stand back, i’m going to do science! (and it will be delicious and might make you fat)

as i have blogged before, i’ve been working to perfect my pie crust recipe and technique for years now. The Cook’s Illustrated recipe that uses vodka (more on that below) is nearly ideal, but there was one nagging problem left: it uses Crisco, which violates all of my whole food principles. one hundred years has given Crisco a good long time to get its claws into the baking industry (according to their website, it was invented in 1911. warning signs one and two: ingredients shouldn’t be copyright-protected proper nouns, and they should be “invented” in a factory), and there are critics and foodies who swear that perfect pie crust cannot be achieved without Crisco. but i disagree. pie crust existed before Crisco. the key is in quality ingredients and rigorous technique.

so the challenge was this: replace the Crisco in the given recipe with butter, since Crisco is Not A Food, and butter, while bad for you, is at least an actual food. there may not be room for Frankenfoods in baking, but baking is still science. so science is how i tackled this problem.

Chris and Teresa gave me this awesome book called Cooking for Geeks last Christmas. think of it like Alton Brown, but with less hipster foodie, more straight up geeky science. last weekend, a page explaining why substitutions can be tricky in baking caught my eye. as an example, it compared a Joy of Cooking pie crust (which uses Crisco as the main source of far) with a Martha Stewart recipe (which uses only butter as a source of fat). though on the surface they appeared to have different quantities of water and butter/Crisco, using baker’s percentages the book demonstrated that the two recipes were nearly identical in structure. one of the keys to understanding the comparison was to know that while Crisco is 100% fat, butter is about 85% fat and 15% water.

so in my quest to de-Frankenfood my pie crust recipe, a straight substitution of butter for Crisco doesn’t work. it decreases the overall amount of fat and increases the amount of water. so i made adjustments. calculated how much butter was required to maintain the correct percentage of fat, then figured out how much extra water that brought to the recipe and decreased the water accordingly. awesome, right?

not quite awesome. i went about making the recipe, and when it was time to incorporate the liquid into the chopped butter/flour mixture, there didn’t seem to be enough liquid. great sections of it were still dry and crumbly. i ended up having to add a significant amount of water to get dough to form, which is not ideal in that it makes for stickier dough and tougher pie crust.

what went wrong? the answer popped into my head about 6am the next morning*. water. liquid. the water in the butter, though technically present in the baking process, is not accessible as a liquid when it comes to making the dough come together. though i had maintained the ratios of fat and water, i had not maintained the ratio of LIQUID. of course! that’s why it seemed so dry and crumbly.

now here’s where the Cook’s Illustrated pie crust recipe really rocks my world: it doesn’t use water as the only liquid. it uses half water, half vodka. vodka, you say? yes, vodka. here’s the secret: vodka is 60% water and 40% ethanol (80 proof). and, drum roll please, gluten doesn’t form in the presence of ethanol. though revered in the world of bread-baking, gluten is the enemy when it comes to pastry dough. so the ethanol provides extra liquid that helps pull the dough together, and then you put the crust into the oven and presto! the ethanol evaporates, leaving awesome pie crust behind with a comparatively low water content.

so, in reworking with this recipe, i had another element i could tinker with. if i wanted to maintain ratios of fat, water (both the liquid portion and the portion bound up in butter) AND liquid (the water and water/ethanol mix that is vodka), i had to adjust the amount of vodka, not just the butter and water.

an hour passed. i converted volume measurements to weights. i filled a piece of scrap paper with long-forgotten algebra, solving for multiple variables. i built a little spreadsheet to double check my math. i cut the Crisco, added more butter, decreased the water but increased the vodka. presto! all bakers’ percentages intact, but a new recipe that uses no Crisco:

Original Recipe:
*from Cook’s Illustrated

weight volume ingredient bakers’ %
354.0 2.5 c unbleached all-purpose flour 100.0%
6.0 1 tsp salt 1.7%
26.0 2 TBS sugar 7.3%
170 12 TBS cold unsalted butter, cut into 1/4-inch slices 48.0%
144.5 (fat in butter) 40.8%
25.5 (water in butter) 7.2%
 47  4 TBS cold vegetable shortening, cut into 4 pieces (100% fat) 13.3%
56 4 TBS water 15.8%
56 4 TBS vodka 15.8%
22.4 (ethanol in vodka) 6.3%
33.6 (water in vodka) 9.5%

No-Crisco version:

weight volume ingredient bakers’ %
354.0 2.5 c flour 100.0%
6.0 1 tsp salt 1.7%
26.0 2 TBS sugar 7.3%
225.3 16 TBS butter 63.6%
191.5 (fat in butter) 54.1%
33.8 (water in butter) 9.5%
35.3 2.5 TBS water 10.0%
76.8 5.5 TBS vodka 21.7%
30.7 (ethanol in vodka) 8.7%
46.1 (water in vodka) 13.0%

* so sometimes i wake up at 6am thinking about pie. shut up.