Sorry I don't have the data. Just the trends...If you want some hard numbers it would be better to agree to disagree. I don't think any numbers would make you happy. Not only can I not give them, I also admitted that if I had the raw numbers I wouldn't give them to you. It's not only bad for me but raw numbers can be easily manipulated by misusing stats in the same way you and I disagree on how to interpret how changing a physical construct of a trigger can make something better or worse.
To be fair to departments that not only will not keep the stats but they usually keep just enough to see the trend and move on. Why? Because our disagreement that we have here isn't an interwebs disagreement. That's just something silly that people say on the internet. Our disagreement is the same argument that has to be won by officers and trainers during every lawsuit around the country. Ultimately the numbers then becomes an interpretation. There are plenty of people who interpret them incorrectly. If you're an activist or lawyer there are plenty of classes at university that will teach how to lie using them. They make them show something that logically they can't show. Ultimately what we've been talking about is the same thing. Break it down to the geek level, think about it, make your choice and move on. As I think Tom Givens said (or close to it), "You make your choices and take your chances". I add, "Choose wisely." For instructors I add, "Make sure you're correct because you're not screwing yourself up. You're screwing the very people that need you the most, the below average."