored as a fantasy game, winning players almost always outscore losing players regardless of how good they are.
Here's why. In League, as a team starts gaining advantages in its match, its champions get stronger. Imagine if basketball had a rule that changed 3-pointers to 5-pointers if a team were up by five points and 10-pointers if it were up by 10? Instead of seeing a lot of 101-98 games, you'd see more scores like 190-80. In that case, if LeBron were on the losing team and Danny Green were on the winning team, Green would probably be the more desirable DFS play, because all of his 3-pointers would be worth 10 points and his assists and rebounds would be worth eight points apiece.
Let me give some examples: Here are the win/loss splits for Pilot, the Attack Damage Carry (ADC) for MVP (a team in the South Korean league).
- Wins: 89.82 fantasy points per match (FPPM)
- Losses: 28.6 FPPM
Pilot is not an outlier. Here are the FPPM win/loss splits for Edge (the mid-laner for Kongdoo Monster, another team in the LCK).
- Wins: 88.4
- Losses: 35.1
In League DFS, rostering players on winning teams is more important than anything else. The only instance in which you should ever roster a player from a team you think is going to lose is if you literally cannot afford any player who has a chance to win. Also, since winning players are much more valuable than their losing counterparts, you should almost never game stack players on opposing teams.
Mistake No. 2: Diversifying the Composition of Your Lineup
It's important to roster winning players, but not just any combination of winning players will do. Let's say that you've analyzed a slate, and you think the winning teams will be Fnatic, Splyce, H2k and Vitality. You can't just grab the best players from each team and build a lineup. Even if the roster and salary constraints would let you do that, it wouldn't be an optimal strategy. In general, you want to stack winning teammates as much as you can. Why? Because kills (three points) and assists (two points) in League DFS are extremely important — they're the main means by which players score — and kills and assists are accumulated in a highly correlated manner.
Let's return to the LBJ-Green NBA example. Let's say that LeBron's Cavs had an implied Vegas total of 100 points and Green's Spurs were implied for 120 points. As already established, you'd want to roster Green over LeBron — and you'd also want to stack Green with his teammates, because every time one of them scored it would be highly likely that all of the Spurs on the court at the time would get an assist. That's right: One basket, four assists. That's basically the dynamic we have in League DFS. The majority of points are scored in team fights, and in team fights when one player gets a kill, it's common for all teammates to be credited with assists.
It's not enough to roster players on winning teams. You want to stack many players from a winning team — hopefully a team in a high-scoring game.
Mistake No. 3: Going Underweight on Carries
Carries are a type of champion, and they are usually the ADC and mid-laner. Carries will almost always accrue the most points on a team for a few particular reasons.
- They get most of the kills.
- They get more minion kills than anyone else.
- They (along with the jungler) are most likely to pick up the early game kills before everyone else on the team gets involved in the action.
Because of these factors, you should always look to load your team with carries when possible. If two rosters are entirely filled with winning players, the lineup with the most carries will almost always be the one with the higher score.
Because moneyline odds and fantasy pricing are often correlated, it's expensive to stack players from the biggest favorites on a slate, which means that if you go heavy on the biggest favorites you'll probably have room in your roster for only two carries. With such a lineup construction, you might have a good chance of rostering a lineup full of winners, but you won't have the necessary points to dominate in Guaranteed Prize Pools (GPPs), because other lineups loaded with winning players and more carries will have higher scores. In the long run, especially in GPPs, it's better to roster more carries (who still have decent moneyline odds) than to roster players who are the biggest favorites to win.
Mistake-Free League
There are lots of mistakes you can make when playing League DFS, but if you focus on teams likely to win, stack heavily, and prioritize carries then you'll be avoiding the biggest mistakes that keep people from finding success.
Until next time, good luck . . . and I'll see you on the rift.
