Bread: How much do you knead?

My ideal bread—the bread I want to have for breakfast every morning, around my sandwiches at lunch, and to sop up the remains of whatever sauce adorned my dinner—is a crisp-crusted, chewy, open-crumbed bread, flecked with bran. This is the kind of bread perfect with a slice of cheese, some large-grained cured sausage and a big swig of coarse red wine to wash it all down. Rustic bread.


The concept is one thing, the creation of this imagined bread is another. Recipes from cookbooks have their virtues, but ultimately none has been totally satisfactory. Over the past year, I’ve tried to understand the techniques underlying the recipes, to manipulate the variables and create a bread that lives up to my ideal. I experimented with hydration percentages, finding that a wet — but not too wet — dough helped to create the open structure I was after. Next, I tested delayed fermentation to see what effect it had on my breadmaking. Lately I’ve been thinking about how I was mixing the stuff: kneading.

Never impervious to trends, I went through a no-knead phase. The results of the various no-knead recipes I tried (my favorite was Cook’s Illustrated’s No-Knead Bread 2.0) were always very consistent, and actually pretty close to what I was after: big open crumb, slightly sour flavor, crackly brown crust. That was all well and good, but I couldn’t shake the feeling that I was turning into little more than a bread machine: mix the given amounts of flour, water, salt and yeast, let them rest and bake for the prescribed amount of time, and then poof! bread. It was a good bread but not one over which I felt much ownership of or had any control over. Using the no-knead method, breadmaking felt more magic than craft.

Having rejected not-kneading, I went on a kneading binge. No bread passing through my oven would be kneaded any less than ten minutes, vigorously and by hand. I settled on this method mostly as a sentimental reaction against no-knead — good bread was something you worked for, dammit — but I also had a somewhat technical justification: the repeated working of the dough was helping to create a strong gluten framework that would support the airy internal structure I was after. And sometimes, it did. But I also found that often my kneaded bread would be very fine-textured, lacking the big holes that make me think “good bread.” Rereading Harold McGee’s On Food and Cooking to research delayed fermentation, I came across an explanation for what was happening in my kneaded bread:

Kneading also aerates the dough. As it’s repeatedly folded over and compressed, pockets of air are trapped and squeezed into smaller, more numerous pockets. The more pockets formed during kneading, the finer the texture of the final bread. Most of the air pockets are incorporated as the dough reaches its maximum stiffness. (538)

All my diligent kneading may have been making strong gluten strands, but it was also crushing and dividing the tiny gas pockets that explode in the oven into my sought-after holes.

I needed a third way: a technique by which I could strengthen and bond long gluten chains while seeding the dough with large gas pockets. The answer was stretch & fold, a technique which I had first encountered as a way of dealing with extremely wet doughs, but only began to consider seriously as a general technique after reading and baking Samuel Fromartz’s baguette recipe. In stretch and fold, after dough is initially mixed it is allowed to rest for ten minutes. Then, using a bench scraper, the baker stretches the dough into a long strand in one direction before folding it in half over itself. The stretch and fold is repeated in the other three directions (check out this video!). The dough is then rested half an hour before being stretched again. I suppose this process could be repeated indefinitely, but I usually stretch and fold the dough four times over an hour and a half. After the final fold the dough can rest overnight in the refrigerator (I just can’t give up on delayed fermentation) and it is ready to shape, proof and bake.

Stretch and fold has given me impressive results and I have been tempted to say that it is the technique for achieving the bread I am after. But there had been times when I felt the same way about no-knead, and ten minute kneaded dough — those techniques had just fallen out of favor with me lately. To ensure that stretch and fold really was something different (and better) I conducted a head-to-head-to-head kneading technique breadoff.

I started by preparing a 3# batch of 68% hydration dough using:

  • 6 oz wild-yeast starter (100% hydration)
  • 16 3/8 oz water
  • 23 5/8 oz white all purpose flour
  • 2 oz whole rye flour
  • 1 T sea salt
  • 2 t instant yeast

Immediately after mixing to form a shaggy ball, I divided the dough into three 1# balls. One was placed immediately in a plastic bag and left to rest on the counter: this was the no-knead bread. Another I left in the mixing bowl to rest ten minutes (ample resting seems crucial to the stretch and fold technique). I spent that ten minute resting time kneading the third ball of dough, using no additional flour so as to keep the recipes constant.

