Why does my sourdough spread out flat instead of rising up? (4 causes, and the tell for each)

The four causes, and the tell for each

A sourdough that spreads sideways instead of springing upward has one of four problems, and they are distinguishable if you know what to look at. The mistake is treating "flat loaf" as a single diagnosis.

The two most confused with each other are the first and the fourth, because both produce a dough that will not hold a shape. The crumb tells them apart after the fact: over-fermented is open and gummy, under-developed is tight and dry. If you have a flat loaf sitting on your counter right now, cut it and look.

Over-fermented dough spreads because acid activates the flour's proteases

The popular explanation, that the bacteria in your starter eat the gluten, is not what the research shows. The enzymes doing the damage came in the flour bag, and the acid your starter produces is what turns them on.

Di Cagno and colleagues (2002) fermented wheat doughs with sourdough lactic acid bacteria for 4 and 8 hours at 37°C, reaching a final pH of 4.3 to 4.6, and analyzed which protein fractions were broken down. The albumins and globulins were almost completely degraded and the gliadins were partially hydrolyzed. But for the glutenins, the fraction responsible for dough elasticity, they recorded no degradation detected.

Eight hours of bacterial fermentation left the structural protein intact. So what slackens the dough?

Two things, both driven by pH. Wheat carries its own dormant proteases, aspartic proteinases active at pH 3.0 to 3.5 and serine carboxypeptidases active at pH 4.0 to 4.5, as set out in a 2021 review in Fermentation. As the dough acidifies it walks into their operating range. Acid also changes the dough's mechanics directly: Koceva Komlenić and colleagues (2010) measured extensibility on an extensograph and found it significantly decreased with acidification, comparing a control dough at pH 5.72 against acidified doughs at pH 5.05 to 5.08. Their biologically acidified doughs softened more than chemically acidified dough at the same pH.

Hydration past what your flour's protein can hold

If the dough was slack from the first mix, before fermentation had time to do anything, the problem is that you put in more water than the flour can carry. This is the cause most often misdiagnosed as over-fermentation.

Higher protein flour absorbs more water. Nobody disputes that, and here is where the actual numbers sit:

Two things in that table deserve comment. Bob's Red Mill publishes ranges while King Arthur publishes point values, so a bag of Artisan Bread Flour can legitimately arrive a full percentage point either side of what you planned around. And Pillsbury publishes no protein specification at all, only a rounded nutrition panel, which is not enough to calculate from.

Swapping a 12.7% bread flour for an 11.7% all-purpose at the same hydration is a real change in what the dough can carry, and it is the most common unnoticed variable when a recipe that worked stops working.

No authority publishes a maximum hydration for a given flour protein

There is no published table telling you the most water a flour of a given protein content can hold, and the closest thing to a hard number is lower than most home bakers would guess.

Central Milling's published certificate of analysis for their Organic Type 110 flour lists 13.10% protein against a measured farinograph absorption of 68.7%. That is an instrument determining how much water a strong flour actually takes to reach a defined dough consistency, and it lands under 70%.

We went looking for a published table mapping flour protein to a maximum workable hydration. There isn't one. Not from King Arthur, not from The Perfect Loaf, not from Modernist Cuisine, not from Central Milling. The closest published guidance is The Perfect Loaf's rule that swapping all-purpose for a high protein flour warrants an increase of 5 to 10% water, and his own worked example on the same page runs to 10 to 15%.

The absence of a table is itself the answer. The ceiling is a property of your specific bag, and the only way to find it is to drop 5 percentage points and see whether the dough behaves.

Weak shaping leaves no surface tension to resist spreading

If the dough felt strong in the bowl and relaxed into a puddle within ten minutes of being shaped, the shaping is the problem and no amount of fermentation adjustment will fix it.

Shaping does mechanical work on the gluten network. Modernist Cuisine puts it plainly: when dough is properly and evenly shaped, the gluten strands have been realigned to support expansion during final proofing and the early stages of baking. The taut outer skin is a tensioned membrane, and its job is to push back against the dough's own weight.

King Arthur describes the visible test: as you drag the dough across the bench, its skin stretches slightly and becomes taut. If you cannot see that skin form, you have not built tension. You have only made the dough round.

• Preshape, then rest 20 to 30 minutes. Skipping the bench rest is the most common error. Dough shaped straight after division fights you, and you compensate by handling it more, which degasses it.
• Shape on a dry bench, not a floured one. You need friction against the counter to build tension. Flour removes exactly that.
• Seal the seam properly. An unsealed seam is where the tension leaks out, usually within the first fifteen minutes in the basket.

