Every couple weeks or so, we send out a little roundup of new recipes, techniques, and tutorials that we’ve recently posted on the site. Occasionally we announce exclusive giveaways to newsletter subscribers as well. We won’t spam you with ads or share or sell your email address. Every email we send has a 1-click unsubscribe link if you decide it’s not for you.
Challenging Sourdough Starter Convention
Can you guess which of the two whole grain breads in this photo was risen with only a teaspoon of cold, not-fed-in-three-weeks sourdough starter and which was made with a standard amount of active refreshed starter? They look the same, right? In this blog post, I’m going to describe two rounds of experimental baking in which I evaluate whether the robust, often complex, sourdough starter refreshing protocol of most home bakers actually results in a better loaf of bread than using a small amount of cold, essentially neglected starter.
BACKGROUND
Over the years I’ve watched @homebreadbaker’s sourdough starter regimen evolve from feeding his starter the night before baking and using a standard amount of starter in his dough, to using small amounts of starter, and finally to using tiny amounts of unfed starter straight from the refrigerator. All while producing lovely loaves of bread like the one below.
100% Whole Grain Sourdough Leavened Bread
Intriguing to say the least. While the speed of dough fermentation is undoubtedly driven by starter activity and quantity, @homebreadbaker and I wanted to know if increasing these factors truly made better bread or an easier process. We discussed some questions that his great results had raised (Ha! Pun intended) and we came up with an experiment that I would do a couple of times to try to answer them.
QUESTIONS
Do you risk gluten degradation from a very slow ferment or does the dough hold up for the long fermentation time needed when the starter inoculation is small? How do the crumb, score bloom, and oven spring compare?
Given the amount of time it takes many people to build their ideal refreshed starter, does using ample ripe starter actually result in time savings?
Same start time. In this test round, the Unfed Micro loaf of bread was actually baked first.
How is the flavor of the bread impacted by using a small amount of unfed starter vs. a standard amount of ripe starter?
Does the impact of starter amount and quality vary with wheat types and with whole grain vs. refined flour? I tested bread flour and I tested all whole grain turkey red and hard white (75:25). Innumerable more tests are possible to try to answer this question.
METHOD
I did two rounds of experiments. In one test, I baked two loaves with bread flour and in the other test, I baked two loaves with whole grain, home-milled flour. In each test, one loaf had a tiny amount of unfed starter and the other had a fairly standard 80g of recently refreshed, ripe starter.
My goal was to keep all variables constant except fermentation time. I moved the doughs to the next step in the bread-making process when they looked and felt ready, not simultaneously or looking at the clock.
In each flour-type pair the unfed micro inoculation dough and the well-fed standard inoculation dough had the same weight, hydration, flour composition, gluten development approach (no knead), preshaping (boule), bench rest time, shaping (batard without stitching), scoring (down the center) and baking processes.
I used identical mixing approaches (dissolving the starter in water, as this seems especially important given the small amount of starter in one of the two doughs and the no-knead approach), straight-walled bulk fermentation containers, no stretching and folding or other gluten development that might impact the doughs differently at different times in the fermentation process, identical proofing baskets, and central side-by-side positioning in the refrigerator when retarded. I tried to make the same motions with my hands when pre-shaping and shaping. I used a ruler to check dough growth and I checked the pan temperature after preheating.
In the bread flour experiment round, a variable that I didn’t keep constant was the ratio of cold-to-warm fermentation time. Both doughs slept in the cold when I slept, even though one had been mixed up later and had been fermenting at room temperature for less time. In the whole grain experiment round, I didn’t retard the doughs during the bulk fermentation (all warm), and I kept the cold-to-warm final proofing times similar with a 4 a.m. bake for one of the loaves to ensure this. Yep, for science.
One variable that I thought I was keeping constant but came to doubt was shaping. I made the same motions with my hands and folded each dough in the same ways, but as soon as I plopped the shaped dough into the bannetons, I felt they looked different. Initially, I blamed myself for inconsistent shaping, but when it happened again in the second round of doughs, I began to wonder if it was the dough, because in both rounds, it was the Unfed Micro dough that felt more floppy during the transfer to the banneton.
RESULTS
The ultimate result of these experiments is that I ended up with four attractive and tasty breads, and the realization that I don’t have to micromanage my starter. There were some differences in flavor and oven spring that might warrant using the standard starter protocol, but they certainly don’t demand it.
In both the bread flour and whole grain experiments, the crumb was essentially the same between Unfed Micro and Well-Fed Standard.
Four half loaves. In each set, Unfed is on the left and Well-Fed is on the right.
