The penultimate prep

[Caradoc]

Checking through the archives and noticed some interesting data.

I need to re-confirm the math and interpretations but it appears 69 horses ran either tops or pairs in the Derby between 1997 and 2010. Regarding those 69, some conclusions are intuitive, such that as a contestant\'s pattern is very strong if it ran only pairs and/or tops in its final two preps for the Derby.  Of the 69, 51 had run only P\'s or T\'s.  On the other hand, only 3 of the 69 ran well in the Derby (a P or a T) with running at least one P or T in its final two preps, also intuitive because it is very difficult for a horse to get back to its top in the Derby if it has been struggling to do so in preps.

What was less expected was the importance of the penultimate prep.  If my math is right, 59 of the 69 (86%) ran either a P or a T in the race before their final prep. (58 of the 69 ran a P or a T in their final prep.)  Running well in the penultimate prep seems to be at least as important as running well in the final prep.  Any theories as to why that might be, assuming the math and interpretations are correct?