I worked on a project where there was a ‘bug bounty’ with prizes for fixing the most bugs in a month. The developer that won in the first month fixed an incredible number of bugs compared to all other developers and won an iPod. How did he fix more than everyone else? He searched far back in the backlog and picked up and fixed all the simplest/easiest/quickest to fix bugs. A lot of those issues resolved were trivial ‘nice to haves’ that didn’t add any substantial value.
Be careful what you measure because by incentivizing something that’s what you’ll end up with, whether that’s really what you we aiming to achieve or not.