Those two setups are very close – 130 DPS difference, with a +/- 400 DPS margin of error on each run. Because the difference is smaller than the margin of error, you cannot say with confidence that one is better than the other. If you were to repeat the experiment, there is a high probability that the results would flip.