Originally Posted by
gauvins
The problem is that both halves share the same sprockets. If one section elongates faster, it'll wear the sprockets to match it's pitch. The shorter chain half's links will then be submitted to proportionally greater force, which will accelerate wear.
That would reduce the observed difference in rate of wear between the two half-chains, but shouldn't reverse the observed effect. When the experiment was done the cleaned half actually wore faster - so if anything it would be the part of the chain that was wearing the sprockets more.