Missed it live but pulled it up on the recording - that's no camera artefact, the movement pattern is all wrong for compression ghosting or a moth flying past the IR lens.
Been running a Reolink 810A on my barn out here on Dartmoor for two years now, so I've seen every flavour of false positive you can imagine. Lens flare, IR bounce off fog, the lot. What's on that Myrtles footage doesn't tick any of those boxes for me.
What timestamp exactly? The 2am window is interesting because that property sits on what some researchers believe is a convergence point - and we know the Myrtles already has documented historical trauma going back to the Chloe legend. That kind of residual energy doesn't just switch off.
Couple of things I'd want to know:
Was there any precipitation or humidity spike logged that night? (Fog can do weird things to IR), Has anyone run the audio through Audacity and checked for EVP in the sub-20Hz range around that timestamp?, Did the figure cast any displacement shadow on the column behind it?
If nobody's done a spirit box cross-reference session timed to that location yet, someone should - my Sangean ATS-909X would be overkill for this but the methodology still stands.
Dartmoor has taught me that the most credible captures are usually the ones that don't look dramatic. Quietly sliding past a doorframe at 2am? That's the stuff that actually keeps me up.
Post the raw file if you've got it, not just the clip - metadata tells you more than the image half the time.