Excellent point. I think there are a lot of characteristics of those totals that are misleading and I don't think one can derive (even on average) that "if only I had transacted 3 times instead of 0 I would have done better". However, it doesn't mean it isn't interesting.
good point. It seems the only relevant metric would be something to the effect of:
Did the trades help the trader?
In other words--take the ratio of the current value of traded-for stocks vs. the theoretical value of original stocks. If this is greater than 1, then trading has helped.