Ash:

While the plural of "anecdote" is not "fact", there exists enough anecdotal evidence that I felt it necessary to keep an open mind.

François:

"I was certain that there was some kind of treshold to trigger the real benefits of low-carbing. ... I guess this treshold must have variations from individual to individual."

I believe you're correct...and I'll be exploring metabolic individuality in detail in future installments.

pzo:

Much of what passes for "nutrition information" is just provocation marketing.  Make bold claims that YOUR CURRENT DIET IS KILLING YOU!!11!, throw in a couple half-assed scientific references to an obscure biochemical pathway with no real-world relevance, and you too can sell a $39.95 e-book.

That said, my purpose here is not to advocate low-carb for everyone: I'm not even low-carb myself!  I'm suggesting that, based on the evidence, "mostly-sorta-low-carb" is a bad place to live: you're getting most of the negatives without most of the benefits.  Result: either go keto or don't.

Also, it's not clear that alcohol has the same effect as sugar or maltodextrin: see, for instance, this study.

Yes, going keto will leave you with a much smaller glycolytic reserve for sprinting, on a bike or anywhere else.  You have to decide for yourself if that's worth it.  And "feeling spacey" is keto-adaptation: again, being "mostly-sorta-low-carb" generally stops you from keto-adapting.

Sofie:

I'm not ready to dismiss fine liquor as nerga, but I take your point.  We all have to decide what's most important to us.  Relevant: "Just Say It," Jim Steel.

Anniee:

I suspect it'll matter, but I'm not sure by how much.  My guess is that it's is likely to be an insulin-mediated response and fructose doesn't stimulate insulin secretion, so other carbs would be at least as bad or worse -- but I'm not going to invest much in that hypothesis.

Frankly, a combination of measurement error and metabolic individuality means you'll most likely have to experiment on your own anyway.

Sherwood:

"Wonder what the results would be if various long chain carbs were used instead?"

As I said above to Anniee, I suspect they won't be much different unless they're so long as to have a meaningfully lower GI.  However, if they're so long that they're soluble fiber (i.e. indigestible except by colon bacteria) instead of carbohydrate, they probably won't have the same fattening effect.

"It would appear that at low levels carbs are acting as a catylist, or are an essential starting material for some metabolic process. No carbs means that some process is blocked, and another process is used.  What happens to the extra calories?"

I agree.  One guess would be peroxisomal oxidation, which would waste much of the energy as heat, but I don't know. 

Note that some of the studies in previous installments measured fecal losses, oxygen consumption, etc...unfortunately, this one did not.

More soon!

JS

Paul Lee:

"Begs the question of how to loose weight on high carb without starvation/fasting or spending hours on a treadmill."

1. Restore metabolic flexibility if possible, through a combination of moderate exercise over time

2. Pay close attention to nutrient density, as hunger is primarily nutrient-driven and it's easy to consume lots of empty starch

"If you are largely an inactive coach potato then zero/vlc will be the only way to go"

I think that's usually true: if you're not fortunate enough to be naturally skinny, and you're not willing to exercise to keep your met flex, you'll probably need to force your RQ down with VLC/ZC.

MargaretRC:

"That explains nicely why my husband was able to lose a fair amount of fat when he went very low carb and exercised a lot, but gained it all back as soon as he reduced exercise and allowed a relatively small amount of carbs (mostly wheat, not sugar) back in."

That's a double whammy: he lost the advantage of VLC, and he probably lost some met flex due to no longer exercising.  Maybe even a triple whammy, given how sensitive some people are to gluten.

I'm glad you find my work valuable!

Paleobird:

P*atarians are remarkably resistant to the introduction of facts.  If you want a laugh, read this and note the lack of response.

Chris Highcock:

As I noted above, this experiment doesn't tell us where the extra energy went — though other experiments measured it, or attempted to (e.g. Part VI). 

We know it didn't vanish, because physics is true — but that's not helpful.  Analogy: If my toilet overflows, it's not helpful to say "Your problem is obvious: more water is entering your toilet than leaving it."  We're only interested in how to get more water to leave the toilet, and "how?" is a much more pressing question than "why?"

