The Fresh Loaf

News & Information for Amateur Bakers and Artisan Bread Enthusiasts

Pain Au Levain | What's the secret?

marc's picture

Pain Au Levain | What's the secret?

I'm on the verge of throwing in the towel.

After too many attempts to count, I do not seem to be getting any closer to a great loaf of bread. Some have bascially good flavor but do not rise up in the oven into beautiful loaves of bread.

I've tried multiple formulas and it seems that the only remotely acceptable results is when I use a high gluten flour like KA Bread or Wheat Montana Natural White.

That said, my latest attempts have been with Giusto's Artisan Bread flour which I believe is around 11.5.

If I were to follow a recipe that called for 68% water, the dough is entirely too wet—regardless of how many french folds I do. I say regardless, but the most I have done is 4.

I set up a calculator in excel to get the amount of dough I need to fill two bannetons and came up with the following:



Stiff Levain 25.00% 190.4

Water 65.00% 495.1

Flour 100.00% 761.7

Salt 2.00% 15.2

TOTAL % 192.00%




25 minute autolyse.

5 minute mixer knead on speed 1

5 minute mixer knead on speed 2.

I added a bit of flour at the end of the knead during the last minute or so because the dough was sticking to the bottom of the mixer bowl.

The above I used with today's dough and had cut the water from 68% to 65% and still ended up with a pretty floppy dough.

I guess my question is: What's the secret? Should I do stretch and folds til the cows come home? Until the dough is stiff and muscular? Should I add more flour—less water? Today I did 1 stretch and fold—but it was about a minute or less of folding and turning folding and turning etc, until is felt like a springy mass

I used wet surface and hands with the stretch and folds.

I know today's batch will not be very good. The dough, even though doubled in the bulk fermentation, was simply droopy, and began to oooze off my work surface. I'm thinking that when the dough is dumped out is should more or less hold it's form in the shap of a big mound. My dough starts out as a mound and then oozes into a large flat shape. I would essentially say that it has no gluten at all.


flournwater's picture

You don't use yeast in this naturally  leveaned bread.  That means it will take much longer to rise than a yeasted formula.  Maybe you're not giving it enough time to mature.

I think you're contaminating your formula as you process beyond the initial kneading effort.  You're using a power drive kneading method (as opposed to hand kneading) and you're adding flour much too late in that process.  Try starting with a bit drier mix and, during kneading, add water a tablespoon at a time if you feel it's necessary.  I don't know what "1" and "2" means with respect to the kneading with your mixer.  But it's possible that "2" is too fast (or "1" is too slow) and your dough is getting too warm or not warm enough during that part of the process.  Actually, I dont' know why it would be necessary to change speeds half way through your kneading step.  If you kneaded by hand you wouldn't knead with greater pressure or increased speed when you had the dough about half kneaded.

When you wet the stretch/fold surface you're adding water to the formula.  If the dough is already wetter than you want it to be, working with it on a wet surface only exacerbates the problem.  A lightly floured surface would probably work better but if you do that you'll need to allow enough time for the lose flour to hydrate or you'll end up with gluey strings of dough in your loaf.   You could, of course, use an oiled surface for working with a slack dough.

You don't describe your oven temperature, humidity in the oven, type of surface (stone, bread pan, ???) you're using so those issues remain mysterious possibilities too.  The slack dough you describe, when gently degassed and gently handled, makes a pretty nice foccacia or ciabatte style bread.

Whatever you do, don't give up.  It took me years to make a loaf of bread that was worthy of human consumption.  You can do it too.

Post Script:

Forget about those decimal points.  Bread making isn't structural engineering.  Round the results of your calculations and it'll work out fine.

marc's picture

My loaves are almost done now, and have risen enough to resemble biscotti when sliced.

I guess I'd be curious as to whether any pain au levain bakers out there actually use 68% hydration or even 65%. AND, are they truly using all purpose (not high gluten bread flour).


hansjoakim's picture

Hi marc,

A 65-68% hydration for this pain au levain sounds ideal to me. What's the hydration of your stiff starter?

