“gravestone reading ‘RIP macros’” – great job, computer!

The more I learn about nutrition, the less I think some of the popular lenses are worth. CICO is one of them, a lens so strong it blinds you to obvious realities.

But I suspect more and more that the idea of “macros” is just as useless, unless you subdivide each of the macronutrients so much further as to dilute the concept completely.

“Macros,” of course, refers to the 3 macronutrients: fat, protein, and carbohydrate. (I guess alcohol is its own macronutrient, too.) It is a very easy distinction to make when looking at the traits and composition of human food. It has also been subject to much attention from nutritional scientists, and many popular diets are based on macronutrients: low-fat diets, low-carb diets, ketogenic diets, high-protein diets..

The misleading part isn’t that these distinctions aren’t valid, they are. It’s just that, for most goals or outcomes (e.g. weight loss), they’re probably a bad heuristic. And, worse, they blind you to the higher importance of distinctions within each macronutrient.

What if saturated fats and unsaturated fats have very different roles in our metabolism? What if certain amino acids signal completely different things in the body than others? What if fructose is metabolized very differently from glucose?

Of course, we know that all these are true.

But here’s the obvious implication: What if the difference between polyunsaturated fat and saturated fat is much more important than the difference between fat and protein, and similar things are true within the other macronutrients?

What if an anti-obesity diet ideally consists of saturated fat, glucose, and gelatin (a protein that is low in BCAAs), but cuts out polyunsaturated fat, fructose, and isoleucine (a branched chain amino acid)?

I now think it’s very likely that our focus on sorting these by one dimension (fat/protein/carb) first, treating the internal distinctions as secondary, has misled us.

There is a hilarious amount of secret implications when people talk about macros. Try talking to a “If It Fits Your Macros” adherent and suggest your diet be made up exclusively of the appropriate amounts of table sugar, soybean oil, and rice protein. After all, it fits your macros!

“That’s not what I mean! Of course you should eat HEALTHY macros!”

Oh, so there are certain foods that are healthy or unhealthy, despite having interchangeable macros?

Something similar is at play in the keto community. Keto technically simply means that you’re in some level of ketosis, which only requires you eat a low enough amount of carbohydrates (and avoid excessive protein, though that’s much more lenient.)

Ketosis is, in that sense, truly just macro-dependent.

But if you suggest to ketoers that they eat a diet made largely of soybean oil, they’ll probably flip over the table. Everybody has an opinion on what a healthy ketogenic diet looks like. “That’s not what we mean!”

I don’t talk to low-fat people much (do they still exist?), but they probably wouldn’t approve of a fructose-whey-shake diet either, despite its very (heart healthy!) low fat content.

This alone should make us skeptical of “macros” unless our use case is a therapeutic ketogenic diet for an epileptic, with all other goals secondary.

So if macros are such a bad lens to view diets and nutrition through, why do we keep using them?

Probably just because they’re simple. We’re like the proverbial drunk searching for his keys under the street light, not where he lost them.

But if the solution to fat loss doesn’t neatly fit into the macronutrient (low-carb/low-fat/low-protein) heuristics, what good is it? If we have to qualify which fats, which carbs, and which proteins, why stick to macros?

Unfortunately it isn’t easy to come up with a convenient way to describe some of the modern diets that don’t neatly fit into such categories.

Example ex150: in terms of macros, it is a very-low-carb, very-low-protein, extremely-high-saturated-fat, pretty-high-monounsaturated-fat, extremely-low-polyunsaturated-fat diet. At nearly 140 characters, that description barely fits into an original-sized tweet.

And it might not even be that simple. What if it’s just the BCAAs? It could easily evolve into a very-low-carb, very-low-BCAA, moderate-protein, extremely-high-saturated-fat, pretty-high-monounsaturated-fat, extremely-low-polyunsaturated-fat diet.

Let’s hope we don’t get down to the stearic acid vs. linoleic acid or isoleucine vs. valine level.. although we might.

Notice that this is a description of the contents of the diet, but I’m not even sure that they are causal. Maybe it’s really about all the cholesterol in the cream? Or about low-variety/low-palatability? Or about timing, with insulin spiking foods largely separated from the majority of (fatty) energy intake?

It’s hard to describe reality if you don’t know what matters. You have to describe every atom.

That’s why even the most basic diet descriptions already imply some sort of mechanism or root cause.

