They could have run control experiments where they didn't actually start the reaction and just looked at how much background polymer to expect from contaminatio

They could have run control experiments where they didn't actually start the reaction and just looked at how much background polymer to expect from contamination. Of course, if there's a lot it would drown out any signal they got in the actual experiments, but if you get an order of magnitude more polymer then you can be confident you're not seeing an artifact.