Gluten-Free Bread Is Built on Guesswork (For Now)
If you've ever followed a gluten free bread recipe exactly and still ended up with something gummy, dense, or structurally confused, I want to offer you some relief: it's probably not your fault.
Here's something worth saying out loud: GF baking is a young field. The science, the technique, the accumulated knowledge that conventional baking has built over centuries… GF bread doesn't have that yet. Wheat baking has generations of iteration behind it. GF baking is still in its first few decades of serious development, and a lot of that development has happened in home kitchens and small bakeries by people who needed to eat safe food and figured things out as they went. That matters. It means the knowledge base is real and hard-won, and it also means there are still significant gaps.
GF bread recipes are, in large part, built on intuition and iteration rather than principle. People find something that works and they write it down. That's not a flaw, that's how early-stage knowledge gets built. But "works" and "works because of a sound understanding of what each ingredient is doing" are very different things. And when you're troubleshooting a failure, only one of those is actually useful to you.
I've been spending a lot of time lately reverse-engineering GF sourdough recipes, building a spreadsheet that tracks every ingredient by its functional properties, hydration absorption, protein content, and how often it appears across formulas. What I'm finding is a field that has made real progress through trial and error, but is still largely learning as it goes. And a lot of what it's converged on so far is oat flour, sorghum, and tapioca starch. Not necessarily because those are the optimal choices, but because they've been iterated on enough that they work.
That's not a criticism. It's a starting point. And it's exactly the kind of starting point worth examining.
Starches Have Jobs
Here's the thing about starches: they are not interchangeable. Not even close. Tapioca, potato, white rice, and cornstarch all behave differently under heat, under fermentation, and under the mechanical stress of shaping. Using them interchangeably is like swapping bread flour for cake flour and wondering why your loaf didn't have any structure.
Tapioca starch is elastic and stretchy. It helps hold a dough together under fermentation pressure and contributes to chew in the final crumb. It's doing structural work.
Potato starch gelatinizes at a lower temperature and creates a soft, tender crumb. It's fluffy. It's great in yeasted breads where you want lift and lightness. But it's not a structural ingredient in the way tapioca is, and using it in a sourdough where you need the dough to hold tension through a long ferment may actually be working against you.
White rice flour is mild and relatively neutral, but it has active amylase enzymes that will degrade starches over extended fermentation -- which is exactly the environment you're creating in a sourdough. Heat treating rice flour before use stabilizes it. Most recipes don't tell you that.
Psyllium husk powder is not really a starch at all. It's a binder and structural component. It absorbs water aggressively and creates the scaffolding that holds CO2 during fermentation. Too little and your dough won't hold gas. Too much and you get gumminess. It has a pretty narrow functional window.
Oat flour is a nutty, sweet, and almost milky tasting flour. As anyone who has eaten anything with oats knows, it’s got its own distinct flavor. Oat flour is high in protein and is fairly elastic, which makes it a great addition to breadmaking with alternative grains. It’s one of the more studied alternative grains, which means there’s a lot more readily available recipes that are easily replicable with it. It’s higher in absorption capacity, which means it can lend to a drier crumb if your hydration is on the lower side.
Sorghum flour is another common grain in alternative breadmaking. Like oat flour, it’s got a higher protein content. This helps, along with binders like psyllium husk, to trap CO2 during fermentation. That CO2 is vital to creating a light, airy bread. It’s not as elastic as oat flour, but it’s closer in flavor to wheat. Sweet yet earthy.
The point is: every ingredient in a GF bread formula has a job. When you understand the job, you can troubleshoot. When you don't, you're just hoping.
The Fermentation Problem
Most GF sourdough recipes use an enormous amount of starter. Like, a lot. And I've been sitting with a question: is that actually necessary, or is it a compensation strategy?
This is one of the places where I think GF baking is still catching up. In conventional sourdough, there's a well-established body of knowledge around inoculation rates, fermentation timelines, and how those variables interact with dough structure. That knowledge took decades to develop. GF sourdough is working from a much thinner foundation, and in the absence of established principle, more tends to feel safer than less.
Here's my working theory. If your dough structure isn't built to capture and hold CO2 efficiently, more fermentation isn't going to fix that. More CO2 production past the point your structure can hold just means collapse, or degradation of the starch matrix you're relying on to replace gluten. Starch is structure. It's not gluten, and it's not psyllium, but it is load-bearing. And running aggressive fermentation through a starch-dependent dough risks breaking down the very thing holding it together.
The same principles of fermentation that apply to wheat bread still apply here. Yeast multiplies at the same rate. CO2 is CO2. The question is whether your structure can capture what you're producing.
I think a lower inoculation rate with a longer, slower ferment (and a well-built dough structure going in) might actually perform better than the high-starter approach most recipes use. That's one of the things I'm planning to test once my sorghum starter is established and my flours arrive this week.
What The Recipe Data Actually Shows
Across the sourdough recipes I've analyzed so far, hydration in the final dough ranges from about 103% to 145%. That's a huge spread. But it doesn't feel crazy to me. Wheat bread hydration varies by about the same proportion, just for different reasons. In wheat, hydration tracks with protein content and desired crumb structure. In GF bread, hydration varies because every flour blend has a different absorption profile. You're essentially re-engineering the matrix every time you change your flour combination.
What I'm also noticing is that certain ingredients only show up in starters or sponges, not in the final dough. That matters. An ingredient acting as a fermentation substrate is doing a completely different job than an ingredient acting as a structural component. Recipes don't always make that distinction clear. Part of that is because the field is still developing the language and framework to make those distinctions. The knowledge exists in pieces. It just hasn't been systematized yet.
The oat/sorghum combination shows up consistently across formulas, usually with a tapioca component. My read is that oat and sorghum are gravitating to the top because they're the closest GF analogs to wheat in terms of protein content and flavor. People are pattern-matching, consciously or not. Whether that pairing is genuinely optimal or just well-iterated is one of the questions I'm trying to answer.
Why This Matters For You
If your GF bread keeps failing, the answer is probably not "follow the recipe more carefully." It's more likely that you don't yet have a clear picture of what each ingredient is supposed to be doing, which means you can't diagnose what went wrong. And honestly, that's not entirely on you. The recipes themselves often don't explain it either, because the field is still working that out. In a lot of ways, professionals don’t even fully know. That’s kind of why this entire sphere of the baking ecosystem is so fascinating to me.
Dense crumb might mean your structure couldn't capture fermentation gases… But that could be a psyllium issue, a starch issue, a fermentation issue, or a shaping issue. Gumminess is often excess tapioca or psyllium past their functional threshold, or underbaking, or both. A dough that spreads instead of holding its shape is a structure problem, not a hydration problem, though hydration can make it worse.
GF baking is going to keep getting better. The knowledge is accumulating. The goal of everything I'm building with Breaducated is to be part of that accumulation: to help close the gap between what we know and what we can actually teach. Not better recipes. A better understanding of why the recipe works, so you can fix it when it doesn't.
More to come as the R&D gets underway. I’ll be starting a couple GF starters this week. I’ll keep you posted on my journey.