Some dough

After kneading I placed the dough in a bag next to the no-knead dough. Because the stretch and fold technique requires the dough sit at room temperature for close to two hours as it rests between stretchings, I left the other two bags on the counter as well so all three dough balls would have the same chance at yeast activity. I followed the procedure as I described above. At the end of the stretching/resting period, all the doughs looked similar, although the kneaded and no-knead doughs appeared more voluminous than the stretch and fold, probably due to their extended rest.

Dough in a bag

All three bags spent the night in the refrigerator.

Although three 1# dough balls will fit in my oven at the same time, it’s a tight fit and the breads close to the edges of oven tend to burn and grow towards the center. For optimal results, I needed to bake each bread in roughly the same place in my oven: the center of the stone. I couldn’t just pull out every dough ball out of the refrigerator to proof and then bake one at a time. That would give the third-baked far more time to proof than the first. Instead, I staggered the breads, proofing each bread for one hour in a proofing basket then scoring it once down the center and baking for 25 minutes in a 450°F oven with a preheated steam pan bearing 1 cup of room temperature water. True, this meant that the third dough ball would spend more time in the refrigerator than the first and second, but because the cold temperature means nothing happens very quickly, I thought the influence would be negligible. All the breads were baked within two hours.


The first bread I baked was the no-knead, followed by the kneaded bread, ending with the stretch and fold.

As I pulled the breads out of the oven, I was surprised by the extent of the differences. Where the no-knead bread was roughly cracked and browned, giving a very rustic, rough appearance, the outside of the kneaded bread was smooth and uniform.



The stretch and fold bread was similar in appearance to the no-knead but almost a half-inch taller.

Stretch and Fold

Circumference (In) Max Height (In)
No Knead 17 1/8 3 1/32
Kneaded 17 1/2 2 23/32
Stretch and Fold 16 3/4 3 15/32

The different external appearances were a sign of unique internal structures. The interior of the no-knead bread was familiar: haphazard large holes here and there, largely concentrated on the edges.

No knead autopsy

The kneaded bread, I was surprised to see, had much larger holes, although it also had large areas of uniform, fine texture. The large air pockets were possibly the result of my technique of forming a loaf; tucking the edges of the dough under it might have trapped large air pockets that were maintained by the strong gluten network.

Kneaded bread or swiss cheese?

The stretch and fold bread seemed like a combination of the other two. Although its structure was similar to the no-knead bread, the holes were larger and more evenly distributed. Of the three, here was the closest to the crumb structure I imagined for this style of bread.

Holes, evenly distributed, voluminous: bread!

But bread was not meant to be looked at; it should be eaten! Would my different techniques result in dramatically different flavors? Although I have been told that mouthfeel (texture) influences perceived flavor, I can say that these differently textured breads tasted essentially the same. All the breads were chewy and substantial, with a deep flavor of grain. I thought that I noticed the crust of the kneaded bread was slightly more chewy and less crispy than that of the other two, but after a few more bites I couldn’t be sure. The stretched and folded bread had slightly more fermented flavors than the other two. Overall, though, once the bread was in my mouth I couldn’t notice a major difference. A blind tasting panel, a more sophisticated palate, or a battery of chemical and mechanical tests would all have helped to better discern the differences. As far as I’m concerned, it was all pretty good.


Looking over the three slices, stretch and fold is the best technique for making rustic bread. Both other techniques yielded good enough breads, but neither could compete with the open crumb and lofty structure of the stretched and folded dough. In some ways, this is also the most involved technique: no-knead bread is over almost before it starts, and kneaded bread takes just ten minutes of intense activity. The act of stretching and folding is not particularly time consuming, but the dough does require attention every half hour for a couple of hours. You can’t just walk away from it. Maybe the technique’s appeal comes back to the sentimental: after working with stretched and folded bread over the course of an afternoon, it feels like I actually did something.


«   »

15 comments on “Bread: How much do you knead?”

  1. Peter 10 December, 2009 at 12:29 am

    This was a really interesting post. I’m not really known for baking much bread, but if I ever do I’ll definitely remember this.

  2. Tom 10 December, 2009 at 8:39 am

    Any time you want to bring some wort and some homebrew we can spend the day perfecting spent-grain bread and doing our unique brand of sober and scientific beer tastings.

  3. Kevin 10 December, 2009 at 8:47 am

    Geeze- And I thought a bread machine was adequate for kneading!

