You Need More than Statistics

10 Dec 2021
4 minute read

There are many apps and sites out there that track statistics for you: damage per minute, farm per minute, deaths, kills, etc. Like any other sport, it’s possible to generate all kinds of statistics, and most of it is crap.

I don’t want to put down stats-based sites. There’s a lot of them out there, and I make use of them myself. In particular, I find Porofessor useful to check opponents pre-game, op.gg to track my match history (especially since Riot has stopped doing that themselves), and LoLAlytics to look up popular build paths.

But I think people should beware of looking for the magic solution. I’ve mentioned this before. In my very first post, I said:

Reiterate is not a magical app that can guarantee you a spot at the top of the ranked ladder

And then in my History of Self-Help I expanded on this. No single self-help book is going to fix all the problems in your life. But that doesn’t make them useless. You can pick up useful tips from even a bad book, and if you study and work on it, you can synthesize a custom program that will work for you, combining techniques from a variety of sources.

Improving at a game like League is similar. Any one site, no matter how jam-packed with statistics, will not help you improve your game by itself. But if you use the stats they provide, you might find them helpful.

How do you best make use of game statistics? That can be tricky. I think there’s a problem with game stat sites, in that the statistics can be of low quality. Take for example, damage-per-minute. It’s too easy to look at a stat like that, and perhaps see that your personal DPM falls in the lower 20% of players, and take it to mean that you should teamfight more.

But that can be the wrong lesson. Context matters greatly. Was your DPM low because you were playing a scaling champ and your team didn’t get to late game? Or perhaps there was a smurf on the other team who enjoyed making it so you couldn’t play at all. Maybe you were focused on taking objectives instead of team fighting. In my own games, I’ve seen my DPM vary from bottom 10% to top 10%, and that’s not because I’m inconsistent, but because the individual games will vary that much. So I think a lot of people place too much value in a stat like that, while its variability demonstrates otherwise.

There was an interesting thread on Reddit recently where many commenters noted similar disappointment with stat-based web sites. One commenter noted that analysis is always hindsight-based. A statistic can only tell you how your game was, not how you should play in the future. You can find out that you did a lot of damage, or very little damage, but that won’t help you play better in the future, unless you evaluate that performance in its proper context. Once you understand why your damage was high or low, you can focus on reinforcing the behaviors that increased your damage (and stop behaviors that hurt your damage).

I made Reiterate in part as a response to stats-based improvement sites. Reiterate helps you to fix your behavior, which is something a stats site will never be able to tell you. Statistics can only provide clues; you must dive deeper to divine the true reasons for your performance.

Tagged with

Comments and Webmentions


You can respond to this post using Webmentions. If you published a response to this elsewhere,

This post is licensed under CC BY 4.0