I'm talkin' 'bout practice

[quarrel]

There’s a game designer I know of who is a vocal proponent of competition as a means to determine true, objective merit. He’s a huge fan of games like chess or poker that let everyone compete on an equal footing, where how well you perform depends solely on how hard you work at it and how much skill you develop, and he has little respect for games like basketball where some players possess material advantages that give them an undeniable yet inescapable edge over opponents who would otherwise be equally skilled.

One day, this designer did something I didn’t expect. He blogged enthusiastically about a psychological study that showed practice is more beneficial to people who have more natural talent to begin with. To be precise, the study showed that people who perform consistently poorly when trying a given activity for the first time don’t get as good at it in the long run as people who consistently performed well at it at the outset. It confirmed another of his philosophies, which is that people are better off playing to their strengths.

I pointed out this contradiction to him. How could he be such a fan of the idea that you deserve success if and only if you work hard at it, yet be happy that science confirms that someone with more inherent skill than you will always stay better — in fact, pull even further ahead — if you both practice? It kind of put the kibosh on his whole “ultimate success through self-improvement” philosophy.

His response was handwavy but fairly sound. He pointed out I was assuming everyone gains skill at a linear rate and takes the same amount of time to reach their personal ultimate mastery level. Thinking about it more, I came up with several additional complications on my own. Not only might the graph of a person’s skill over time easily be a curve instead of a line (in fact, it probably is), it might even dip in places rather than always rise. The worse early performer might learn swiftly at the beginning and slower later, while the better one does vice-versa, which would flip the tables at least temporarily. There’s no guarantee that the better initial performer will practice as much as the other competitor, or even at all. It’s also possible that the person with the stronger initial performance didn’t achieve it due to having more innate ability but rather from having more past experience in similar activities, which means he’s actually already climbed part-way up his learning curve rather than having a higher origin for it, and so may not truly have the higher ultimate plateau.

All that said, none of this may matter. This isn’t the whole picture. I double-checked the study that started this whole discussion and discovered this designer never mentioned its most important finding. While the study did find that people who do consistently poorly at a new task usually only ever become so-so at it, and that people who do consistently well at a new task usually become quite good at it, it also discovered that people who display a wide range of early results — amazing “beginner’s luck” outcomes and abysmal failures — tend to do best of all in the long run.