Smart Kills And Worthless Deaths In League of Legends

How much does an individual contribute to a team's performance? In this episode, we seek to answer this question, and explore the most important factors that improve team performance in League of Legends.

❓ An Essential Question

"How much does an individual’s performance contribute to a team’s likelihood of winning?... Like points scored or defended in traditional team sports, eSports measure kills and deaths. But, again like traditional team sports, these standard metrics do not do a great job of predicting team performance." [1]

To answer this question, the researcher used the League of Legends API to collect game data. In total, ~150.000 games from across all ranks were analyzed. A key twist to the study is its advancement of player data. Typically, kills, deaths, and assists are used to assess performance. However, worthless deaths and smart kills appear to predict performance way better.

"Worthless deaths are ones that do not increase your team’s win probability. Smart kills are ones that do increase your team’s win probability." [1]

To illustrate this, have a look at the figure below. The basic game stats (API)—both charts on the left—are okay but not great. "The better you do there individually, the more likely your team is to win. But, if you are really good, your team doesn't win as much as it should; and if you are really bad, it doesn't lose as much as it should."

Think about it this way: killing someone at a crucial time in the game would be a smart kill, whereas killing someone who's already 0:12 may increase your stats but doesn't add any real value. Similarly, a death without having any impact would be worthless.

🏆 What Influenced Team Performance?

In addition, some additional variables were used to get a deeper understanding of what influenced performance. As you can see in the figure below, only three stats were relevant to individual performance: favorable fights (smart kills and percentage of favorable team fight), deaths, and time management (how much useful activity players had across time—the game).

"The implied advice to be a better player is: fight good fights, don’t die, and don’t waste time." [1]

These three things have the highest impact on a team's win probability. Interestingly, the fourth variable that had an impact was survivability—damage taken per death. Surprisingly, the higher a player's survivability, the worse it was for team performance. Basically, a team full of tanks would do worse than a team full of carries. I assume that survivability comes at the cost of damage and therefore not winning team fights.

👍 A Rule Of Thumb And Map Vision

The researcher gave a rule of thumb, here it is:

1. "The final decision rule is simple indeed: as a rule of thumb, if your bad death percentile was below average (meaning you had more worthless deaths than players at your rank typically do in games), then you likely lost." [1]
2. "If, on the other hand, you didn’t die that much, then, if you had more smart kills than your peers, you likely won." [1]
3. "Finally, if you had both above average worthless deaths and below average smart kills, then whether your team won or not depends on how much influence you had over the map. If you had more map coverage then average, you probably won; otherwise, you probably lost." [1]

"Thus... don’t die for no reason; if you do, try for some smart kills; if you can’t do that either, at least roam and have decent map coverage." [1]

Thanks for reading. See you next time,

Christian 🙂

References

[1] Maymin, 2020