The Fresh Loaf

A Community of Amateur Bakers and Artisan Bread Enthusiasts.

Adding a Poolish To An Existing Low Hydration Recipe

AsburgerCook's picture

Adding a Poolish To An Existing Low Hydration Recipe

Just because we can, doesn’t always mean we should.

This is a companion piece to my Solving the Mystery of Polish post, and overcoming the many problems I ran into adding a poolish to sandwich bread.

Pull quote: The dough was extremely hard and dry, and with different iterations of the same recipe, I tried all kinds of different tips for kneading stiff dough. When I thought maybe it was kind of ready, I’d use the stand mixer to knead it longer. I never got any kind of “windowpane” dough, and it always ripped apart no matter how long I kneaded. But the recipe said it would be more like a “lump of clay.” Sure. When I baked it, it was awful.

Not all bread needs a pre-ferment like a poolish. Hotdog buns, burger buns, a lot of sandwich rolls, and very soft dinner rolls would become just too complicated and time-consuming. And you don’t win anything with little bits of bread. I’ve found that simply adding some dehydrated potato flakes (i.e., boxed instant mashed potatoes) is more than sufficient. But for an everyday white or whole wheat sandwich bread, the poolish is magic!

I’ll use as a reference for this post, my go-to 100% Whole Wheat Sandwich Bread from King Arthur (KA).

Simply following the recipe, produces a really wonderful bread. After discovering poolish, I emailed KA to ask if I could use a poolish, and also whether the potato flakes changed the hydration. They wrote back that poolish would be great, and that in their experience, since flour and potato flakes absorb about the same amount of water, it’s a net zero.

Here’s what we’ll be talking about:

113 g water
113 g milk
113 g orange juice
339 g liquid.

425 g Whole Wheat flour (white or dark are both excellent, and either one works)

64 g Dry Potato Flakes.

The original hydration calculates to 339 / ...what? Problem 1: Do I include the potato or not? That’s why I wrote KA. Every article I’ve read about using potato flakes (instead of tangzhong), they always use the word “Replace.” As in; “Replace about 1/4 Cup of flour with instant potato flakes.” Therefore, and as KA replied, I’m including the potato in the hydration calculation.

If we do not include the flakes: 339 / 425 = 80% hydration (79.7).

If we include the flakes: 339 / 489 = 69% (69.3).

We have a 10% variance in the hydration, and haven’t even got started yet! My solution was to ignore it entirely, and just make the recipe – as written. The bread was terrific! Was that enough to leave it alone? Of course not! I’m Me! LOL! So when I later learned about poolish, I absolutely had to use it with this bread.

My intention was to use a 15% Poolish.

Problem 2: “Place the dough in a lightly greased bowl, cover the bowl, and allow it to rise till it's expanded and looks somewhat puffy, about 60 to 90 minutes.” I did this in the microwave (MWO) with no drafts and a constant, decent temperature. After about 60 minutes, I eyeballed it and it “seemed like it was double.” Later, when I started using a small dough sample, I found out it doubled a lot faster, here in North Carolina in the summer. Plus, my larger, graduated rising bucket wouldn’t fit in the MWO.

NOTE: Because the yeast is eating, living and breeding over a long period of time in the poolish, you’re also growing a lot of new yeast. If the recipe calls for 2 tsp of yeast, I’m seeing you can actually use less (I’m thinking half), because Nature has increased the amount of yeast in the overall dough.

Problem 3 – Use The Right Pan!: This bread is over 400 grams, and so they call for a 9”x5” pan. I made it, by accident, in a 10-1/2 x 5-1/2” pan. It didn’t rise above the rim, and the resulting loaf was short and stubby. Use the right pan! When I made it in an 8-1/2 x 4-1/2” pan, it rose so fast and so high, it spilled out over the rim.

Because of my pound cake experiments, I bought a 9 x 5” pound cake pan, which is just right for this recipe. I did (at first) just set a timer for 75 minutes, and ended up with The Blob From Outer Space! That’s when I decided I needed a more accurate system for the length of rise times.

