That is rather puzzling. Given the wind direction yesterday, is it possible gusty outflow perhaps led to the strengthening of an inversion? Breaks of sun though would have certainly negated that though. Perhaps the difference had to do with the dewpoints? We were mid-70's down in CT which helped to create a great amount of low-level instability. One thing I just thought of but didn't look at yesterday was DCAPE. I've never thought about using this in these type of situations but I think it provides value. Here is DCAPE values at 16z yesterday. This is right around the time when winds started to ramp up in CT. One thing I've noticed over the years too is there is a disconnect between max DCAPE values and strongest winds with convection (differences in that you won't see max DCAPE coincide with strongest winds).
I also think that the system evolved a little erratically...perhaps as a result of a transition to a more ET nature. But looking at the course of the H85 low, it actually kinda of elongates a little bit and I guess you could say "weaken". At first I thought maybe this tracked a bit farther east in the last few hours but I think the system overall just became a bit more elongated/tilted. This was happening as the LLJ max moved into CT so we maximized it, but it also began to shove the LLJ max farther east...which I think explains why even PVD got into some solid gusts.
Here is 19z and you can see the MAX over CT
Here is 21z. Based on this I would think the LLJ max is shunted east rather then going north into MA/northern New England