The trouble with raw statcounter hit data is that it's hard to know what it's tracking. Even statcounter itself cautions that the stats should not be considered

The trouble with raw statcounter hit data is that it's hard to know what it's tracking. Even statcounter itself cautions that the stats should not be considered representative samples. You'd need some in-depth analysis to distinguish any real effect from, for example, those that are an artefact of the sites that use it (and don't use it).