. . Most of QuidNYC's notes still apply here; I would suggest reading his before reading mine . .

Changes from QuidNYC's Superfood

1. changed oil composition -- MCT gave me a lot of problems that never went away
2. added blueberries for taste, another source of Vitamin K, phytochemicals
3. added mangos for taste, another source of Vitamins A and C
4. added parsley for a source of vitamins A and K
5. deleted most of the nutrients listed for whey protein--there is little evidence that those numbers were accurate
6. made several changes to the protein powder aspect of the recipe; see the section on protein later
7. increased the amount of potassium. There is evidence that increased potassium intake is beneficial, some evidence that "low" amounts are harmful, and no evidence that there even exists an upper limit for healthy people; therefore, given that potassium citrate is cheap, adding more is simply insurance against deficiency for extremely little risk/cost. I added the equivalent of a few cents per day.
8. increased salt to keep pace with potassium.
9. removed soy lecithin - see below for details
10. increased masa flour to help bump kcal up around 2000
11. added fish oil capsules
12. the combination of the above changes added a lot of kcal...losing weight is not my goal, and the amount of kcal in QuidNYC's Superfood is not sustainable for me

Safety/Completeness

(vocab notes on nutrients:

1. RDA = Recommended Daily Allowance: strong evidence that this level will meet the needs of at least 97% of the population
2. AI = Adequate Intake: not as much evidence exists as RDA, but enough that at least 50% of the population should have their needs met
3. UL = Tolerable Upper Intake Level: the highest level of daily intake that is likely to pose no risk to almost all of the population)

In general, I tried to make it so that there was no single source of any nutrient if reasonably possible. For example, previously, Vitamin A was provided solely by a multivitamin (if the hypothetical contribution by the whey protein is not considered). While this is not necessarily bad, especially considering an analysis of the particular multivitamin in question suggests that it used multiple sources itself--beta carotene and retinyl acetate--the engineer in me makes me uncomfortable with one "point of failure", so to speak. As much as I want to trust a given source's information, such as a manufacturer via their Nutrition/Supplement Facts, or the USDA database, I've found that some sources are less reliable than others, and therefore having a particular nutrient be spread across multiple ingredients means minimizing the chance that one of those ingredients is actually supplying a radically different amount of that nutrient than the information suggests. If I can't be extremely confident about how much of a nutrient is in a particular ingredient, I look at the totals of the ingredients that I AM confident of and check for two things: (1) if the questionable ingredient were removed completely, would we still hit RDA/AI? and (2) if the questionable ingredient does in fact have the level of [nutrient] claimed, would we hit UL? This is why generally my tweaks have tended toward getting nutrients above RDA/AI but well short of UL, if one exists.

Protein - Soy Free

I changed the main protein source to a 50/50 mix of whey concentrate (not isolate) and casein. Whey concentrate is a very fast-absorbing protein, and whey isolate is a purified form that is even faster-absorbing than concentrate. This massive bolus of protein causes a big insulin spike. Insulin spikes make me feel weird, and I mean that both in the actual physical sense (increased heart rate, nausea) and the "health anxiety" sense. There seems to be a lot of debating among broscience types about whether continual insulin spikes due to large bursts of amino acids hitting the blood stream are not as harmful as insulin spikes due to sugar. Even putting that aside, it still makes me physically feel uncomfortable to have a big dose of insulin on board, possibly because it drives down my blood sugar to hypoglycemic levels (the insulin spike happens very quickly, whereas the glucose deriving from the carbs in the masa take a lot longer to get into the blood stream; there is not a lot of sugar from the fruit so I'm thinking that has a small effect). Casein, on the other hand, is very slow-absorbing, and does not cause insulin levels to spike. Having this mix, as opposed to 100% whey, helps blunt the big insulin release. (Incidentally, 50/50 whey:casein is closer to the mixture of these proteins in human breast milk, as opposed to 20/80 in cow's milk. Without getting into weird Oedipus complex territory, I feel more comfortable matching my protein needs along the lines of human evolution than cow.)

I also chose a particular brand (True Nutrition) for the protein because they use sunflower lecithin rather than soy lecithin. By choosing sunflower lecithin-instantized protein, and removing the soy lecithin as a separate ingredient, I've completely (as far as I can tell) removed soy from the recipe. I get that the case against soy is not bulletproof, and that the amount of soy added to protein powders is not great. But the idea of this recipe is to, in theory, be something that could be consumed indefinitely and be the sole source of energy and nutrition in one's diet, which means you want it to be as solid as possible. Part of my philosophy and motivation behind this recipe is to optimize. Optimization, at a certain point, becomes a game of incremental tweaks and diminishing returns. Honestly, removing the small amount of soy in the original version probably won't have any appreciable health impact. Having said that, it wouldn't be optimizing if I were only thinking in terms of the best cost/benefit ratio; that is a goal of other recipes, but not mine. If I have some reason to believe I can improve the recipe, and it doesn't add undue complexity or inordinate cost, then I'll do it. This is, after all, a "super" recipe :)

Finally, I slightly lowered the overall protein powder contribution. The other changes added protein that made up for the decrease.

Phytates

One last thing about phytates. QuidNYC addresses this in his notes, but I wanted to add that the nutrients that are known to chelate with phytic acid (zinc, iron, calcium, magnesium, niacin), and phosphorus itself--phytic acid is a form of phosphorus--are in abundance in this recipe anyway. I can't find even ballpark figures as to the rate at which phytic acid binds to these things, but I think it's safe to say there's enough "left over" after the mean old phytates do their dirty work to still meet AI. This was another factor in my nutrient tweaks--to make sure that whatever effect phytates do have is hopefully mitigated by having enough "extra" that we are still nutritionally complete.