To me it looks like a test that is designed to avoid all the built in crash structure of a vehicle. The hit looks to be outside the frame rail and as a result parts never intended to absorb the impact are expected to do something they are not designed to do. To me it looks like a bogus test.
So how should the tests be designed? Should they cater to the design of the car or should they instead test the ability of the vehicle to resist a more likely crash scenario?
The small front offset test depicts the overwhelming majority of frontal impacts. After 30 years of full frontal impact testing, shouldn't we be focusing on the weak points again? I work in the testing industry on the aerospace side. I've seen the occasionally disastrous effects of the results of a test being ignored because the product didn't perform like it was supposed to. Instead of improving the product, the test is altered to produce more favorable results (on paper). That's not how the real world works, though.
In any case, you can't just throw a test procedure together and expect it to mean ANYTHING. A huge amount of work goes in to developing a standard test procedure, and the results should not be taken lightly, even if it paints one of our beloved (insert your beloved object here) in a poor way.