  4. Tom 10 December, 2009 at 8:51 am

    Well, the bread machine might be, but only if you want that really fine, uniform texture. Has anyone ever successfully made “rustic” bread in a bread machine? Seems sort of a contradiction.

  5. Kim Ode 10 December, 2009 at 11:12 am

    I love this post! And appreciate your due diligence. I use stretch-and-fold for my ciabatta, but hadn’t extended that technique to anything else. (And I second your somewhat perverse view of no-knead as being more magic than technique; we must crave the flavor of labor – or at least of attention.)

  6. Tom 10 December, 2009 at 11:19 am

    Stretch and fold is a little harder with lower moisture dough because they don’t stretch quite as readily as something like ciabatta, but it can be done. Regarding no-knead bread being too easy, I liked this quote from Thomas Keller (which I found in the recent NYTimes article on the Sous Vide Supreme): “There’s some danger in cooking techniques that don’t require much attention,” Mr. Keller wrote in “Under Pressure.” “Eliminate the need to pay attention and you eliminate the craft.”

  7. Uncle Don 10 December, 2009 at 12:09 pm

    Wonderful explanation of why my bread is so fine. I shall give this method of stretch-and-fold a try. Now if I could only get you to write your recipes using metric units!

    U. D.

  8. Tom 10 December, 2009 at 12:13 pm

    I don’t know why, but I am a reactionary about metric measurements. Who needs millimeters when I have 32nds of inches?

  9. Linda 10 December, 2009 at 9:10 pm

    All of your bread posts rank among my favorites. I admire your patience. Nice proofing baskets :)

  10. Tom 10 December, 2009 at 9:15 pm

    I haven’t made a bread since October that hasn’t passed through those baskets, save a few baguettes. Thank you Linda!

  11. Uncle Don 11 December, 2009 at 8:12 am


    It’s grams I want to see! And 32nd of an inch! Are you using a caliper?


  12. Cindy 14 December, 2009 at 2:03 pm

    Wow, thanks for that detailed experiment and post. I have been wondering what the difference in results would be between kneading and stretch and fold (I had not considered no-knead).

    I am part of the Bread Baker’s Apprentice Challenge, one of 212 bakers from around the world, baking our way through Peter Reinhart’s book “The Bread Baker’s Apprentice”. I am on Bread # 26, Poolish Baguette, and decided to try the stretch and fold technique. I was pleased with the results. Your post was very well written and helpful. I plan to link to it for others to see.

  13. Tom 14 December, 2009 at 9:09 pm

    Thanks for the kind words Cindy. What a fun — if not a little daunting — project you’ve undertaken! The Bread Baker’s Apprentice is one of my most used cookbooks and has taught me a lot of what I know about bread. I can only imagine what could be learned by going through it systematically.

  14. alexandra 14 July, 2011 at 7:30 am

    Dear Tom, I came across your 2009 post today and just want to say that you write very well, are a sound thinker and thanks for doing this time-consuming experiment for me! I have baked regular kneaded breads with a lot more success than the 4 loaves I recently did with the no k nead artisan bread in 5 min. The no knead aren’t rising well and are not real impressive tasting. I would love to know which book I can buy to teach me about this stretch and fold method. Would you let me know please. Thanks,

  15. Tom 14 November, 2011 at 12:38 pm


    I am not aware of any books on the S&F method, but they are probably out there. I learned a lot from the Fromartz post referenced above. Here’s another post: It’s also fun to experiment: take a dough recipe you like (preferably one that’s pretty wet) and instead of following the usual mixing instructions take it through a series of stretch-and-folds and see if you prefer the result.

    Posts linking to this post

  1. December 11 Morning Roundup « The Heavy Table 11 December, 2009 at 10:16 am

    […] a review of the The Red Mug in Superior, the art of “salvage” kimchi, Martha and Tom make some rustic bread, and a report from the First Annual South Minneapolis Blizzard Pub Crawl. var addthis_pub = […]

  2. MARTHAANDTOM » Focaccia Pizza 12 November, 2010 at 8:19 am

    […] is that its hydration level is 80%: that’s way to wet to knead by hand. You could use the stretch and fold method, but it will greatly increase mixing time and should really involve an overnight rest. If […]

  3. # 26. Poolish Baguette Interruptus | Salt and Serenity 26 January, 2013 at 12:25 pm

    […] my research I came across a very interesting post by Martha and Tom, where they do an experiment and bake 3 loaves side by side for comparison.  The first […]