Perfecting Choux Pastry: The Liquid Variable (part 1)

Milk or water, water or milk, perhaps a bit of both. Does it even make a difference? Let’s find out. However, before I start testing, I want to see whether there are any patterns in the recipes that might be illuminating. Below is the recipe table again but with only the relevant columns included and I’ve grouped the milk and no-milk ones together.Choux recipes

Initial observations
  • All recipes, barring the outliers of Delia and Raymond Blanc, use between 200 and 250g liquid total (recalling that egg is constant)
  • When milk IS used, it’s half and half with water
  • Milk recipes seem to include less butter

At this point, I’m starting to guess at some patterns, but I need to be scientific, so it’s time for a spreadsheet. However, having now made that spreadsheet, two outlier recipes are messing things up – Raymond Blanc and Delia. I’m going to ditch them. Delia is the only one using all bread flour, while Raymond’s liquid and butter quantities are really odd. Delia is also using huge amounts of water, which I doubt is a good thing.

The spreadsheet proved to be pretty helpful. Here’s a picture of what it looked like in the end.

Choux spreadsheet

I’ve grouped by flour type as well as milk, and bold numbers are averages for the group. Of course, these are averages for very small sets of numbers so they don’t mean a great deal, but even in this small group there are some patterns that I think can be supported.

Post-Spreadsheet Observations

  • Liquid to butter ratios are fairly standard – a little over 2:1 – for milk/water mix but they get rather scattered for water only recipes (and remember poor Delia who got the boot? A ratio of 3:1!).
  • Liquid to flour ratios fairly standard – about a bit under 2:1.
  • Water-only mixes used less butter in relation to flour. These were all plain/AP flour mixes, but comparing to both the cake and the plain groups in milk/water, the butter relative to flour was still lower.
  • No one used cake flour and just water.

What does this all mean? Those using water/milk tended to also use more butter. Those who used cake flour over plain flour used more butter still but only marginally more (apart from the surprisingly stingy La Duree). From one perspective, this is the opposite of what I would have expected. Milk adds fat as, of course, does butter. So I thought possibly those mixes with milk would offset this with a drop in butter. But they did the opposite. My guess is that the test will reveal that milk/water produces a more tender crumb, and I suspect butter does the same, so if you want that kind of crumb, you boost your chances of achieving it by adding fat via milk, too.

So the next step is to test and then think about it once the results are in. As everyone left in the mix is actually using plain or cake flour, that simplifies things – 2 types of flour, 2 types of liquid. No one uses cake/water so I’m not going to bother with that. I’ll keep butter levels constant at around the average (which I’ll call 110g) but not flour because I think if I mess about too much I’ll just bake a pile of rubbish. So with that scientific (I should just stop using that word) approach decided, the recipes will be….

Choux liquid recipesAnd now I need a little lie down after all that.