actually, now that I think about it: do you think January 2013 Wii U numbers were depressed in any way compared to what it should have been?
I honestly don't feel the January-->February bump it got last year was really legit
Most unsold systems initially bought in November or December by scalpers were most likely returned in January when the demand plummeted.
NPD records a sale and records a return in the months they occurred.
So:
Console bought by a scalper looking to make a quick buck in December from Wal-Mart
NPD records it as +1 sales in December NPD results, adding to December's total
Console returned by scalper in January
NPD records it as -1 sales in January NPD results, subtracting from January's total
NPD is trying to be a benchmark for potential install-base. The system "had the potential to contribute to" the install-base from the time it was purchased up until the time it was returned, so it's relevant to include it on December's figures but subtract it from January's figures.
So yes, this phenomenon dragged down Wii U sales to 57K when they'd normally be higher. But I'm not sure how prominent this phenomenon is to the extent where Wii U sales would be significantly higher in January than if the rate of console return was per the norm.
That is, I do think that the existence of a January -> February bump last year was legitimate, but it was probably smaller than what the raw NPD figures (56.9K -> 66.0K) represented.