I have a question about your mixing times, however. Whenever I make a pain au levain, I use a flour with 11% protein. That's more like your all purpose flour. High gluten flour is good for pan breads and certain multigrain breads, but not European style hearth breads. Pick a flour with a protein content in the 11% - 12.5% range, such as the Artisan flour you've tried. When you autolyse the flour, you should be very careful not to mix too long in the mixer. Especially with lower protein flours. After a 25 min. autolyse, I usually mix my pain au levains for 2 mins. on 2nd speed and that's it. If you let the dough autolyse prior to mixing, it can break down pretty quickly in the mixer, since it's quite far developed before you even put the hook into action.

marc's picture

I'm using Giusto's Artisan Bread flour. I think it's 11.3  or 11.5%

My levain is 50% hydration.

I made bread yesterday and cut the hydration to 60%. The dough and resulting loaves were much better—though the crumb was a bit on the tight side. In general though, and much more manageable dough. I'm thinking that tomorrow I will increase the water a bit—to maybe 63%. At 60% the dough and water mixture for the autolyse just came together and I was nervous that there would be dry clumps in the dough—but luckily none.

Your mixing theory has peaked my curiosity. I'll give your suggestion a try tomorrow. Honestly, after 5 minutes on speed 1, my dough yesterday looked ready to go. Between the lower hydration and possible overmixing, I may have made a dough that was too tough to rise up nicely in the oven with a loose open crumb.

Thanks for your feedback.

Liam's picture

When I first started out with my pain au levain, or in fact any bread recipe I was trying;  I bought and used a digital thermometer to check the temperature the dough was sitting at during each stage.  It turns out that my kitchen is perfect  for rising at about 78 degrees most of the time.  I guess it's a bit too warm for the chef/levain, but I just make sure I can refresh about 9-10pm and try to begin my bread by 8 am. 

I've also run out of time or had some sort of emergency and stuck the risen levain back in the fridge.  Then I take it out the next day and just measure and begin the bread.  It's up to my mood whether I let the levain come to room temperature first.  Since my water and flour are at room temp, and I'm kneading by machine and rising the dough at my usual kitchen temp of 78 deg F it all seems to work out in the end.

All I can add is that if your current recipe is giving you headaches, go to the library, if you have access and check out as many bread books as you can.  Start with the levain/sourdough technique that seems easy to you.  Work from there.

Again hope you find this helpful or encouraging


marc's picture

Day before yesterday, I did the Susan's Simple Pain Au Levain method with very successful results. Of course that recipe called for higher gluten bread flour.

On the same morning I also mixed my formula that I had posted here (using 63% hydration)However, I followed a different mixing strategy. After the autolyse, I kneaded in the pieces of levain and salt for a minute on speed 2, the ratcheted up the speed to 4 and kneaded for 8 minutes. Then, having great success with the simple pain au levain method I followed those steps for the stretch and fold. One at 30 minutes. Another at 1 hour, and then a final at the 2 hour mark. My resulting dough was great to work with, poofed up beautifully in the oven. And, once sliced—there was a large, open crumb. BUT...the loaves were very heavy and when feeling the inner part of the bread, it felt very dense and solid. Very unlike the simple pain au levain that felt light and airy—even though the simple pain au levain used high gluten flour.

This made me wonder—did I over mix and thus make the dough too tough? You mention mixing your pain au levains for only 2 minutes. Are you making yours with an all-purpose, or a high gluten bread flour? Are you doing any stretch and folds.

Tomorrow, I'll make another attempt, and trying kneading in the mixer for only the 2 minutes you suggest.

I could probably stand to increase the hydration a bit more than 63%, but I am nervous since so many past attempts have resulted in loaves that simple do not rise up very well in the oven and are just too stickly and messy to work with.

Jw's picture

For me, commercial yeas (instant) was a lot easier, then getting on with natural levain. Do you have pictures of the various stages? Do you time the stages?

I recall baking once and leaving out of bit of the dough, for other couples of hours. Then that started to rise, and became the best bread ever. I really had to experiment, since the recipes don't account for local temperature and theight.

It took me ages to ge to the level I am at now, and I still make 'mistakes'. Mostely it is the form that does not work out, but the taste is always good.

