I mean yes and also no
Mathematically speaking your analysis is sound, but your logical approach to how to analyse the problem is flawed. Granted, 6 officer deaths may equate to one civilian death in typical circumstances, but this number is irrelevant unless you use it in conjunction with the other relevant statistics.
For instance, it's important to consider the ratio of officer to civilian deaths, and when civilian deaths do occur, the ratio of confiscated guns to guns picked up and returned to the owner. With the ratio of officer deaths : gun confiscations, you could make a way more accurate ratio of cop time wasted to civilian time wasted per shootout, which would be way a way more relevant and accurate statistic to discuss.
Another issue with this is a lot of the guns you take averages from (the "middle class of guns, so to speak,) are rarely used. Stuff like the benelli, mossberg and most SMGs don't see much use. It's usually only rifles and pistols (with the occasional sniper and shotgun.) I'll be using the HK45CT as an example because it's commonly used. Using your calculations, a civilian dying with this would waste about 12.6 minutes, compared to the officers 5 minute death time.
Honestly I'd just use a range of values instead of an average of all of them because of the large range of values.
You've also failed to take into account that if all the cops die they lose out on confiscation money, something that may be significant here, even using your model. Cops get about $500 for each confiscation, so if they die and miss out on one that's 0.7 minutes per rifle per officer, again using your baseline for money earned per hour (keep in mind for an officer main this baseline would be lower as they typically earn less than growing drugs, so relatively speaking this number would be higher.)
And of course, it's not nice to lose. Regardless of how much you lose or how much you don't lose, both sides want to win and there's a side to this argument that can't be quantified by statistics.