@karsten -- of course, in an ideal world, the numbers that we report for footstrike type would match regardless of mounting location. Unfortunately, the reality of this particular metric, is that it isn't something that we can 'directly' measure. Even if we were able to reliably (100% of the time) capture the absolute angle at FS, we still wouldn't know the specifics of the terrain for that FS. Instead, our algorithm uses the Stance Excursion, as I mentioned in your prior post) to estimate what type of FS has occurred. We built the algorithm based on data collection across a number of runners, capturing data (as you have done) simultaneously at the heel and laces locations. Unfortunately, as @Alang points out, subtle things like shoe flex can alter the amount of angular (StanceExcursion/pitch) change that the heel sensor sees compared to what the laces sensor sees, resulting in differences based on the mounting location, which result in different estimates of FS Type.
I absolutely take your point, that what you're after is an absolute measure which is 100% in agreement between the two locations. It's unfortunately not something that's directly measurable, which means that absolute accuracy across all FS types in all shoes is challenging. This is different than things that we can directly measure, such as ContactTime (which match very well between mounting locations in your example).
We'll continue to refine/improve our algorithms (FS type is definitely on the list).