Given the counterintuitive conclusions of the accuracy comparison, it behooved me to consider possible reasons why the results might not actually reflect reality. These include some accusations I know will be leveled by infuriated AR enthusiasts who can't stand the thought of their precious fighting carbine weapon system platform ever being less than absolutely perfect under all conditions. Please stop hyperventilating.

Here's why the Mini might have outshot the AR, ranked from most likely to least likely.

This is not a scientific study and proves nothing.

I mentioned this earlier in the article, but prior responses show that it could probably bear repetition. The small sample size makes for a low statistical confidence. (Although it should be noted that the confidence is not zero.) Shooting only five rounds per group doesn't help. My only goal with this test was to see if the Mini was so abysmally inaccurate compared to an AR that it was instantly obvious with only a handful of groups.

The AR didn't like the ammo.

Possibly. Different guns like different loads, which is why I tried two different bullet weights at different pressure levels. I'd be saying this was definitely the case if one type of ammo showed the AR grouping better than the Mini. However, the Mini shot better groups with both types of cartridge. I happened to pick the boxes of ammo that the Mini-14 shot well and the AR shot poorly, but like the lucky rifle argument above, what are the odds this happened to the first guy to actually compare the two?

The Mini-14 has all sorts of aftermarket enhancements.

As mentioned earlier in this article, the Mini has a trigger job, a giant chunk of aluminum clamped to the barrel, a reduced-size gas bushing, and a thinner front sight. (That last one is stock.) Maybe those make it more accurate than it otherwise would be. It should be noted, though, that when I first bought the Mini and it was still box-stock I managed to squeeze a 3 MOA offhand group out of it. That's what made me notice that it was actually pretty accurate, an observation that was borne out by subsequent informal observation. The addition of the Ultimak rail and gas bushing didn't seem to change its accuracy to any significant degree.

You suck at shooting.

My skill with a rifle could be better, I agree. That said, the groups work out to between 2 and 3.6 MOA, which is enough to clean a 200 yard High Power match target and, I'm told, not bad for irons. I've also had people tell me they can do better, which I believe despite the fact that none of them have yet to produce photos of their groups shot under similar circumstances. My skill with a rifle notwithstanding, the same guy shot both the Mini-14 and AR-15, so you'd think I would suck equally badly with both.

A variation of this argument would say that my shooting skill is so abominable that it obscures the difference between the rifles. However, principle of root-sum-of-squares additive error means that if I'm a bad shot and the difference between Mini and AR are measurable, then there is an even more significant difference in mechanical accuracy on the part of the rifles.

Take the closest groups between AR and Mini that use the same ammo, Set 2's 55-grain bullet groups. The Mini-14 shot a 1.4" group and the AR shot a 1.5" group. If my inaccuracy is masking the difference between the two rifles, this group would be the best example because the difference between the Mini and AR is the least, so the factor I introduced would be the dominant one. Using root-sum-of-squares approximation of additive sources of error and assuming the AR shoots MOA, that would make my hold alone responsibe for about a 1.25 MOA error. Plugging that back in to the Mini side of the equation, it would mean the Mini was shooting a little over 0.7 MOA.

The take-away here is that human inaccuracy actually obscures rifle accuracy differences. If I'm a bad shot, then the accurracy differences between the AR-15 and Mini-14 needed to show up on the targets would have to be dramatic indeed.

You're better at shooting the Mini.

Unlikely. My AR has twice the round count of the Mini and I find it much more comfortable to hold and shoot. It also permitted locking in more solidly to the bags; I couldn't shove the Mini-14 hard against the front bag because it would shove its rock-and-lock magazine back and risk misfeeds.

The AR is homebuilt/not mil-spec/cheap and that's why it didn't do well.

While a more expensive AR would probably have turned in tighter groups, this argument is a form of the "no true Scotsman" fallacy: This AR isn't accurate? Then it's not a good AR! The AR used for testing cost about $200 more than the Mini in its stock configuration, raising the question of how much more one would have to pay before greater accuracy was assured.

You didn't clean the AR correctly.

Both the Mini-14 and AR-15 had been freshly cleaned prior to shooting.

You intentionally falsified the results to make the Mini-14 look good.

First, I have no dog in this fight. I own both a Mini-14 and AR-15 and like both. I actually prefer the AR-15 in some regards due to how it handles and where its point of balance is. I fully expected the AR to be more accurate. There are far more profitable uses of one's time than lying on the internet to get people riled up.