A few people in this thread are using this case to justify a policy of "I defend me and mine only". Now, there's people who take that tack, and if that's your position, that's fine, you don't have to justify it with this or any other case.
The lesson *I* take from it is "don't use a gun to solve the problem unless the alternatives are FAR, far worse!"
In other words, I won't defend a business' money or property. But if I think a robber is about to kill somebody, yes, I absolutely will act. Been there, done that, the bad guys backed down because they saw I wasn't bluffing.
In this case the cop had pretty good evidence that murder was not foremost on the goblin's agenda. The goblin and at least one employee had been in the back room together, no shot fired, goblin was now leaving. That's a very strong sign that gunfire wasn't needed. The cop should have faded from view so that he COULD (at the cop's option) gone into "be a good witness mode". And then if necessary, ONLY if necessary, switched to "sniper in a covered position mode", which is a really good place to be at the beginning of a gunfight.
Instead, by standing out in the open with a drawn gun, the cop caused a gunfight to go down in a bystander-dense environment - PLUS the cop's got no cover or at least concealment. That's doubly-bad-tactics right there.
Me, I'd rather be proned out behind a really sturdy garbage-box-that-holds-a-can thingie like most fast food places use, looking like somebody who's avoiding trouble, gun NOT visible, peeking out with one eye. Odds are I won't have to shoot, that's just how the statistics break down - most armed robberies don't involve a murder.
Fun fact: one of the biggest predictors of a successful outcome in a police gunfight is whether or not the cop(s) make it behind hard cover - at least something like a car engine block. When they do, in one piece and still armed, the successful conclusion rate goes through the roof, one number I saw put it at 95% success rate where "success" is measured as "won the fight and not hurt too bad".
In most of the fast food places I've ever been in, if you have a few seconds you can set that up in a case like this. If you're not looking for and using cover the *instant* there's even a possibility of a gunfight, you're Doing It Wrong[tm].
True story:
I once had to call 911 because a bum was freaking out late at night in the area in back of some small office buildings right next door to where I live. He was banging on the windows and generally going nuts. I called 911 where the guy couldn't hear me, told 'em what was up, then walked around to the front of the office complex to guide 'em in. This was in AZ. I was strapped, didn't let the cops know it. So they wait until they have three cops, one stays near the front with me, the other two draw guns and go straight in - the guy's more or less in a dead end.
I stepped a few feet to the side so that I was just around a brick building corner from the cops with guns out. The one who was left behind with me asked me not to leave, I explained that I was just getting out of any possible line of fire and wasn't going further. The cop just looked annoyed. To him, cover was no big deal. Yeah...I don't think so buddy
.
(Please note: I absolutely did NOT reach for my own hardware. Ain't. NO. Way...not in the presence of a twitchy cop. If somebody made it through the first two and was still a threat, I'd have beat feet first and let that no-cover moron try and slow whatever it was down. In the end they carted off a mostly-harmless meth head...)
Cover and even the remotest possible possibility of gunfire need to fit together in your head, always, so that in an emergency they'll flow together instinctively. Part of your "condition orange" drill should be the question "where's the nearest cover?"
Fun drill: when you're walking down the street, make yourself do a fast-as-you-can scan for available cover whenever a yellow car/truck/whatever is first in sight. Or a bus. Whatever. Point is, an unpredictable trigger. You get good at figuring out where the cover is in a hurry.
Remember kids: we can
defend cover if we're attacked.
And defended cover is a total nightmare for an attacker. It's one of the few factors that can give a handgunner an even footing or better against a rifleman. Every time a small number of soldiers have done well against a larger force, cover was part of the deal - the Jews of the Warsaw Ghetto racked up way more than their own number in dead Nazis, and Cochise with 34 guys
successfully held off 5,000 US cavalry by sniping from behind rocks way up in the hills. These were not flukes.
In the case this thread is about, it appears the cop doesn't realize even after the fact that cover on his part
might have also concealed the gun long enough to prevent a gunfight from breaking out if it wasn't necessary.