What has actually been tested?
Penn State looked at adult female deer around winter storms and did not find a consistent movement increase in the preceding 24 hours. That is useful research, but it is a different question from bucks moving a couple of hours before a fall front. [1]
I would not throw out a hunting observation because one study asked a broader question. I also would not turn one busy afternoon into a rule. What I want to do is write the timing down and see whether the same thing keeps happening, including the fronts when nothing shows up.
Get the front timing right
A cold front brings advancing colder air. A wind shift, temperature change, and pressure trend can help us recognize its passage. Falling pressure by itself is not enough to say exactly when a front arrives. That matters if we are trying to look at a two-hour window. [2]
If two forecasts disagree about the arrival time, I want HNTED to show that. A front expected at three in the afternoon is a different planning question from one expected after dark. We should keep the forecast we saw beforehand and compare it with the conditions that actually arrived.
What I would compare
I would start by comparing separate camera visits in the three hours before passage, then the three-to-six-hour and six-to-twelve-hour periods before it. Those are test windows I would choose, not proven deer rules. I would compare similar times of day and parts of the season.
The camera's capture time is important here. A picture delivered late to the app could put the deer in the wrong weather window. Keep camera downtime and quiet periods in the record as well. If the same pattern holds across several fronts, then we have something more useful than a memory of one really good sit.