One of the longest standing debates in sports today is Training vs Talent or Nature vs Nurture. People always wonder whether they can compete at the highest level if they put in endless hours of practice or if they are destined to fail by genetic predisposition.
The amount of scientific literature surrounding this debate is tremendous, and the issue has gained even more popularity with the publishing of Malcolm Gladwell’s Outliers, which stresses that expertise in any field follows the 10,000 Hour Rule. This book takes a big picture view of the issue and grossly oversimplifies it, so let’s dive into some of the details and how the specifics apply to eSports.
Training vs. Talent
Let’s examine the truth of this by looking at a detailed article that compares training vs. talent in running and sprinting. A lot of the content is contributed by Dr. Michael Joyner, a colleague and former mentor, and he uses some powerful graphs to illustrate the effect of innate talent on the ability to perform.
It’s shocking to think that innate talent is so important, but running is about as mechanical as sports get. Sprinting is simply a measure of how fast you can get from point A to point B. In fact, it is used as a physical attribute that defines cutoff parameters for drafting of professional players of other sports.
Other sports continually add more complexity to the quantification of a player’s ability. Your ability to run fast is important in basketball, but there are a lot of other aspects that are also equally important. Experts debate attributes like sheer athleticism and basketball intelligence, and which one is more important for a player to have. The greatest players like Jordan and Kobe possess both, of course, attained through a tremendous amount of dedication and hard work.
In addition to technique, game intelligence and other sport-specific attributes that make quantification of talent more difficult, the team aspect adds another layer of complexity. How well players work within the structure of a specific institution, the coaching, and management determine how well a team functions cohesively. Analyzing the ability of a team to act as a single entity is valuable, but incredibly complex due to the multi-factorial nature of the problem.
Physiological testing and player ability quantification in traditional athletics is very well established.
What Happens When We Apply Similar Concepts To The Gaming Realm?
Let’s try and start with a very basic game that can compare with the pure mechanical demands of running. This game would strictly require hand-eye coordination, speed, reaction time, and reflexes. Since games are inherently more complex systems than that, finding such a game is difficult. Even on-rail shooters have subtle nuances and strategy requirements in addition to mechanical skills, although they may rely heavily on these mechanical aspects. The only thing I can think of are rhythm games, like OSU, but I don’t have much experience with them, so I am open to suggestions from the readers.
As the complexity and skillset required to play different games changes, we can see how difficult it can become to quantify a player’s raw ability. Some games are more focused on mechanical skills (Quake 3), while others require considerably more strategy (Dawn of War), and still others require a heavy dose of both (Starcraft, Streetfighter). Team games like League of Legends and Counter Strike attain the final layer of complex talent quantification.
Why is quantification of player ability so important? Sports science and analysis is a huge business, and rightfully so, if players are being paid through 100-million dollar contracts. Players in traditional sports are drafted by their physical quantification metrics, the performance statistics and the intuitive assessments of scouts and recruiters.
Moneyball, worth two hours of your time to watch, gives you a good idea of how scouting and recruiting has changed in baseball with the advent of computer analytics. As eSports continues to grow, you can be sure that teams will use every tool available to them to recruit top talent. Korea is way ahead of the US in this department, eyeing talent from a young age to begin the training/recruitment process.
Currently, eSports has no physical quantification metrics to evaluate talent, and league structure outside of the professional arena isn’t well established. This means it is difficult to extrapolate results and predict performance as we do in as in traditional sports, which have well-structured leagues for varying talent groups and ages. In LoL, all we have is our intuitive evaluation, their ladder performance record, and maybe a few amateur tournaments – so taking a gamble on solo queue heroes doesn’t always work out (Seraph).
How do we quantify recent TSM pick-up, Lustboy’s, player ability under a new team infrastructure?
How Do We Quantify Players?
The present article sets the stage to understand scientific talent quantification in the world of eSports and LoL. Science brings in a certain amount of cold, hard facts to supplement the vast amount of game experience and intuition already being used at the professional level to make decisions.
What metrics can we quantify to assist us with player assessment? How can these metrics help us target weaknesses and improve performance? Finally, we will try to answer the initial question posed by this article, “What separates the Pros from the Joes?”
The video serves as a good introduction to the work to be described in this series of articles. The follow-up articles to be published here will dive into detailed methods and results of each specific topic in the video.
- The first article will tackle mechanical skills: hand eye coordination, dexterity, reaction time, and accuracy.
- The second article will take a look at the information processing, attention, and focus of players.
- The third topic will investigate how players handle stress in game, and how much of an effect that has on performance. We all know how stressful life in the majors can be, but studying how players react to stressors in game can bring a wealth of information on performance optimization.
I hope you are as excited by these articles as I am. I wrote a few articles for Cloth5 earlier in the year, but my work calendar proved too difficult to be consistent, especially when I could no longer play League. I would like to make a more sincere effort this time and provide more consistent and scientific eSports content. To do this, I am starting a Twitter feed, and hopefully you will join me to participate in the discussion. So what do you want to know about the science of gaming?
Special thanks to Team Curse (Congratulations on a great Super Week!), and particularly their manager, Liquid Steve, for being gracious and forward thinking. They were also willing to put up with constant prodding and probing for the collection of data, no easy feat. We also appreciate the USC computer science group and their willingness to help gather the data.