Good question: What sort of 'average' is used?
Is it the arithmetic mean, the median, the mode, the harmonic mean, the geometric mean? I guess, if they are looking at stars which are all the same sort of distance away and 'averaging' their recession speeds, then the arithmetic mean would be appropriate as long as the spread is not too great with respect to any non-linearity that may arise when the distances are great. Average density would possibly need to include some consideration of time of observation as a consequence of distance (because of finite light speed and expansion).
A lot of statements made abut the Universe seem to imply that an observer is outside looking in and that everything is happening at the same time. Whereas, in fact, the impact (strength and timing) of what happens in one place is affected by the distance away of the phenomenon 'when' it happened.
Verne; the idea of stretching avoids the problem of exceeding c, doesn't it? Metres per second is still the same -it's just that the metres may have changed.??