I applaud your idea of testing and thank you for sharing. However, I must say that your test only produced anecdotal evidence which is not sufficient to draw a valid conclusion. I’m not saying your conclusion is wrong, just that your data was not ample enough to draw a valid conclusion.
You did one test with a sample and a control. What happens when you repeat the test? Do you get the opposite result half the time? Based on your single test subject you do not have enough data to draw any valid conclusion. How were you able to balance the difficulty in ranking between the test and the control? How where you able to isolate rankings due to QDF versus normal ranking?
There is so much more data needed to draw the type of conclusion that you made. I challenge you to continue with your experiment to see if you can repeat the results and how consistent those results will be with varying ranking difficulties. You are off to a good start, but you stopped too soon and drew conclusions based on anecdotal evidence.
Another option would be to simply conduct a study. Use ample data that already exists. You don’t always need to build websites and promote them yourself. You may be able to simply data mine from a database of SERPs and group web pages the emulate your test parameters to see how often and consistent your hypothesis is repeated.