I started using a sample register.

Problem 4 – The Big One: I figured I’d use a 15% poolish next. That’s 74 grams flour (I used some whole wheat from the original recipe), and so 74 grams of water (also from the original). I put in about 1/32 tsp of Instant Yeast. I was going to let it sit overnight, but it doubled so quickly I just went and used it.

489 (total flour + potato) * 15% = 74 g (73.5)

It never really did triple, I think because it was lower protein whole wheat. But many people say that as long as it rises, it’s okay to use. Plus, I’ve learned you really CAN smell the poolish and experience all the complex aromas when it’s done.

Not knowing when to add the poolish, I just put all the remaining ingredients in the mixer bowl, scraped in the poolish on top of that, then added the remaining water – 265 grams – and mixed it up. I’d already learned not to put in the butter until mixing in the stand mixer. But combining the ingredients, I did by hand.

339 g (liquids) - 74 g (for poolish) = 265 g (remaining)

And this is when things came to a screeching halt.

The dough was extremely hard and dry, and with different iterations of the same recipe, I tried all kinds of different tips for kneading stiff dough. When I thought maybe it was kind of ready, I’d use the stand mixer to knead it longer. I never got any kind of “windowpane” dough, and it always ripped apart no matter how long I kneaded. But the recipe said it would be more like a “lump of clay.” Sure. When I baked it, it was awful.

Think about it: I now have (maybe) 265 g water / 415 g flour = 64% (63.8) hydration. Keep in mind that I’d taken 74 g water from the recipe, and likewise, 74 g from the flour. The original bread, including the potato flakes was 69%.

Mystery: So where did my 5% of hydration go?

In “theory,” the poolish “should have” added back in the flour and water, “keeping the hydration” the same. In reality? Not so much.

Everyone explains that making a poolish uses portions of the water and flour from the original recipe. But nobody writes out the hydration calculation before-and-after. Nobody is talking about “vanishing water or flour.” And nobody is saying anything about where that missing 5% of hydration may have gotten off to! Because...they don't think it's missing!

I'll tell you what, though: That dough was WAY more dry, WITH the poolish included, than it had any right to be! And I gave it a severe talking to, wrote a stern letter, and slapped it around with a phone book!

Almost every baker, pro or amateur warns not to add flour if the dough feels wet. And don’t add water if the dough feels dry. I took that too heart, long ago when I first started trying to bake bread. But I never calculated the before-after hydration with a poolish!

This calculation sets the poolish aside, and calculates the hydration of only the remaining ingredients and remaining water. My suspicion was that the poolish was NOT adding back the flour and water.

Solution: This is when, after many trials and failures, I finally sat down and DID calculate the before-after hydration. I just “assumed” that the water and flour and vanished into the main ingredients. It took me a long time to remember the idea of “bound,” and “free” water.

I was so fed up, bent outta shape, discombobulated, frustrated, disheartened and bamboozled, I was about to abandon the entire concept. So, as we all should do when faced with a problem, I took a nap. :-) And when I woke up, that’s when I realized I should do that hydration calculation. I also should consider adding back in some water. Rules be damned! Full Speed Ahead!

Ultimately, I found that for these kinds of breads – sandwich breads made in a loaf pan – the poolish should be made entirely separately from the recipe. The only decision-point is what percentage poolish. I’m finding that 10-15% gives me all the wonders of flavor, and the poolish fits in a small-ish jar.

Now: 10-15% of...what?!

My contention is that you’ll first choose your percentage number (experience or arbitrarily), then calculate how much flour that will be by using the Entire Amount of Flour (plus starch) in the original recipe.

If I want a 10% poolish for this loaf, then: 489 * 10% = 49 grams. Because that just “looks like” it’s pretty low, I tend to use 15%. That would be 489 * 15% = 74 grams of flour (and then, matching weight of water). And just seems more reasonable.

