In this LAB note
Question · Hypothesis · Method · Results · What Failed · What We Learned · What Changed at MOMS
Question
If we sort all 201 MOMS recipes by how much fruit and vegetable they actually contain, does fiber rise with produce?
Hypothesis
Fiber should track produce. That is the intuition the entire category runs on — more vegetables in the cup, more fiber in the glass — and it is why "add greens" is the first move anyone makes when they want a shake to be more filling. We expected a clear positive slope. Recipes in the top quartile for fruit and vegetable mass should carry meaningfully more fiber than recipes in the bottom quartile. Our guess before running it was a gap of roughly 3 to 4 grams.
Method
Published research. Mitchell and colleagues (2009) analyzed NHANES 1999–2002 intake data for US adults aged 19 and over and reported that consuming approximately half a cup of dry beans or peas resulted in higher intakes of fiber, protein, folate, zinc, iron and magnesium, with lower intakes of total and saturated fat. Worth noting for our purposes: US dietary guidance files dry beans and peas inside the vegetable group. In the official accounting, the beans already are the vegetable. Brick and colleagues (2022) measured total dietary fiber directly across 275 dry bean entries grown in Colorado and North Dakota using AOAC Method 2011.25. Total dietary fiber ranged from 22.82% to 30.23% of dry seed weight, composed of insoluble (54.0%), soluble (29.1%) and oligosaccharide (16.8%) fractions. Roughly a quarter of a dry bean, by weight, is fiber. Internal recipe analysis. We took all 201 recipes in the production database. Every one is a 16 oz serving, so volume is held constant by construction. For each recipe we summed the mass of every fruit and vegetable regardless of which slot it occupies — the Fruit & Vegetable slot and the Body slot both, because cauliflower in the Body slot is still cauliflower. White rice, oatmeal, seaweed, nuts, seeds and cocoa were excluded from that total. Separately we recorded grams of intact legume in the Protein slot: chickpeas, lentils, black beans, kidney beans, adzuki beans, white beans, mung bean. Tofu was deliberately not counted as a legume here, because the okara is strained out in production and soy curd is not a fiber source. We then sorted all 201 recipes by total produce mass, cut them into quartiles, and read off the fiber.
Results
Total produce mass across the database ranges from 0 g to 528 g — a spread of more than half a kilogram inside an identical 16 oz cup. Sorted into quartiles by that mass, fiber does not move. Q1 averages 62 g of produce and 7.7 g of fiber. Q2 averages 105 g of produce and 7.6 g of fiber. Q3 averages 138 g and 7.7 g. Q4 averages 219 g and 7.5 g. Three and a half times the produce, and fiber falls by 0.2 grams. The Pearson correlation between produce mass and fiber across all 201 recipes is 0.01. The correlation between grams of legume in the Protein slot and fiber is 0.64. Split by legume presence, the picture inverts. The 122 recipes containing a legume average 8.7 g of fiber on 109 g of produce. The 79 recipes without one average 5.9 g of fiber on 167 g of produce. The lower-fiber group is carrying 53% more fruit and vegetable. The dose response in the Protein slot is monotonic, and it happens at flat calories and flat protein. Zero legume: 5.9 g fiber. 1 to 39 g: 7.9 g fiber. 40 to 59 g: 8.4 g fiber. 60 g and above: 9.8 g fiber. Across those four groups calories move only between 370 and 383 kcal, and protein between 32.5 and 33.6 g. Thirty recipes clear 10 g of fiber. All thirty contain a legume. Zero exceptions. The two endpoints make it concrete. MY-W1-COOL-009-V1 "Crisp Reset" (Filling, 323.3 kcal, 30.6 g protein) carries the most produce of anything we make — cucumber 173 g, cabbage 130 g, strawberry 140 g, blueberry 55 g, papaya 30 g, 528 g in total — with egg white 240 g as its only protein. It lands at 8.7 g of fiber. MI-W3-VITAL-006-V1 "Vital Rise" (Meal Replacement, 397.2 kcal, 33.5 g protein) has an entirely empty Fruit & Vegetable slot and a Protein slot of chickpeas 30 g, lentils 35 g, egg 27.5 g and egg white 195 g. It also lands at 8.7 g of fiber. Same number, 528 grams apart.
