You have to look at the percentages before you decide if it was a smart or dumb move.
I couldn't find an analysis of college on-side kicks, but there have been studies on NFL on-side kicks and here are the results: 20% of EXPECTED on-side kicks work. 60% of UNEXPECTED on-side kicks work. Despite what the announcers were saying about a possible on-sides kick, BYU was not expecting an on-side kick based on their defensive positioning.
So, based on the data, UConn had a 60% chance of recovering the on-side kick and based on the play execution, UConn probably should have recovered the ball.
Even though UConn didn't recover the ball, the defense would have forced a 3 and out had Adams not kicked the football into the stands. In my opinion, the on-side kick had a better than 50/50 shot and the defense did do their job. I think it was a decent calculated risk.
This is a much better team than last year and Diaco seems to be turning the program around. This was expected to be a rebuilding season with 2016 probably showing significant improvement. Enjoy the ride.