I made this bread with the external 15% poolish and it’s just incredible! I don’t even LIKE whole wheat bread! But the Proximate Lady prefers it. This bread is so good (even better when I used honey instead of granulated sugar), that I really enjoy it. Plus, it tastes similar to rye bread, so I don’t have to buy some different flour.

Along the way, I decided to make the poolish separately from the recipe. That kept the original recipe alone, which already made nice bread. Adding in a “new ingredient” of an external poolish, does change the original hydration. But as people do say:

Hydration is a description. It isn’t a mandate.

Since this is all still new for me, I’m thinking I may do a sort of “split the difference.” I’ll maybe use half the poolish weights of water/flour from the original, then “create” the other half with new water/flour.

Another option is to go back to extracting the poolish from the original ingredients. Add the poolish to the prepared dry ingredients, and add the water. Start trying to combine them. BUT!...have some extra water in a measuring cup (for easy pouring) standing by.

Start hand-mixing the poolish, dry stuff and liquids. When there’s no more water left, but visible dry stuff at the bottom of the bowl, just start slowly drizzling in some more water. Use it to pick up some of that dry flour-stuff. Add some extra water! Become ungovernable! Live dangerously! :-D

Continue drizzling in water and picking up dry stuff until it’s all sticky. Pat it together in a shaggy ball. Do a little hand-kneading on the counter, then autolyze for 45-min to 1-hour. Proceed from there. I’ll think about it, and maybe try it. (A little gun-shy, from all the utter failures before. But I wasn’t adding in new water back then.)

tpassin's picture

NOTE: Because the yeast is eating, living and breeding over a long period of time in the poolish, you’re also growing a lot of new yeast. If the recipe calls for 2 tsp of yeast, I’m seeing you can actually use less (I’m thinking half), because Nature has increased the amount of yeast in the overall dough.

There isn't much difference between a poolish and a sourdough starter, except that the poolish hasn't had a chance to develop lactic acid bacteria and their associated flavors (and possibly different varieties of yeast, too).

When the poolish or starter is first mixed there will be air in  the dough and the yeast will be in aerobic mode - mostly multiplying, producing some CO2 but not as much as it will later.  Once the air is mostly used up, the yeast change over to metabolize in anaerobic mode, where there is little multiplication and much more gas production.

Some of the flavor improvement from the poolish, I'm sure, comes just from the flour having been hydrated for those X hours.  You can check that by making a dough or poolish-like mixture without yeast, leaving it overnight, and then combining it with the rest of the flour and ingredients.  You will find the bread has a richer flavor. Again, that's going to be more noticeable for a lean dough.

Almost every baker, pro or amateur warns not to add flour if the dough feels wet. And don’t add water if the dough feels dry.

Like so many other pronouncements, this one should have an unstated understanding: "within reason".  If your dough is like dry sand, adding water would be a good idea.  Anyway, most experienced bakers do make adjustments in the liquid or flour as they experience the dough.  What you want to avoid is adding some water, not waiting long enough for it to be absorbed, and adding more because the dough still doesn't feel right, and so on.  Before you know it you've got a wet mess.  It's easy to fall into this mistake and overdo the additions.

Fortunately, for sandwich breads baked in loaf pans, you can get excellent loaves even if they are way wetter than expected.  Can't form a log to put into the pan?  Just scrape the dough in and smooth it down a little. Bake longer to drive off the extra water.

Hydration is a description. It isn’t a mandate

Remember also that a given hydration with all white flour can act quite differently than the same hydration with a grain-heavy dough (not to mention if the flour is rye instead of wheat).

AsburgerCook's picture


What you want to avoid is adding some water, not waiting long enough for it to be absorbed, and adding more because the dough still doesn't feel right

Yes, I found it's very important to add water very slowly, and only based on actually seeing dry flour. It's why I prefer to hand mix (squish around) the combined ingredients in a separate bowl from the mixer's bowl.

In a more rounded, larger bowl I can put in half the water and try to incorporate it into the ingredients. It almost never happens. Then I can take a sort of sticky blob and move that over to the side. I think a lot of bakers to this same thing on the counter, but I don't have a lot of room here.

