The Sourdough Experiment: design of experiments (DOE), explained with a loaf of bread
Six settings, eight bakes, one best loaf. Find it by gut feel or with a designed experiment.
Six settings took eight bakes. Your formulations have forty.
How the experiment works
You control six settings: hydration, starter, salt, wholemeal flour, bulk temperature and bulk time. Each bake scores volume, crumb, sourness, crust color and ease of shaping, combined into a loaf score out of 100.
What drives the loaf
The model tracks yeast gas, bacterial acid and gluten strength. Above about 28 °C the yeast slows while the bacteria keep going. Salt slows both. Wholemeal soaks up water and adds tang. Leave the dough too long and it spreads instead of rising.
Loaf score = weighted geometric mean of the five results, so a poor result on anything you care about pulls the whole score down
Every bake varies, like a real kitchen: about 10 points per result at home, about 5 in the test kitchen.
Gut feel or a designed experiment
In gut-feel mode you change anything between bakes. In designed mode you pick two or three settings, and the tool plans five bakes, fits a model and suggests three more.
Bakes 1 to 4: the corners of a half-fraction design. Bake 5: the center. Bakes 6 to 8: the model's suggestions.
Each setting is high in half the bakes and low in the rest, so every bake informs every setting. Some effects look identical after four bakes. Bake 6 tells them apart.
Why one-at-a-time fails
Changing one thing at a time misses interactions. The best bulk time is about 8 hours at 20 °C but under 5 hours at 30 °C, so testing time at the wrong temperature points the wrong way. It also trusts single bakes, and one unlucky bake can mislead you.
Across 2,000 simulated home-kitchen games, a designed experiment reached 91% of the best loaf on average, one-at-a-time 84% and random changes 60%. In the unluckiest 10% of games, design still reached 82%, against 66% for one-at-a-time.
What you study matters most. A design on the wrong settings loses to careful one-at-a-time on the right ones.
What the simulation does not do
This is a simplified teaching model. Effect directions come from the research below, but the numbers are illustrative and not fitted to real bakes. Proof, shaping and baking are held constant, sourness is a single number and salt is not capped for taste.
Worked example: the warm-day over-proof
The game's starting loaf, with everyday flour and default scoring:
- Start: 70% hydration, 20% starter, 2.0% salt, 10% wholemeal, 7 hours at 28 °C. Flat, very sour and sticky: 30% of the best loaf.
- Best for this flour: 78% hydration, 18% starter, 2.2% salt, 15% wholemeal, 5 hours 45 minutes at 25 °C.
- Cut the time to 4 hours 15 minutes: 66%.
- Cool the dough to 20.5 °C: 70%.
- Add water to 77% hydration: 42%.
- Cool the dough and add water together: 97%. That jump is an interaction, and one-at-a-time testing struggles to find it.
Each game uses one of 25 flour batches, so the best recipe shifts every time you play.
From the game to Uncountable
- Choosing what to study: models trained on your past results show which ingredients and conditions drive each property.
- Planning the runs: a proprietary design gets the most learning from the fewest experiments, inside your ingredient limits.
- Picking the next run: suggested experiments come with predicted results and uncertainty.
- Keeping what you learn: every result stays linked to its formulation and sharpens the next model.
See the full workflow on the DOE and predictive tools page.
FAQs
A way to plan tests so a few runs show how several factors affect a result, including how they interact. You change several factors together in a balanced pattern, then fit a model.
It misses interactions and trusts single comparisons that noise can mislead. Here, the best bulk time drops from about 8 hours at 20 °C to under 5 hours at 30 °C, so testing time at the wrong temperature gives the wrong answer.
A full factorial tests every combination of high and low settings. A fractional factorial tests a balanced subset: here, four bakes cover three settings. The cost is that some effects are confounded until you add runs.
No. Effect directions come from published sourdough research, but the numbers are illustrative. Use it to learn about experiments, not to pick a recipe.
Every bake gets random variation, like a real kitchen. That is the point: one bake can mislead you, and a design that uses all your bakes is more reliable.
References
- Gänzle, M.G., Ehmann, M. & Hammes, W.P. (1998). Modeling of growth of Lactobacillus sanfranciscensis and Candida milleri in response to process parameters of sourdough fermentation. Applied and Environmental Microbiology 64(7), 2616-2623. pmc.ncbi.nlm.nih.gov
- Gänzle, M.G., Loponen, J. & Gobbetti, M. (2008). Proteolysis in sourdough fermentations: mechanisms and potential for improved bread quality. Trends in Food Science & Technology 19(10), 513-521. doi.org
- Thiele, C., Gänzle, M.G. & Vogel, R.F. (2002). Contribution of sourdough lactobacilli, yeast, and cereal enzymes to the generation of amino acids in dough relevant for bread flavor. Cereal Chemistry 79(1), 45-51. doi.org
- Belz, M.C.E., Ryan, L.A.M. & Arendt, E.K. (2012). The impact of salt reduction in bread: a review. Critical Reviews in Food Science and Nutrition 52(6), 514-524.
- Miller, R.A. & Hoseney, R.C. (2008). Role of salt in baking. Cereal Foods World 53(1), 4-6.
- Noort, M.W.J. et al. (2010). The effect of particle size of wheat bran fractions on bread quality: evidence for fibre-protein interactions. Journal of Cereal Science 52(1), 59-64.
- Gómez, M. et al. (2003). Effect of dietary fibre on dough rheology and bread quality. European Food Research and Technology 216, 51-56.
- Kim, J.H., Maeda, T. & Morita, N. (2005). Application of polished-graded wheat grains for sourdough bread. Cereal Chemistry 82(2), 144-151.
- Arendt, E.K., Ryan, L.A.M. & Dal Bello, F. (2007). Impact of sourdough on the texture of bread. Food Microbiology 24(2), 165-174.
- Purlis, E. (2010). Browning development in bakery products: a review. Journal of Food Engineering 99(3), 239-249.
- Calvert, M.D. et al. (2021). A review of sourdough starters: ecology, practices, and sensory quality with applications for baking and recommendations for future research. PeerJ 9, e11389. peerj.com
- Scanlon, M.G. & Zghal, M.C. (2001). Bread properties and crumb structure. Food Research International 34(10), 841-864.
- Cauvain, S.P. (2015). Technology of Breadmaking, 3rd edition. Springer.
- Derringer, G. & Suich, R. (1980). Simultaneous optimization of several response variables. Journal of Quality Technology 12(4), 214-219.
- Box, G.E.P. & Wilson, K.B. (1951). On the experimental attainment of optimum conditions. Journal of the Royal Statistical Society B 13(1), 1-45.
- Czitrom, V. (1999). One-factor-at-a-time versus designed experiments. The American Statistician 53(2), 126-131.
Model reviewed on 13 October 2026. Model spec version 1.0.
Disclaimer
The Sourdough Experiment is a free teaching tool built on a simplified model of sourdough baking. Results are illustrations, not measurements or baking, nutritional or professional advice, so do not treat the simulated best loaf as a recipe.
We have taken care over the model, but the tool is provided as is. To the extent the law allows, Uncountable gives no warranty that the results are accurate, complete or suitable for any purpose, and accepts no liability for any loss that results from using them.
