How hydration % affects stretchability
I had been wondering precisely how dough hydration affects a dough's ability to stretch to form a membrane, so this morning I did a quasi-scientific experiment to get a rough idea of how this works.
I made five batches of dough, at different hydration levels, using the following recipe:
525g bread flour
The hydration levels were 50%, 57%, 64% and 71%
The sugar, salt and water were added to a 6qt KA mixer bowl, and the flour and yeast were floated on top. The mixture was then kneaded with a spiral dough hook for 15 minutes on setting 2, and then a 200g ball from each batch was set to rise for two hours in an air conditioned environment (~74 deg).
It should be noted that the spiral dough hook wasn't able to form a proper ball with the 64% and 71% batches, so I imagine that the kneading that they received wasn't quite as rigorous as the others.
After rising, the dough was given a french fold, and left to rest for 5 mins, after which, three 10g balls were plucked from each 200g ball. These 10g balls were then rolled out in to as thin a membrane as I could manage using a rolling pin and dusting flour. The 64% and especially the 71% doughs were tricky to handle.
The rough average area that I managed to produce from each batch is as follows:
To make the most of the experiment, I also made a 57% batch where I substituted 10% of the water for oil (270g water, 30g olive oil). I was able to get an average membrane of 23.3" out of it.
There has to be a point where adding more water will do more harm to the extensibility than good (soup doesn't make a very good membrane), but it looks like I didn't quite reach it in this experiment. Seeing as the gains past 57% were modest, and the increase in difficulty of handling was significant, I probably wouldn't go much past it for stretchiness sake alone. My go-to pizza dough recipe uses 62% hydration, and as a result of the results found here, I might experiment with lowering it for practicality sake. On the other hand, it could be that if the 64 and 71% batches balled-up properly in the KA, they might have proven to be even more extensible.
I do have pictures, and more procedural data if anyone wants it.