In answer to your question in the last paragraph: Supporting data in most cases would be a positive analysis for the original contents of the ruptured aerosol can at the remote fire location.

The Factory Mutual experiments concentrate on the wrong end of the skunk. Rocketing aerosol cans are inherently interesting but they tend to fly in the opposite direction from the expelled accelerant. When a can blows its bottom, it accelerates away from its liquid content and usually (but not always) becomes an empty vessel before it has moved a foot. Thus the can becomes a relatively poor fire starter compared with the combustible ejecta.

The liquid from a blown-bottom can moves in the opposite direction from the can at a somewhat lower velocity and behaves much like pellets from a shotgun accompanied by a finer mist. In the experiment by Dr. Fox, liquid from the wasp spray can first slams into a board, then bounces 180 degrees back across a 5-foot gap and deposits a large splash on the target blanket – all in the blink of an eye.

If a can is punctured or suffers a broken seam below the liquid level, the liquid will be ejected from the hole by the vapor pressure just as water is ejected from a garden hose or a kid’s squirt gun, without necessarily moving the can at all. The trajectory of the high-velocity stream can carry the liquid to a remote target without depositing any significant quantity in the intervening space. The behavior of liquids expelled under pressure from an orifice has been well understood since the time of Bernoulli.

By far the most important bottom line in this discussion is that ruptured class 2 or class 3 aerosol cans present at a fire scene cannot be eliminated as the source of apparent multiple origins in the absence of additional supporting evidence.