Neither King Arthur nor Modernist publishes a measured surface tension figure for dough, and we could not find one anywhere. This part of bread making is craft knowledge with a mechanism attached rather than a measurement.

Under-developed gluten spreads even when fermentation is right

A dough can be perfectly fermented and still spread, because the gluten network was never built strongly enough to hold the gas it made.

The distinguishing feel is short rather than slack. Over-fermented dough is sticky, extensible and soupy; it will stretch a long way and then tear wetly. Under-developed dough resists, tears early, and feels dense rather than flowing. If you pull a walnut-sized piece thin and it rips before it goes translucent, you never passed the windowpane, and no amount of bulk time will produce structure that was not developed in the first place.

This is the most common cause in doughs that are mixed once and then left alone. Fermentation makes gas. Folding makes the container. A no-knead schedule with a single set of folds is enough for a 70% hydration white dough and it is not enough for an 80% dough or one carrying 30% whole grain.

The correction is more sets of folds early in bulk, when the dough still responds: four sets at 30-minute intervals during the first two hours does more than eight sets spread across six. Whole grain doughs also benefit from a longer autolyse, since bran needs time to hydrate before it stops acting like sand in the network. If your loaves are coming out both flat and tight, start here rather than with fermentation timing, and see why is my sourdough dense for the crumb side of the same problem.

The poke test cannot tell these four causes apart

The poke test is the standard advice for judging proof, and its own author documents three conditions under which it reads wrong. Two of those three are on the list of causes above.

The Perfect Loaf's guide to the test notes that cold dough taken from the fridge springs back quickly in a way that misleads, that high hydration doughs display little spring back, and that doughs with a high percentage of whole grains or lower quality protein show sluggish spring back.

Read those three against the diagnostic table. A high hydration dough reads as over-proofed when its real problem is water. A weak-protein or whole grain dough reads as over-proofed when its real problem is structure. The test that is supposed to be telling you about fermentation is being answered by hydration and by flour.

The test is also entirely qualitative. There is no specified poke depth, no timing, and no measured validation of it that we could locate in any literature or from any craft authority. It is a useful feel developed over many bakes, and it is worth learning. It is not an instrument, and it should not be the only thing you are using.

Percentage rise matched to dough temperature is the better test

Measure how much the dough has grown and match that number to its temperature. This is the single change that fixes more flat loaves than anything else on this page.

Tom Cucuzza of The Sourdough Journey publishes targets that vary with dough temperature, which is the part most recipes leave out entirely:

The logic behind the inversion is worth stating, because it looks backward at first glance. A warm dough is acidifying quickly, so by the time it has grown 30% it has already accumulated the acid load that a cool dough only reaches at 100%. You are not measuring gas. You are using gas as a proxy for how far the fermentation has gone.

To use it: take a small sample of dough after mixing, put it in a straight-sided container with milliliter markings, note the starting level, and end bulk when it reaches the target. Cucuzza's recalibration advice is to adjust in 5 to 10% increments based on the crumb you get, and to reduce the target by 10% if the result is over-proofed. His targets are not differentiated by flour type or whole grain content, so treat them as a starting point for a white dough. The microbiology behind why temperature changes the answer is covered in sourdough fermentation temperature.

Why a flat dough cannot be rescued in the oven

Oven spring is a window measured in minutes, and a dough that cannot hold gas at room temperature will not hold it at 220°C either.

The sequence is agreed on even where the numbers are not. The yeast keeps producing gas as the loaf warms, then dies. The gas already present expands thermally. Then starch gelatinizes and gluten coagulates, the crumb sets, and expansion stops permanently. All of it happens in roughly the first five to eight minutes.

The specific temperatures depend on who you ask, and the spread is wide enough that you should know about it:

• Yeast death: around 50°C per Modernist Cuisine, 50 to 60°C per BAKERpedia, 60 to 70°C per the American Society of Baking.
• Starch gelatinization onset: 55 to 65°C per BAKERpedia, 76°C per the American Society of Baking.
• Crumb sets: above 85°C, per BAKERpedia.

What this means practically: steam and a hot vessel help a good dough expand into the window it already had. They do not give a weak dough a window it did not have. If your loaves are flat, the fix is upstream of the oven, and buying equipment will not move it. Our methods for baking sourdough without a Dutch oven cover what the vessel does and does not do.

Sources

Related reading

• Why is my sourdough dense? The opposite failure, and it shares two of the four causes above.
• Sourdough fermentation temperature Why the percentage-rise targets change with how warm the dough is.
• Sourdough without a Dutch oven What the baking vessel actually contributes to oven spring.

FAQ