The qualifiers for ERChamp 2023 are finished, and the top ten teams were selected to compete in the finals in Wroclaw. How exciting! At the same time, I’m thrilled to get my hands on the log data from the event to put together some of the stories from the online portion of the event.
This year I already put together one article with findings from the ERChamp stress test data. It was published on Escape.Buzz, the escape room news site managed by the same folks behind ERChamp, LockMe, the Escape Tales game series, and more. Here is a link to that article.
As I start to compile the next article for Escape.Buzz, one featuring details from this year’s eliminations round, I get very excited about the opportunity to blend my two passions together – analytics and puzzles. ERChamp has an unheard of opportunity to collect tremendous amounts of data from over five hundred individuals solving a puzzle simultaneously. Frankly, I’m surprised no university, or even an aspiring Statistics or Psychology PhD, has reached out to the ERChamp team to try reviewing what these games have to show about the way smart people think.
As for me, I’m a sucker for customer experience analytics, so I’m delighted by anything that can shed light on what people like, dislike, and anywhere in between. Which puzzles stuck people the longest? Did we have any common incorrect answers that we should have provided more clues around? How does team size or country of origin impact a player’s experience? Did any teams find unorthodox approaches to the correct solution?
To potentially set aside some fears, I want to also clarify a few details on how I use the data. During analysis, the only time we ever map any data back to the team name is for the bar chart race diagram showing the top teams. Nothing else is ever mapped back to the individual team, nor is anything ever mapped to individual participants (that mapping is not even contained within our dataset). Even if you wrote a particularly hilarious entry into one of the code entry fields, that secret is safe.
The end result of a good analytics approach is that every year should be better than the last. We learn a little about ourselves and our players and design the next iteration accordingly. I love this stuff. Though I spend most of my day designing novel approaches at customer experience challenges using analytics, I’m thrilled to have such a cool opportunity with puzzles. If I go back for my PhD, you can bet on what my dissertation will be about.
Will we ever get to the point where we use our data to make an AI-driven ERChamp game that tailors itself to every player? Probably not. It wouldn’t be fair from a competitive angle anyway. But there’s a lot of fun stuff in the meantime. The bar chart race is just one cool way to reward teams that were in contention for the prize but didn’t quite make it- it’s fun to see your team name briefly jump into the top spot even if you didn’t maintain the position. We can also work toward giving tailored insights- for example, today you might know your team took 56th place, but maybe you don’t know that there were actually two puzzles in the game you had the fastest solve time for.
I hope you enjoy some of the insights that come from the ERChamp data, and I look forward to seeing how the data guides us to make better and better experiences.
Leave a Reply