Don't give up, it is still worth the road to get there.

Happy baking!


dmsnyder's picture

Hi, Marc.

Your formula looks good. I have some suggestions on your procedures, although they are not specific enough for me to tell exactly what you have been doing. In any case, I think you are closer than you think to great bread.


  1. Mix flour and water to a shaggy mass. Cover and autolyse as you are doing.
  2. Sprinkle the salt on the autolyse. Cut the levain into chunks and distribute over the autolyse.
  3. Mix at Speed 1 with the paddle until the salt and levain are completely incorporated. (5 minutes should be more than enough.)
  4. Switch to the dough hook and mix at Speed 2. Mix longer than 5 minutes - up to 10 minutes - if necessary to get moderate gluten development. (The dough should stretch some before tearing.)
  5. Transfer the dough to a lightly oiled bowl and cover. Note its volume.
  6. Set your timer for 30 minutes. After 30 minutes, scrape the dough out onto a lightly floured board and do a stretch and fold. Round up the dough and return it to the bowl.
  7. Repeat 6. two more times.
  8. Let your dough double in volume from the original volume noted in 5.
  9. Transfer the dough to a lightly floured board. Handle it gently, deflating it as little as possible. Divide and scale it as desired. Pre-shape the pieces into rounds. 
  10. Cover the pieces and let them rest for 10-20 minutes.
  11. Form your loaves and place in your bannetons. Handle the dough gently but seal your seams well. Note their volume.
  12. Pre-heat your oven, stone, steaming apparatus.
  13. Proof the loaves until 1.5 - 1.75 times their original volume.
  14. Transfer the loaves to a peel (or equivalent) and slash them as desired.
  15. Bake with steam for the first 1/3 to 1/2 of the baking time, cool, slice, enjoy.

Note that, for bulk fermentation and proofing, I did not give times. This will depend on levain activity and room temperature. Watch the dough, not the clock. That said, fermentation may take 2-4 hours after your 3rd stretch and fold. Proofing may take 1.5-3 hours. These are rough estimates.

If you have questions about any of these steps, please ask.

Happy baking! 


marc's picture

Thanks for the feeback David.

I'll review and incorporate your suggestions in tomorrow's bake.


Of course, now I'm a bit confused: Hans says to knead less—you say to knead more. I think my dough yesterday at 60% hydration was too dry and maybe I overkneaded and made it too tough.

Tomorrow, I'll try 63% and see how the dough feels in the mixer. Would you say that the dough should pass the window pane test before doing any stretch n folds? I've been confused about that since I started doing them as part of the process.

dmsnyder's picture

Hi, Marc.

Hmmmm ... I think hansjoakim makes good points, and it is futile to argue with his results. If you do the stretch and folds, the gluten should develop adequately. I retract my recommendation regarding longer mixing.

When I plan on doing 2 or 3 stretch and folds, I mix to the point I'm just starting to get some window paning. (Yes. Window panes have a spectrum of development.) Truth is that, for familiar breads, I stand over the mixer and watch what the dough is doing and stop when it's "ready." So, I'm not really timing the mix. It may well be shorter than it seems.


marc's picture

Hey David:

I made new dough this morning using 63% hydration, and kneading for a longer time. Did one set of stretch and folds. And now the dough is resting for a 3 or 4 hour fermentation. The dough consistently seemed much better, being just a bit wetter and having the extra kneading.


davidg618's picture


I checked your formula's math: (I rounded to the closest gram)

190 gm starter @ 25% Hydration contains;

152 gms flour

38 gms H2O


762 gms flour

495 gms H2O yields

914 gms. flour

533 gms H2O

and, Dough Hydration:

533/914x100 = 58%

That hydration doesn't correspond to the wet dough you describe.

If your total flour weight is 762 gms, and total H2O weight is 495 gms, that corresponds to 65% Hydration, but your total dough weight (including the salt) is only 1272 gms, not the 1463 you report.

I suspect, but have insufficient information to prove it, you are not adding the correct amount of flour and/or water. (My guess would be too much water, or not enough flour)

Note: another clue is 2% of 762 is 15.2 gms which is your formula's salt weight.

Apologies if I misread or misinterpreted what you wrote.

