In this LAB note
Question · Hypothesis · Method · Results · What Failed · What We Learned · What Changed at MOMS
Question
Banana and avocado occupy the same slot in a MOMS recipe and do the same structural job. In a shake built to a fixed calorie target, does it matter which one we use to make it thick?
Hypothesis
We expected this to be a texture decision with a rounding error attached. The Body slot exists to give a shake weight in the mouth, and we assumed that because every recipe is held to a calorie target, whichever ingredient fills that slot simply gets dosed until the calories match and the rest of the formula absorbs the difference. Avocado would bring some fat, banana some sugar, and the two groups would end up two or three grams apart on something. What we did not expect was that they would be the same shake by every summary number we report, and a different shake by the only two that were left.
Method
Two published isocaloric-substitution studies frame the question. Appel and colleagues (JAMA, 2005) ran the OmniHeart trial, a three-period randomized crossover feeding study in which the only variable across arms was which macronutrient carried the calories. Replacing carbohydrate energy with unsaturated fat lowered systolic blood pressure by 1.3 mm Hg - 2.9 mm Hg among hypertensive participants - raised HDL cholesterol by 1.1 mg/dL and lowered triglycerides by 9.6 mg/dL, all at matched calories. Wien and colleagues (Nutrition Journal, 2013) ran the same manipulation on a single meal using the actual ingredient: 26 overweight adults, three lunches - control, half a Hass avocado added (+112 kcal), and half a Hass avocado included isocalorically in place of other food. The added arm raised satisfaction 23% and cut desire to eat 28% over five hours. The isocaloric arm did neither. It did, however, produce a lower three-hour insulin AUC than both the control and the added meal (P = 0.04 and P = 0.05 respectively). Internally, we parsed the Body column of all 201 recipes in the formulation database, extracted each ingredient and its exact dose in grams, and split the recipes using banana without avocado (n = 32) from those using avocado without banana (n = 32), holding the 11 that use both as a separate group. We compared calories, protein, carbohydrate, fat, fiber, Score, Cal Label distribution and per-recipe carbohydrate-to-fat energy ratio across groups, and correlated dose against macros within each.
Results
Banana and avocado are the two most-used Body ingredients, and they are used exactly the same number of times: 43 recipes each, out of the 149 that carry a Body ingredient at all. Taken exclusively, the banana group (n = 32) averages 377.40 kcal, 32.16g protein, 45.38g carbohydrate, 8.30g fat, 8.15g fiber, Score 14.31. The avocado group (n = 32) averages 376.29 kcal, 32.94g protein, 31.45g carbohydrate, 14.18g fat, 8.34g fiber, Score 14.84. The two groups are 1.11 kcal apart, 0.78g of protein apart, 0.19g of fiber apart and half a Score point apart. Every number we normally use to describe a shake says these are the same shake. The two numbers left say otherwise: 13.93g of carbohydrate and 5.88g of fat. Expressed as the ratio the body actually receives, the mean per-recipe carbohydrate-to-fat energy ratio is 3.22 in the banana group and 1.09 in the avocado group - the banana shake delivers three times as much of its energy as carbohydrate as it does as fat; the avocado shake splits it almost evenly. The 11 recipes that use both land in between, at 41.21g carbohydrate, 11.53g fat, ratio 1.75. A matched pair makes it concrete. EI-W1-CLEANSE-013-V1 (Whisper Reset) and EI-W1-CLEANSE-015-V1 (Mellow Rinse) share body type, week, goal and Standard label. Whisper Reset uses Banana 60g on 300ml of almond milk and lands at 45.5g carbohydrate and 5.1g fat. Mellow Rinse uses Avocado 26.3g on almond and soy milk and lands at 28.3g carbohydrate and 18.2g fat. Their protein differs by 0.9g. Doses differ as physics requires: among recipes whose Body is a single ingredient, the median banana dose is 50g and the median avocado dose is 26.3g, because avocado carries nearly twice the energy per gram and so fills the same slot with about half the mass. And our own labelling is blind to all of it - 38% of banana recipes are labelled Filling against 31% of avocado ones, with Standard, Light and Meal Replacement distributed almost identically. Which is precisely what Wien's isocaloric arm predicted. The swap does not change how filling the shake is. It changes what arrives.
