Obviously if they optimized each and every one of these benchmarks we would see different results, but it would take a massive amount of time to learn the ins and outs of each framework to the point where you can do so effectively.
For one person do a benchmark over this many samples, they have to just go with the out of the box setup for each.
As our blog post suggests, where we are not experts we had to rely on the tutorials provided by each framework's authors to build a test setup. If a specific framework seems low on the list, it could be due to the fact that the best practices guides we found for getting set up were not correctly configured for production use.
Draw what conclusions you would like from this statement, but we did aim to be as fair and unbiased as possible.
For one person do a benchmark over this many samples, they have to just go with the out of the box setup for each.