Moving that blob aside inevitably leaves behind some dry ingredients at the bottom of the bowl. And that's where your comment comes into play. I'll only drizzle in maybe a tablespoon, right onto that dry stuff. Then mix the water only into that, and move that over.

It provides for the time you're talking about -- the time to ensure that the water is being fully incorporated into the dough ball.

When there's no more dry stuff left, then if there's any leftover water in my container, I'll just pitch it. Seems like this let's the dough "tell me" how much water it wants, rather than my "forcing" it with a recipe or lack of experience.

I know I'm re-inventing the wheel here, but only because of the large amount of reading or viewing I've been doing, to figure out my problems. I'm pretty sure a lot of other new bakers have similar problems, and there's a chance they may run into these discussions with a web search. I know TFL routinely shows up at the top of search results. Thanks to Floyd's good work! :-D


AsburgerCook's picture

You can see that the hydration doesn't change all that much, but the "assumption" that the poolish is contributing to the overall hydration is where I found the problem. 


Abe's picture

How can it not be contributing to the final hydration? 

AsburgerCook's picture

@Abe: Yup. That's the rub! Does the poolish add water to the system? Yes it does, like the above spreadsheet. BUT....does it add water to the initial step of combining the ingredients? No.

"Combine" was meant to be with regard to that first step: combining the dry and wet, then knead.

I may be using odd syntax, but I'm not saying that the poolish does not combine "with the dough hydration." It does add to that calculation. What I'm intending to mean is that it isn't "combining" in such a way to make the dough workable. No elasticity or extensibility -- not soft, no windowpane, no stretch, always ripping, tearing and cracking.

Every time I tried adding the poolish -- before, during, or after the autolysis -- the loaves failed. I blew off the autolysis entirely and just threw everything together. Didn't work. I finally said, "This is stupid! (or words to that effect) This needs water!"

Imagine a kitchen sponge. That's the flour for the poolish. Place that sponge in a bowl of measured water and let it sit. The sponge will soak up all the water in the bowl.

Now take that sponge and set it on a dry plate. Do you think the water will leak out onto the plate? 

If you set that wet sponge on a paper towel, the paper will "leech out" the water, drawing it from "high wetness" toward "low wetness." So, presumably, if I lay the poolish on the dry ingredients, the water will be "pulled out" of the poolish, into the dry ingredients (flour).

But if you do that, and given the nature of flour, the flour directly in contact with the poolish will try to balance the moisture. So what about the rest of the flour, that isn't in direct contact?

The other way to get the water out of the sponge is with pressure -- pressing or squeezing. What if I put the poolish in with ALL the rest of the ingredients, using a low "mix" speed? Then, the kneading process will be like "squeezing." Right?

This is by far, the most common video example or blog advice: "Put your liquids into the bowl, add the poolish, stir it around to partially mix it. Then add the flour on top. Mix it with the mixer, then knead." 

Or, as TomP has said, just as many will put the dry ingredients in, make a well, then add in the liquids. They'll often mix the poolish with the rest of the water, and dump it all into the well in the dry stuff. Then knead.

I tried both. It didn't work! One time, I think I said, "Screw it!" and let the mixer knead the dough for something like 12 minutes. It never changed, and remained dry enough to crack during the kneading rotations.

I never saw a video where someone measured out the flour, then extracted their (percentage) of poolish flour and made that poolish. Then, later, added it all back to the recipe ingredients and showed the combining process. I never saw the results of removing (same percentage) the water from the recipe water.

They SAY that's what they're doing, but I never saw anyone do it.

That's when I decided that Sandwich Loaf Bread in a pan must be different from free-formed loaves on a stone, artisan bread, or high-hydration breads. 