In the bread flour round, the score bloom was better for the Well-Fed Standard dough, whereas in the whole grain experiment, the score bloom was the same between the two loaves.
In both the bread flour loaves and the whole grain loaves, the Well-Fed Standard breads had more oven spring. I’m still unsure if this was due to my inconsistent shaping or gluten quality. Although because both the Unfed Micro doughs felt floppier during their transfer to the banneton, I’m leaning toward attributing it to gluten strength. I’d like to do more rounds of experimenting to confirm this though.
In the bread flour breads, the flavor of the Unfed Micro was only a smidge more sour, and this is likely due to the extended cold process I had for both doughs. In the whole grain breads, the flavor of the Unfed Micro was significantly more sour, but not too sour, in my opinion.
EXPERIMENT NOTES AND DETAILS
These experiments were performed on doughs of particular flour composition and hydration, and I suspect that changing some of these variables could impact the results I had. For example, at a much lower hydration or with a different wheat variety, perhaps a long process could have a different impact on crumb or oven spring. Additionally, my process here involved no autolyse or gluten development maneuvers (stretching and folding, lamination…etc), so I can’t claim to know how they would interact with starter amounts and activity. However, the dough of @homebreadbaker is usually treated with lamination and stretching and folding, and comes out bloomed, open and full of ear, so my hunch is that these actions would only improve the results of both starter approaches.
BREAD FLOUR TEST BAKE
Unfed Micro Formula
460g bread flour
345g water
20g starter
9g salt (1.5 tsp)
(total flour 470g, total water 355g, 76% hydration)
Well-Fed Standard Formula
430g flour
315g water
80g starter
9g salt (1.5 tsp)
(total flour 470g, total water 355g, 76% hydration)
Bulk Fermentation
The Unfed Micro dough bulk fermented 9.75 hours at room temperature, 8.5 hours in the refrigerator, and another 5 hours at room temperature (14.75 RT and 8.5 COLD). The starter was pulled directly from the refrigerator and had not been fed in close to three weeks.
The Well-Fed Standard dough bulk fermented 4.5 hours at room temperature, 8.5 hours in the refrigerator, and another 2.5 hours at room temperature (7 RT and 8.5 COLD). The starter, built from the contents of my active jar and not the 3-week-old jar, took 4 hours to triple in size.
The Unfed Micro dough spent more hours bulking at room temperature.
Final Proof
Once shaped and proofing, I expected the doughs to move at the same pace, or possibly for the Well-Fed Standard dough to continue to develop faster. Much to my surprise, this was not the case. Both doughs spent 25 minutes at room temperature before going into the refrigerator, but the Well-Fed Standard dough proofed an additional 23.5 hours in the refrigerator (and still looked a smidge smaller when it went into the oven), while the Unfed Micro dough proofed only 17.5 hours in the refrigerator.
Approx 40.75 hrs total fermentation for the Unfed Micro.
Approx 39 hrs total fermentation for the Well-Fed Standard (and an additional 4 hours of starter build).
BREAD FLOUR PHOTO GALLERY
WHOLE GRAIN FLOUR TEST BAKE
Unfed Micro Formula
350g whole grain turkey red flour (75%)
115g whole grain hard white flour (25%)
380g water
10g all purpose flour starter
9g salt (1.5 tsp)
(total flour 470g, total water 385g, hydration 82%)
Well-Fed Standard Formula
323g whole grain turkey red flour (75%)
107g whole grain hard white flour (25%)
345g water
80g starter (built from 10g AP starter, 27g turkey red, 8g hard white, 35g water)
9g salt (1.5 tsp)
(total flour 470g, total water 385g, hydration 82%)
Bulk Fermentation
The Unfed Micro dough bulk fermented 12.3 hours at room temperature. The starter was pulled directly from the refrigerator and had not been fed in close to three weeks.
The Well-Fed Standard dough bulk fermented 5.25 hours at room temperature. The starter, built from the contents of my active jar and not the 3-week-old jar, took 7 hours to triple in size before mixing (1:3.5:3.5 feeding to get consistent whole grain to all purpose ratio with the other dough).
Final Proof
Both doughs spent 35 minutes at room temperature before going into the refrigerator. The Unfed Micro dough proofed 9 hours in the refrigerator. The Well-Fed Standard dough proofed 7 hours in the refrigerator.
Approx 22 hrs total fermentation for the Unfed Micro
Approx 13 hrs total fermentation for the Well-Fed Standard (and an additional 7 hours of starter build)
WHOLE GRAIN FLOUR PHOTO GALLERY
Challenging Sourdough Starter Convention