What Failed
The headline pair is not clean, and that needs saying before anything else. "Vital Rise" also carries 10 g of chia seeds in its Functional slot, which is worth roughly 3 g of fiber on its own. The pair is an illustration, not the evidence. So we re-ran the split with every recipe containing avocado, any nut, any seed, cocoa or oatmeal removed entirely. Ninety-six recipes survive that filter. The gap held and widened: 8.5 g of fiber for the 62 legume recipes versus 5.4 g for the 34 without, on 97 g versus 157 g of produce respectively. That restricted comparison is the result we are actually standing on. Legumes are necessary but not sufficient, and our own database supplies the counterexample. WAY-W4-RENEWAL-008-V1 "Quiet Rebuild" contains chickpeas — 30 g of them — and has the lowest fiber of any recipe we make, 3.0 g. It is a half dose, its Fruit & Vegetable slot is 4 g of seaweed, and nothing else in the cup contributes. Ten legume-containing recipes sit below 6 g of fiber. The presence of a bean is not a fiber guarantee. Mass is. The effect is real but partial. An r of 0.64 leaves most of the variance in fiber unexplained by legume mass alone. Ten non-legume recipes clear 8 g: EY-W3-GUT-007-V1 "Steady Gut" reaches 9.9 g on 100 g of broccoli, 110 g of cabbage, 25.4 g of avocado and 61.3 g of banana. Produce can get there. It just needs roughly four times the mass to do it. We also checked whether our internal Score tracks any of this, because it would have been convenient if it did. It does not. The correlation between Score and fiber across the 201 recipes is −0.17 — slightly negative, and effectively noise. The Score was never built to reward fiber, and nothing in this analysis gives us grounds to claim that it does. Finally, the scope of the claim is narrow on purpose. We measured fiber. We did not measure potassium, vitamin C, polyphenols, water content, satiety or flavor, and the produce slot is doing every one of those jobs. Nothing here says vegetables are optional. It says produce mass is a bad lever if the thing you are trying to move is fiber.
What We Learned
Fiber is bought by dry seed mass, not by produce volume. A dry bean is roughly a quarter fiber by weight. A cucumber is roughly one part in a hundred. Inside a fixed 16 oz cup, that ratio decides the outcome long before anyone argues about which vegetable to use. The practical version: 65 g of chickpeas and lentils and 528 g of cucumber, cabbage and berries deliver the same 8.7 g of fiber. One of them leaves 460 grams of room in the cup for everything else. This also explains something we had noticed without understanding — why our fiber numbers looked so stable across recipes that look nothing alike. They were not stable because the formulation was tuned to hold fiber constant. They were stable because the produce slot was never the variable doing the work, and the Protein slot was quietly compensating. In the lowest-produce quartile, 84% of recipes contain a legume. In the highest-produce quartile, only 29% do.
What Changed at MOMS
Fiber is now specified at the Protein slot rather than the produce slot. Three concrete changes. We set a legume floor. Any recipe we describe as fiber-forward carries at least 50 g of intact legume in the Protein slot. Below that the target is not reliably reachable, as "Quiet Rebuild" demonstrates at 30 g and 3.0 g of fiber. We stopped using "add more vegetables" as a reformulation lever for fiber. When a recipe comes in under target, the first move is now a legume swap or a dose increase in the Protein slot. Adding produce is reserved for the jobs produce is genuinely good at — micronutrients, water, color and flavor. We are reviewing the 32 recipes currently sitting under 5 g of fiber. Twenty-seven of them have no legume in the Protein slot at all. Those are the swap candidates, and they get worked before anything else in the rotation.
References
Mitchell DC, Lawrence FR, Hartman TJ, Curran JM. "Consumption of Dry Beans, Peas, and Lentils Could Improve Diet Quality in the US Population." Journal of the American Dietetic Association, 109(5):909–913, 2009. | Brick MA, Kleintop A, Echeverria D, Kammlade S, Brick LA, Osorno JM, McClean P, Thompson HJ. "Dry Bean: A Protein-Rich Superfood With Carbohydrate Characteristics That Can Close the Dietary Fiber Gap." Frontiers in Plant Science, 13:914412, 2022.
Keep reading
Previous experiment
Not sure which shake fits you? We match the formulation to your body type.
Find your body type