I'd take a look at past performance on Atweak -- run it through a couple of earlier seasons -- it may be a formula that matures mid-season. I've got a couple of college formulas that do that, plod along and then pick up around week six/seven. If it has two winning weeks in a row, I'd keep it out of my go-against consensus at least till it starts losing again.
Back in the days when there was a lot of debate and discussion about formulae on the microbros board, before the pinheads took it over, there was a lot of discussion about how you wanted strong past performance in a formula, and I think that's a lot of crap.
I want a formula that is hot RIGHT NOW, not one that generated +x units per year over the last four seasons.
I'm in the process of monitoring my formula constantly -- a couple of bad weeks and I'll drop a formula from my consensus, three good weeks and I'll add a formula. I think you have to be fluid with it.