So up until now, the only thing that's worked is either making the poolish with non-recipe water, OR...having some extra, non-recipe water available to add after the poolish and ingredients have been combined. Either way, it needs extra water.



tpassin's picture

Every time I tried adding the poolish -- before, during, or after the autolysis -- the loaves failed

You've written this in one way or another quite a few times by now, and I don't really understand what you mean.  My experience has been that it's hard to get loaves fail - meaning I almost always get at least a decent baked loaf.  I also don't understand some of what you've been writing about difficulties combining a poolish.  I would like to get more clarity here, if we can.

Let's start with the poolish. I assume you are mixing equal weights of water and all purpose flour, and a little yeast, stirring it up by hand until the flour has all hydrated. Then you cover the container and leave it alone for some period of time like overnight.  Is this what you mean?

Under the assumption that this is your poolish, it should be pretty nearly the same as a 100% hydration starter.  That's what I usually use and what many other TFL bakers use too.  So I'm very familiar with handling it. The poolish will be a thin but not really liquid substance that is very sticky and has well-developed gluten.  Strands that you try to pull out will be very strong and elastic.  Does this describe what you have been working with?

In my experience, any dough with developed gluten will act as if it doesn't want to accept more water.  That sounds like what you have written.  But it eventually will, you just have to keep at it.

About mixing, I almost always mix by hand.  That's partly because I never can get my KitchenAide mixer to actually mix the dough, and partly because I don't like the noise and the cleanup. I think some of my problems with it happen because I mix smaller dough quantities than it needs to work right, generally 300g - 450g of flour.  Either the flour and water don't get combined enough, or the early dough wraps itself around the dough hook and doesn't ever get stretched out and pulled away from it. (Let's not get into troubleshooting my mixer here!). From here on I will talk about what I experience mixing by hand.

So, using the poolish. If you glop it into a pile of flour,  globs of poolish will grab hold of and get coated with flour and you will have an irregular mess.  You won't be able to work the poolish uniformly through the flour, breaking up those globs, because there simply isn't enough water.  Even if you more or less succeed it will be hard and take a long time.  I'm going for simplicity and low effort in most of my baking and this way is not a good match.

So your practical choices are to mix the poolish with the water first (and possibly include oil and melted butter too), or to combine the other liquid ingredients and add the poolish later.   Let's not talk about enrichments yet and just stick with water. How to choose?

Aside from whether you want to include the poolish only after an autolyse, the practical problem is whether there will be enough non-poolish water to make a dough without you having to struggle with it because it's too dry.  And that in turn will depend partly on the recipe and how much poolish there is.

I finesse the matter by always combining my water with the starter/poolish first.  To reduce the problem of having the starter clump up, I use the time-honored technique of adding a small amount of water first, stirring up the mixture to get it moistened relatively uniformly, then add the rest of the water.  I still get some clumps but they are smaller and wetter and don't cause problems when I go to mix in dry ingredients.

What if the recipe wants you to add the starter after the autolyse? I just blow that off. I don't know why that would make much if any difference; I've never read one thing about why it would, my doughs make good loaves anyway, and it's easier for me my way.  If I were using a good mixer with larger quantities I might do it differently, I don't know.

So here we are - could you (as concisely as possible) say what you have experienced that is very different?


AsburgerCook's picture


When I say "fail," it REALLY fails! It barely rises during the proofing, so tends to be flat across the top, and about level with or below the rim of the pan. I'll bake it anyway -- why not? It's already in trouble, so let's see.

Into the oven: Then, there's zero oven spring. It also usually takes 10+ minutes more to reach 195+ internal temp. 

When I take it out of the oven, the crust has many lumps, bumps, boils, and blisters. None of them edible. The crumb itself, though, is often edible, which is how I determined I really like the added flavor of the poolish. But the crust is thick, hard, dry, and nasty-looking. Very hard to slice, as well.

Because of the extra time to overcome the unincorporated extra moisture from the poolish, the bottom and sides will tend to over-bake, while the top is pale and barely browned.

Believe me: It's frickin' Awful-looking bread! And entirely inedible. Not to mention I wouldn't ever "present" a slice of this to anyone other than maybe a forum like this.

Craig -