David G

pmccool's picture


I think Marc is saying that the weight of the starter is 25% of the dough's flour weight, not that the starter is 25% hydration.  That will probably jigger the results of your calculations.


davidg618's picture

I didn't think of that.

Without the starter hydration I don't know where to start recalculation.


David G

Liam's picture

Hi David

I must sound like a broken record by now, but I swear by "Bread Alone" levain.  I started it with organic hard wheat flour and spring water just like Daniel Leader explains in the book.  (four days to make.  Day 1: 4 oz  20% bran wheat flour, 4 oz. spring water by weight both, plus a pinch of yeast [moist or dry  less than 1/16th tsp]. Mix, cover and keep at about 70 deg F.  Days 2,3: add 4 oz 20% bran wheat flour and 4 oz water, stir cover and store at about 70 deg F.  Day 4 [day before bake day] add 6 oz  20% bran wheat flour to the chef and let stand 8 - 10 hours at 70 deg F.  If you aren't baking on day 4 store the chef in the fridge for up to a week.  The night before, do the all flour thing and store covered at room temp overnight)

The recipes work for me.  For a while I was religious about spring water; quantity and flour type.  Then I ran out of white hard wheat flour.  I had organic AP in abundance so I used that, still with spring water and organic whole wheat at about 20%, the bread was heavenly!  Then I went back to hard wheat, but ran out of whole wheat flour.  I used all white organic hard wheat for chef,levain and bread making.  The bread was heavenly!  For a while I mixed organic white AP and white hard wheat when I made the bread, in whatever quantities the spirit moved, as long as it came to about 29 oz flour.  The bread was great.  Then I bought a lot of organic rye, because I couldn't find a small quantity of rye near where I live.  So I substituted rye for the whole wheat when refreshing the chef.  I used about 20% rye/WHW (total 4 oz more or less) in the next loaves and guess what?  They were divine!  If for some reason I decide I want a bit more chef left over, I add 5 or 6 oz of flour, but I also add the same amount of spring water.

The next loaves following the rye course were not so great, I think I needed more flour, as the dough was kind of slack.  They tasted good though.  So at the moment I am refreshing with 20% organic rye/organic WhiteHardWheat and using mostly white for the bread, with oh, probably 10% rye and using feel to decide if it needs more flour.  The bread is..............guess???  Divine! 

The point of all of this is that I don't bother with hydration and I have become rather cavalier about whether I have 18 oz of chef or 24 oz of chef when I add the 6 oz of flour before bake day for the levain.  (That is to say I may refresh once or twice without discarding).  The bread still works.  I get good oven spring, I seldom have a problem, the bread is fabulous with a variety of additions (onions, nuts, sundried tomatoes, herbs etc.)  It just works, by and large.

I don't fret about protein or ash etc.   I just call my local organic store and ask for a bag of white hard wheat.   Usually they get it from the same supplier.  Occasionally from another but little if any re-jigging happens when I knead the dough, by Kitchenaid at speeds 2-4 depending on what I see, hear (from the mixer) and feel as the dough is kneaded.  I've tried folding........ pffft.  Though I make a mean puff pastry, so I know my technique is good.

I get the bread I'm after, and I don't agonize over calculations.  I stay strictly with organic flours and spring water, that's my only rule.  This works for me.

Hope this helps, rather than confuses.  To me it's a lot simpler than the Bread Bible's wet this and firm that.  I've looked at Reinhart's books, tried a few and was not impressed.  To much brain work for a thing that conjures up warm fuzzy memories of childhood.  I guess I'm just a simple baker at heart.

To those who follow hydration and percentages religiously I say Go with God, if you enjoy what you are doing, I won't criticize at all.

I do use the Bread Bible for a few things, but overall I just stick with my trusty levain. 

Again I hope this helps.



dghdctr's picture

This confusion over what the actual hydration is for marc's dough illustrates the pifalls created by the use of a formula format that expresses levain content initially -- right up front -- as if it were a distinct ingredient in the overall formula, like salt or yeast.  Certainly the amount of any pre-ferment contained in a formula is a vital consideration, but if we choose to focus upon it too soon in our calculations, then it becomes difficult to see in just one glance what the true hydration of a dough will be.

