Eclipse turning blue?

Mid-November, ingress nominally 2/3 over, and the snows are arriving in the mountains.  We made one last attempt to reach Mt.Evans observatory on 14 Nov but huge drifts defeated the stout Unimog vehicles we used.  However, intrepid observers elsewhere are still getting peeks at the star, sometimes between clouds.  Some interesting trends are emerging since mid-October: the B filter photometry shows a decreased rate of decline, as does V filter (less so), but R and I filters appear to continue the decline in a linear fashion.  This means the star is relatively brighter in bluer wavelengths compared to redder ones - suggesting a hotter source - perhaps the F star pulsations continuing (like those seen pre-eclipse) and/or some of the supposed hot B star light is leaking out of the dark disk.  Richard Miles reports an ~0.1 mag jump in V-Ic after mid-Oct.
Equally interesting is a dispersion among observers in V filter brightness - not much disagreement (a few hundredths of a magnitude), but enough to get everyone checking their comparison stars, spectral response of their filters, (seasonal) changes in atmospheric extinction, even software updates all can affect the derived magnitudes.
Spectroscopically, H-alpha profiles continue to grow more absorption, while neutral potassium appears to have stalled.  Both of these trends mimic last eclipse cycle reasonably well, although the UBVRI light curve differences suggest there is no guarrantee that things will proceed exactly in the same manner.  In any event, as winter approaches, it's understandable that observers and epsilon Aurigae are both turning "a little blue"...