or rather, some very rough calculations.
lets assume that when the car is sliding perpendicularly, the sharkfin has a Cd of 1.3 (a number I got from wikipedia for a simple flat plate perpendicular to airflow) and has an area of .827 m^2 (2325 * 355.6 mm, which i roughly got from the spotting guide and the total length of the car) and the center of pressure is halfway up the fin and dead center. we'll take the CoG height to be 6 inches above the ground, with the fin center of pressure 880 mm above the ground. Finally, lets assume that the car lifted off at about 90 km/h.
drag force = .5 * v^2 * air density * Cd * A
with SI units, Fd = .5 * 25^2 * 1.225 * 1.3 * .827
which in our case works out to 411.6 N.
Lets assume half the weight is on each side of the car, and assume Fg on that part to be 9.8 * 450 = 4410 N. We'll neglect downforce as the car was travelling sideways. The question then is can the fin generate enough moment to overcome the gravitational loading
the length of the overturning moment arm is .677 m (880mm - 6 inches) to the overturning moment .728 * 411.6 = 300 Nm
if we assume the track to be 2 meters minus half the tyre width (14 inches), the countering moment from gravity works out to be 3626 Nm.
Yes, its very rough, but its an order of magnitude difference. While it may have contributed, the sharkfin does not appear to have been enough to cause the flip over. I tried to use overoptimistic estimates, erring towards things that would help the sharkfin lift the car off the ground. I have neglected the lift effect from the floor. Thats fine; I'm only trying to determine of the shark fin was the culprit. That does not appear to be the case.
Forgive my odd mix of metric and imperial units. Thats what happens when you live in a country that uses SI but sells 60% of your exports to americans.