I can only guess as to why so many pro bakers and a number of books still adhere to an ingredient accounting system that can leave a reader wondering what the consistency of an unfamiliar dough will be.  I truly believe that it is a pernicious holdover from the days when using that system worked well enough, and that we as 21st-century bakers are in a period of transition where the weaknesses of that system are becoming apparent.  Getting most bakers to see the issue from a global perspective will continue to be a challenge.

I'm just speculating, but I think that bread baking has only recently been esteemed as a subject worthy of intense, scientific analysis.  In the not-too-distant past,  bakers who took on apprentices were analysts and teachers of a sort, but it might be safe to say that theory took a back seat to practicality in everyday lessons.   Also, in any given bread baking community, there was probably less communication with bakers from far-away places who used any significantly different processes.  There's no urgent need to compare the contents of one levain with another if they were most likely about the same (as was poolish).

I think that, at that time, this practice of just listing levain or poolish as if they were merely one more ingredient in your bread dough (instead of parsing out their flour and water content) may have worked well enough.  The head baker/owner in any establishment decided what the hydration of any levain or "yeasted" pre-ferment might be, and it rarely varied.  Since his flour and so forth was the same as most other bakers,  there was less need to precisely translate what "levain" might mean to anyone outside of their establishment.

These days, we're surrounded by more research on the fermentation and rheology of bread dough than we can easily use.  With the internet as an affordable way to talk at length with other bakers, we can even talk shop with people we've never met, and the number of cross-cultural influences in baking is increasing.  Differences in flour characteristics have moved to near the head of the line as considerations in creating balanced dough formulas.

It ain't simple anymore.  At least not if you want to speak effectively to bakers elsewhere who have different ingredients and practices from your own.  We need to use a written baker's language that has meaning across most borders, which conveys the likely characteristics of any dough in just one glance, with no further calculations.

This is not meant as criticism of marc or anyone else already using the system that (I think) is now so confusing.   I think most bakers -- pro or non-pro -- may still be using it, and they continue to pass this method along.

Notice, though, that Jeffrey Hamelman does not.

--Dan DiMuzio


davidg618's picture

The correct ratio of flour and water for 65% Hydration and a 1463 gm (including 15 gm salt) dough weight is:

878 gm flour

570 gm H2O

David G

davidg618's picture

Yes, I routinely bake Dan DiMuzio's pain au levain, and Hamelman's Vermont Sourdough. Both are approximately 68% Hydration.

David G.

marc's picture

I originally tried DiMuzio's formula with Giusto's artisan bread flour and each time I tried the dough was a complete sticky mess. However, when I tried it with high gluten bread flour, the results were better, but then there were the side effects of using high gluten flour—tougher, denser bread and tough chewy, and leathery crust.

I switched to the Artisan flour because I saw a post from Dan that I interpreted him as saying that when he says "bread" flour, he is essentially talking about all-purpose flour. Maybe I got that wrong?

SylviaH's picture

I think what is meant by the all-purpose is the brand used..King Arthur's All Purpose Flour instead of using King Arthur Bread's to strong..their AP works where bread flour is called for in a lot of recipes!  So you will find there is a difference in the hydration.


dghdctr's picture

I switched to the Artisan flour because I saw a post from Dan that I interpreted him as saying that when he says "bread" flour, he is essentially talking about all-purpose flour. Maybe I got that wrong?

 Hi Marc,

The bread flour I used in writing my book is the same I used most of the time working in bakeries -- King Arthur Sir Galahad.  In retail grocery stores, it is branded as "King Arthur All Purpose", but it is the same flour, according to their flour techs.

As Sylvia suggested in this thread, one miller's AP can be very different from another's -- maybe even more so than their bread flours might differ.  Most pro artisans like to use a not-too-strong hard winter wheat flour, and KA's All Purpose (Sir Galahad) is completely that.  I believe Gold Medal's "Better for Bread" flour is a winter wheat/spring wheat mixture sold to commercial bakeries as "Harvest King."   I usually found it to be very strong, and had some occasional difficulty in getting it to extend easily for baguettes.

