In this LAB note
Question · Hypothesis · Method · Results · What Failed · What We Learned · What Changed at MOMS
Question
Every MOMS plan runs four weeks, and the number implies escalation. Across all 201 formulations, mean protein moves 1.1 g from Week 1 to Week 4 and mean calories stay inside a 17 kcal band. If the macros are not what the four weeks are changing, what is?
Hypothesis
We expected the answer to be progression. Our own internal shorthand described Week 4 as the built-up version of Week 1 — more protein for a body that had adapted, or a calorie taper, or a fiber ramp. If that were true, the weekly means should separate cleanly, and a Week 4 recipe should read as a Week 1 recipe with the dials turned. Protein up, calories down, week over week.
Method
We grouped all 201 formulations by Week (1 through 4) and computed mean kcal, protein, carbohydrate, fat, fiber, and Score for each group. We then took one body type end to end — EY, 30 recipes across the four weeks — and measured two quantities against each other: the spread of its macros, and the number of distinct whole-food ingredients appearing across its Liquid, Body, Fruit & Vegetable, Protein, and Functional columns. We repeated the ingredient count week by week, to see whether Week 4 was drawing on Week 1's roster or a different one, and ran the same count across all eight body types. Finally we correlated per-recipe ingredient count against Score, to test whether our own scoring system was tracking variety at all. The published context comes from two sources. McDonald and colleagues (mSystems, 2018), working with 1,596 American Gut participants, compared people eating more than 30 types of plants per week against those eating 10 or fewer, and found the two groups separated on gut microbial composition — plant count, not macronutrient split, was the axis that divided them. Heiman and Greenway (Molecular Metabolism, 2016) had argued the same direction earlier: microbial richness tracks the number of distinct plant substrates a diet supplies, not its calorie or protein totals.
Results
The macro envelope is flat, and that turns out to be the design rather than a defect. Weekly means across all 201 recipes: Week 1 is 375 kcal and 32.4 g protein, Week 2 is 373 kcal and 33.3 g, Week 3 is 384 kcal and 33.2 g, Week 4 is 367 kcal and 33.5 g. Total protein drift across the whole program is 1.1 g. Every one of the 201 recipes is served at 16 oz. Inside a single body type the same pattern holds. EY's 30 recipes span 30.5 to 35.1 g of protein — a 4.6 g range over four weeks — while drawing on 35 distinct whole-food ingredients. The roster is where the movement is. EY Week 1 uses 21 distinct ingredients, and Week 4 also uses 21 — but seven of Week 4's (taro, pear, blueberry, flaxseed, peach, cacao, spinach) never appear in Week 1 at all. The count held constant while a third of the list turned over. The extreme pair makes it visible. Sunrise Surge (EY-W1-ENERGY-001-V1) is Coconut Milk 195 ml, Banana 26 g, Beet 30 g, Carrot 25 g, Egg 55 g, Egg White 170 g, Black Beans 80 g, Turmeric 0.2 g, Cashew 5 g, landing at 393.5 kcal with 34.5 g of protein. Quiet Recovery (EY-W4-CALM-001-V1) is Soy Milk 150 ml, Pumpkin 70 g, Taro 25 g, Pear 50 g, Dates 20 g, Egg White 220 g, Turmeric 0.5 g, Cashew 15 g, landing at 386.3 kcal with 33.3 g. Seven kilocalories and 1.2 g of protein apart, they share three ingredients out of fourteen. A single shake averages 6.4 distinct ingredients. A single EY week supplies 21.
What Failed
Our Score does not see any of this, and we only found that out by checking. Correlating per-recipe distinct ingredient count against Score across all 201 recipes returns r = -0.02. A six-ingredient shake and a nine-ingredient shake are, as far as the Score is concerned, indistinguishable. The variable this entry has just spent four paragraphs arguing for is the one variable our scoring system has no term for. That blind spot has already cost us something. Roster breadth ranges from 27 distinct ingredients (MY, 19 recipes) to 47 (WAY, 26 recipes) — 1.42 against 1.81 new ingredients per recipe. MY's mean Score is 14.6 and WAY's is 15.9, but the two groups differ in recipe count and goal mix, so we cannot read that gap as diversity causing it. The honest statement is narrower and worse: MY's roster is a third narrower than WAY's, and nothing in our system flagged it. Rotation did not improve every tracked nutrient either. Quiet Recovery brings in seven ingredients Week 1 never sees and carries 5.5 g of fiber against Sunrise Surge's 9.4 g. New plants and more fiber are not the same move, and we should not have assumed they travel together.
What We Learned
The four-week structure is a rotation schedule, not a progression. Holding calories, protein and volume near-constant is what makes rotation possible. If Week 4 also had to hit a different macro target, the ingredient roster would be constrained by arithmetic rather than free to turn over. The flatness we mistook for a missing feature is the enabling condition. And the variable it frees up is the one the microbiome literature keeps pointing at. A plan that puts 35 distinct plants in front of someone over four weeks while never moving protein more than 5 g is doing something a macro table cannot show you.
What Changed at MOMS
Plant-species count is now a tracked formulation metric rather than a byproduct. Every body type's four-week set carries a floor of 35 distinct whole-food ingredients. MY's 27 is flagged and is being widened before its next revision. No week may repeat more than 60% of the previous week's ingredient roster. EY Week 4, with seven entries Week 1 never uses, clears this. Several other sets did not, and are being reworked. The Score keeps its macro and goal-fit terms unchanged. We are not going to reweight it to chase a correlation we just measured at zero. Diversity is reviewed as a separate gate before a recipe set ships, standing beside the Score rather than inside it.
References
McDonald D, Hyde E, Debelius JW, et al. American Gut: an Open Platform for Citizen Science Microbiome Research. mSystems, 3(3):e00031-18, 2018. | Heiman ML, Greenway FL. A healthy gastrointestinal microbiome is dependent on dietary diversity. Molecular Metabolism, 5(5):317-320, 2016.
Keep reading
Previous experiment
Not sure which shake fits you? We match the formulation to your body type.
Find your body type