There has to be some physical presence of a footway on the ground for it to be marked in OSM, so no, a Strava heat trace on its own is not enough.
If you look closely at the Bing imagery (mine is dated 2014-2017), by my judgement there is what looks like a foot path matching the Strava heat trace along some sections.
Of course, local knowledge or a ground survey trumps any of this, so if you have better data don't hesitate to remove this way if it really doesn't exist.

@kylebarron
This trail was sourced from the high definition Strava heatmaps, when they were available. Not sure I can access them any more, but at the time I guess that there was a heavy and consistent enough foot traffic along this route to warrant adding a path. There is also evidence of some kind of trail on the Bing imagery.
It's not an official trail and is not marked as such.

Hey Michael

Sorry, not 50%. The boundary of forest in NLCD is defined as 'areas dominated by trees generally greater than 5 meters tall, and greater than 20% of total vegetation cover'.

Cheers

Tom

Hey Michael

The NLCD2011 imagery is a set of geotiffs published by the USGS here - https://viewer.nationalmap.gov/basic/
These show basic land coverage, including tree cover. The boundary between wood and scrub is, approximately 50% tree canopy cover.
I haven't completed this area yet, so it will be updated in the next few days.
There are discrepancies between the NLCD 2011 imagery and the Bing imagery - which is dated 2014 for this area - due to felling and wildfires between those dates - and due to errors in the NLCD data.

Cheers

Tom

Not sure, looks like someone did a preliminary attempt to map Macmillan way by adding the name to footpaths and tracks used by the way.
I'm editing the Macmillan way relation, using other people's gpx tracks, but not touching the names - this will need a ground survey.