If you still care to use my formula, I'd try it again with KA's AP flour and see what you think.  It sounds like your levain may be very enzyme-active, so you might want to examine the feeding schedule, feeding proportions, or maintenance temperature of your levain to minimize the possibility of excess enzyme activity.  It is often too much amylase enzyme that causes stickiness, and too much protease enzyme that weakens the gluten structure.

Hope that helps a bit,

--Dan DiMuzio

hansjoakim's picture

One way to proceed, could be to replace the firm levain with a stiff yeasted preferment, such as a biga or a pate fermentee. That could help spot any excess enzymatic activity in the sourdough.

Another thing that struck me, is the reported stickiness of this dry dough. I can only speculate of course, but I'm curious about the combined effect of autolyse and the long mixing time. 5 mins on 1st followed by 5 mins on 2nd speed is long if you've already performed an autolyse on the dough. I recall Hamelman warns about overmixing doughs that have undergone an autolyse - if the dough is overworked, wouldn't there be a risk that the dough starts "weeping", and lets go of water? That could be a reason for the observed stickiness in this firm dough, right?

I'd say, first replace the sourdough levain with a yeasted preferment (to gauge enzyme activity). A second step could be to eliminate the autolyse stage (to secure against overdeveloping the dough).

Ambimom's picture

This may be anathema to the bakers here, but bread is an art as much as it is a science.  Use ALL of your senses, not just your calculator.  How does it look?  What does it smell like? How does it feel?


flournwater's picture

You have my endorsement.  However, even in art, an understanding of the science usually comes first.  The artist must know the science behind the medium (how to prepare watercolor and oil paints, composition of stone, etc.) before he/she can create successfully.  We can't share artistic talents in a forum like this, at least not beyond trying to describe desirable results the baker can expect throughout the process of making a loaf of bread; things like dough texture, reading the "window pane", the whisper of the dough telling you when it's proofed enough.   But if we can help others understand the science it may, hopefully, reduce the time and effort they will have to apply in eventually becoming bread making artists.


marc's picture

I completely agree!

I'm just trying to get my dough somewhere close to the ballpark so that I can actually tell what it should look like, smell like and feel like.

Right now—at 68% hydration, regardless of the amount of levain—the dough is well beyond sticky and messy. If I add enough flour to pull it together, it still falls apart after the first bulk fermentation. By falling apart—I mean—when I dump it on to my work surface to divide for rounds, the dough is like pouring out a large batch of chocolate mousse. It simply spreads and spreads, rather than holding together in a mounds and being "light and springy" as described in the books where I obtained these formulas.


Liam's picture

Amen to that!

dmsnyder's picture

What kind of water are you using? (Tap? Spring? Distilled?)


marc's picture

I'm actually using Tap water.

I use to only use bottled mineral water (Evian to be specific) until I realized there was no difference in the result when testing with tap water.




marc's picture

started out great...but I may have over proofed them. But I also wonder if the scoring is part of the issue. The photo below shows an example of how I scored the loaves (on the left) but I'm wondering if I shouldn't be scoring so completely across the loaf and maybe score more like the image on the right.

The finished loaves baked up more like shape "A", whereas I am trying to obtain a shape like in "B"

I'm sure that these may have been overproofed, though they didn't completely fall until I scored and slid them on to the stone. But I'm still interested in any feedback about the scoring.

Today's bread was made with 63% hydration & Giusto's Artisan Bread Flour (11.5%). I did a longer knead and the resulting dough felt great. These smelled wonderful in the oven—and I'm sure they will taste great.

nicodvb's picture

did you try to make the final dough rest in the fridge (covered) overnight? It helps a lot to consolidate the gluten and the results are often surprising and much better than without the cold rest. Another trick that helps develop the bread higher rather than larger is to roll the dough flat on a floured surface and roll it tight; when it's cold it can be handled much easier, even if it's very wet.

Susan's picture

Have you tried spiking the dough with just a tiny bit of yeast?  If you did that, and it worked, it would show you that your starter wasn't doing its job properly.  99% of my sourdough problems were related to starter.  Gotta run, good luck.

Susan from San Diego