Therein lies the difference between physics and the naive application of CICO.  Physics says "In a closed system, energy is neither created nor destroyed."  CICO says "Calories out has a direct linear relationship to exercise, calories in (as approximated by Wilbur Atwater based on the average of a series of empirical measurements in the 1800s), and nothing else."

 
Jacquie:

First, optimum health and fat loss are not exactly the same goal.  They're usually congruent — but I view fat loss as a consequence of good health.  If your body is functioning properly and well-nourished, it should be happy to let go of some fat stores — whether VLC or not.

That being said, not everyone has good metabolic flexibility, or can restore it quickly via exercise (there's a study I can't find now showing that people who became T2D at an early age did not restore met flex with exercise, unlike people who became T2D later on), in which case VLC can help.

Moving on to the type of carb: I don't think we can know, based on this experiment, whether it's only the refined sugar, or whether other carbs have the same effect at 10%.  If I had to guess, I'd say what I said above to Anniee: any sugary or starchy carb will have the same effect.  Exception: green veggies that aren't sugary/starchy root veggies (onions, carrots, potatoes), whose few carbs tend to be mostly theoretical and used in digestion.  (I tend to agree with Paul Jaminet that veggie carbs don't count.)

That being said, it's something you have to experiment with on your own, because my thoughts on a rat study don't trump your own direct experience.

Finally, let me be clear: the point of this article is not that everyone should eat VLC or zero-carb!  (I don't prescribe a specific carb intake in "Eat Like A Predator" beyond "don't eat lots more carbohydrate than you need to.")  The message I'm hoping to communicate is this: stay out of the no-man's-land between 5% and 15% carbs.  (Higher if you're very physically active, and carbs from green veggies don't count.)

Does this help?

More soon!

JS

Renaud, Sean:

As I said above to Sherwood, we don't know where the extra energy went.  Re: your second question, I see no evidence that growth was stunted by the zero-carb diet.

1. Zucker rats are sexually mature at six weeks.

2. Note the abstract: "The subscapular and epididymal fat pats were increased 42% and 44%, respectively, in animals consuming the 10% sucrose diet compared to all other groups."

3. We don't have figures for total weight...but a mature non-fatty Zucker rat weighs perhaps 400g, and the rats gained 20g in four weeks, so weight gain during the four-week experiment (20g) was perhaps 5%.  Even if they were half adult size, we're still talking about a difference between 10% weight gain on the most fattening lab chow possible and zero weight gain. 

A 5-10% difference doesn't qualify as "stunted".

Johnnyv:

No, rats aren't human -- but they're much more like humans than mice are, being naturally omnivorous, so I take rat studies more seriously than mouse studies.  (Exception: fatty Zucker rats, and anything else bred to be obese.)

The only human study I'm familiar with that features similarly extreme dietary compositions is Keckwick and Pawan 1956 -- which comes to similar conclusions, though I'm reluctant to rely on it because it was so short.  (Note that the "refutation" of Keckwick and Pawan (Hood 1970) only ran for eight days, therefore featuring exactly the same problem.)

People like to cite one and disparage the other, depending on their preconceived bias, but the only difference I see is that Keckwick and Pawan were honest about the real-world problems of getting people to stick to an extreme diet, even when they're hospital inpatients.  (Therefore, Keckwick and Pawan is a "metabolic ward" study, because all that means is the subjects slept in a hospital.  It's no guarantee they didn't cheat: there's plenty of food in hospitals.)

con:

That's why I wrote this article: it seems to jibe with many people's observed experience.

Congratulations on losing 95# and keeping it off!  Statistically, almost no one does that without gastric bypass.

BawdyWench, Jacquie:

See my reply above to Jacquie.

neal matheson:

I'm certainly "low-carb" compared to average consumption and the nutrition guidelines -- but I don't make any effort to restrict my intake beyond eating like a predator and not cheating too often, so I don't feel comfortable calling myself "low-carb".

Again, I'm not recommending that you go ketogenic...I'm just cautioning about ending up in "kinda-sorta-low-carb" by accident.  Fortunately, dietary experiments are easy to do, and inexpensive compared to medical intervention!

Renaud:

"I just think this study is nowhere near convincing."