E.g., the low-carb diet implies that lack of carbs is what makes it tick. In the beginning, ex150 was designed as a low-insulin diet – not just low in carbs, but also protein. But how confident am I that insulin is really the mechanism? Not super confident. It could even be that ex150 works due to low-carbs and low-protein, but NOT due to insulin.

So a description of a diet is pretty much always part of the map, not just the territory. Whoever picked that diet name had to make some assumptions, and shorten it to what he thought was most important, or he’d still be in the process of writing it down.

One option are somewhat unique diet names, like The Croissant Diet or the Shangri-La Diet. The Croissant Diet (TCD) especially is also sort of a troll name. It’s not really just about eating croissants. It’s neither a croissant-only diet, nor is it an anything-and-croissants diet. But the name catches your attention, because croissants are not traditionally considered a fat loss food.

TCD is a very specific diet, mostly characterized by avoidance of all PUFA and (somewhat optionally) eating a lot of refined starches, like flour, drenched in saturated fat, mostly stearic acid. He could’ve called it the “No-PUFA, high-starch, stearic acid diet” but where’s the fun in that? It’s also quite long. The Croissant Diet is pretty eye-catching and unique, and anybody searching for it will quickly find the definition.

The name Shangri-La Diet doesn’t even imply anything about nutrition, it instead refers to a fictional place. The name conveys inspiration and a dream-like or utopian state of being. The diet itself is interesting because, unlike most, it doesn’t concern itself with macronutrients – instead, it posits that flavor is the culprit, and we associate our food too much with delicious flavors. It recommends consuming 100-400kcal of flavorless sugar water or light olive oil (the tasteless stuff, not the extra-virigin salad dressing type) an hour before a meal, even holding your nose to remove any flavor. That way, you’ll be satiated by the time a real, flavorful meal rolls around.

Of course there is another danger with such names: there can be a lot of drift. Consider the Carnivore diet. If we take it literally, it’d mean only eating meat. But there are people out there eating honey on a “Carnivore diet” because honey “comes from animals” and is “ancestral.” Do we eat eggs? Drink milk? Many carnivores say yes to eggs and no to milk, but the arguments for/against aren’t easily deduced from the name “carnivore.”

Ancestral/Paleo/Primal is another example of drift: there were of course a near infinite amount of diets people could’ve consumed back then. All we know is they didn’t add refined flour and seed oils, because they wouldn’t have had those. But there might’ve been carnivorous ancestral peoples, fish-eaters, maybe some near-vegetarians. Some might have lived heavily off dairy. Some ate a lot of muscle meat, others more fat. The paleolithic era lasted over 3 million years and the earth is a big place.

The downside of these names is therefore that they can drift, or change meaning according to whomever uses the name, or tries to put his own spin on things.

ex150, my current diet, was originally just called “low-protein.” That only ever made sense to me, as I was already in a ketogenic context – that it was also near zero-carb was simply implied.

I changed the name because I wanted it to be more descriptive of what it was, not the proposed mechanism (lack of protein) – after all, I am still not sure that’s the causal factor.

But I also chose the “ex” (for experiment) part because it conveys uncertainty, and that the diet is in flux. It’ll evolve, hypotheses will be disproven.

The number 150 just describes a single part, the amount of meat (in grams) I eat per day. Since “low protein” is still my #1 hypothesis, seems fair to put it in there. That’s the part of the name that changes, to e.g. ex225lean (when I ate 225g of lean meat) or ex150sardines (guess what I ate on this ones).

It is also a unique enough sounding name that people don’t reflexively go “Oh, this is about a specific macro” like they would with “low-carb.”

In a sense, it’s almost like a serial number for an experiment, and I’ve added a few new serial numbers since: ex150deli, ex150sardines, ex150choctruffle, ex225..

Given that there isn’t an obvious, amazing way to name a diet without inviting confusion in some way, I think the conclusion is that we accept this fact, but look behind the curtain.

Obviously, most modern ketoers mean a somewhat Paleo-style keto diet. Most low-fat, plant-based dieters probably mean whole, unprocessed plant foods. Some even incorporate that into the name of their diet, WFPB (Whole Foods, Plant-Based).

We have to recognize that the map can’t be the territory. Almost all modern diets are far more complex in their advice/restrictions than pure maconutrients, and few claim causality based simply on exactly the macronutrient borders (the therapeutic ketogenic diet, unlike the more popular weight-loss keto diet, is maybe the exception).

Maybe we’ll figure out what exactly makes ex150 tick, and then we can nail down a more descriptive name. Until then, I’m hesitant, because it would be speculation and I’d rather have a serial number than a name that’s just flat out wrong.

Read More