What Failed
The dose-response we had assumed. If avocado is the fat ingredient, more avocado should mean more fat. Inside the avocado group, dose against total fat correlates at r = +0.018 - flat. The formula quietly compensates: as avocado grams rise, something else fatty comes out, and dose against calories runs r = -0.402. Banana behaves nothing like this. Its dose correlates with fat at r = -0.505 and with carbohydrate at r = +0.346, so banana grams actively push fat out of a recipe while avocado grams do not pull it in. Ingredient identity, not amount, is doing the work - which is not how we had been reasoning about the slot. Two recipes break the pattern outright. EY-W4-CALM-007-V1 (Soft Landing) uses Banana 14.4g and still lands at 27.4g carbohydrate and 17.9g fat: an avocado-shaped shake with a banana in it, because the dose is small and 240ml of coconut milk plus cacao and cashew supply the fat instead. In the other direction, EY-W1-ENERGY-012-V1 (Blazing Dawn) uses Avocado 35.8g, an above-median dose, and is the highest-carbohydrate recipe in the exclusive set at 52.3g with only 10.1g of fat - below the banana group average - because 21.1g of dates and oat milk arrived alongside it. The Body ingredient sets a default. It does not set the shake.
What We Learned
The Body slot has been filed as a texture decision, and it is not one. It is the largest single macronutrient decision in the recipe, and it has been getting made on mouthfeel. Because calories, protein and fiber all come out matched, the swap is invisible in every summary field we publish and in the Cal Label we assign - and the Cal Label is the field a customer actually reads. OmniHeart's point is that an isocaloric carbohydrate-for-fat exchange is not nutritionally neutral even when the calorie sheet says nothing happened. Wien's point is that the same exchange will not feel like anything either. Both are true of our recipes at once. A banana shake and an avocado shake are the same 377 calories, the same 32 grams of protein, the same 8 grams of fiber, and two different metabolic events.
What Changed at MOMS
Body ingredient is no longer recorded as a texture field. Every recipe now carries a carbohydrate-to-fat energy ratio alongside its Cal Label, and the Body ingredient is selected against that target rather than against mouthfeel: carbohydrate-forward bodies (banana, mango, white rice, sweet potato) for Energy and Build goals, fat-forward bodies (avocado) for Calm, Balance and Renew, where a flatter glucose arrival is the point. Reviewers can no longer sign off on a Body substitution as like-for-like - swapping banana for avocado at equal calories now requires re-checking the whole formula, because 14 grams of carbohydrate and 6 grams of fat move with it. We are auditing the 32 exclusively-banana recipes for the Soft Landing case, where the honest description is that the fat is coming from the liquid and the functional slot while the banana is only there for texture; those should be reclassified rather than defended. And the 11 both-ingredient recipes are now the documented default for goals with no directional reason to pick a side, since a 1.75 ratio is a deliberate midpoint rather than an accident.
References
1. Appel LJ, Sacks FM, Carey VJ, Obarzanek E, Swain JF, Miller ER 3rd, Conlin PR, Erlinger TP, Rosner BA, Laranjo NM, Charleston J, McCarron P, Bishop LM. Effects of protein, monounsaturated fat, and carbohydrate intake on blood pressure and serum lipids: results of the OmniHeart randomized trial. JAMA. 2005;294(19):2455-2464. 2. Wien M, Haddad E, Oda K, Sabate J. A randomized 3x3 crossover study to evaluate the effect of Hass avocado intake on post-ingestive satiety, glucose and insulin levels, and subsequent energy intake in overweight adults. Nutrition Journal. 2013;12:155.
Keep reading
Previous experiment
Next experiment
Not sure which shake fits you? We match the formulation to your body type.
Find your body type