Convincing of what?

Paul Lee:

In general, I believe the effects of food on hunger are more important than their differing "caloric" value...but I already wrote that series of articles.

Dave:

I remember the first time I ate nothing all morning or for lunch, rode my bicycle in the afternoon, and made it back without feeling like I was going to starve or faint...it was wonderful!  I think the best part of regaining metabolic flexibility is the freedom from having to eat every few hours.

One advantage of fasted training is that it increases your basal rate of fat oxidation, so you're less dependent on glucose.  I'm glad it's working for you!

I'm finally caught up!  Thanks, everyone, for bringing an interesting discussion to the table.

JS

Michelle T:

As I've said before, I think most of the benefit of calorie counting comes from the simple act of keeping a food log, regardless of calories: it's easy for junk to sneak back into our diets, and keeping track of what we eat is often enough to cause us to cut down our consumption of junk.

That being said, there can be a place for calorie counting once you've determined a roughly optimal diet and are simply trying to figure out how much of your preferred foods to eat. Otherwise, you're not just comparing apples to oranges -- you're comparing them to coconut cream, potatoes, broccoli, and prime rib!

JS

js290:

You are correct that fat uses less oxygen per ATP generated -- but glucose contains much more oxygen than fat does (remember: C6-H12-O6 vs. long chains of carbon and hydrogen attached to one glycerol), so the net result is that more exogenous oxygen is required to oxidize fat.

This short letter neatly sums up the situation: "...Each gram of palmitate produces significantly more high-energy bonds than a gram of glucose, and the caloric value of the stored high-energy phosphate bonds derived from the oxidation of a gram of palmitate is 2.4 times greater than that derived from a gram of glucose, albeit at a greater relative cost in oxygen. Hence, when oxygen is abundant and food is scarce, there is an advantage in utilizing fatty acids for fuel as opposed to using glucose. The reverse, however, would occur when food is plentiful and oxygen is scarce."

Result: fat should be the preferred and most efficient fuel any time we are not pushing the limits of our aerobic capacity ("oxygen is scarce"). 

And, in fact, this is how animal bodies work.

JS

Howard:

Drug responses are strongly individual.  Some people, like yourself, find alcohol uninteresting, while others center their lives around obtaining it.  Others find cocaine uninteresting, while you find it strongly addictive.

I'm sure this is because different drugs tickle different receptors and boost (or inhibit) different hormones and neurotransmitters -- as do different dietary compositions.  I'll explore metabolic individuality in future installments.

As far as "that last 50 pounds", that's going to be very difficult, because you're strongly fighting leptin dynamics at that point -- you don't get to 150# of extra fat without some adipocyte hyperplasia (more fat cells, not just bigger fat cells).  ItsTheWooo has posted at length about her usually-successful strategies for dealing with this.  Others find that reintroducing some amount of carb cycling helps...but at some point you may have to kill off some of those tiny shrunken non-leptin-producing fat cells, whether invasively (lipo) or non-invasively (cryolipolysis, lasers). 

The good news is that none of this is necessary to be in good health: it's purely a cosmetic problem.

JS

Johnnyv:

That's a very interesting case of equal total weight and fat gain causing dramatically different metabolic effects...

...and I'm sure similar situations contribute to the dramatic inter-individual variation in weight loss outcomes at the same "caloric deficit".  Thank you for bringing it to my attention!

JS

La Frite:

Regarding your question, Romestaing 2007, from Part II, shows that coconut oil is indeed far less fattening per calorie than butter -- but butter causes greater satiation and satiety, causing the rats to eat less of it.  (Butter is still less fattening per calorie than standard chow.) 

The RS stuff is very interesting, but it's being oversold so hard (I've already encountered several straight-up misrepresentations of the scientific literature) that I don't take any of the hype at face value.  And vinegar has many of the same properties (e.g. improvements in blood glucose control -- documented by scientific studies, not just uncontrolled n=1), plus it's an amazing topical AND internal antibacterial (SIBO? Drink vinegar, it'll be gone soon).  The main difference I see is that no one's driving the hype train for vinegar.  (PROTIP: acetic acid is the shortest SCFA.)

JS