I'm getting questions about ketosis, as well as my evolving views on starches, to I'll describe my own experience with ketosis and starch intake.
Before my dietary transition, I always ate a fairly animal-fat heavy diet. I simply stopped eating sugar and flour in the fall of 2007, and I had never really eaten that much in the way of starchy vegetables. So when I read that "carbohydrates" were the likely cause of diseases of civilization, I pretty much axed them. I still ate onions, tomatoes, asparagus, green salads etc. It was an easy transition, with no "carb flu" or indeed any adverse effects to speak of. I had increased energy and slow weight loss without any hunger, which stopped at about 6-9 months. I did have substantial improvements in long-time IBS and atopic symptoms, as well as some unusual symptoms that may have represented a peripheral neuropathy. (Interestingly, these improvements came with wheat reduction, long before wheat was totally eliminated.)
My total carb consumption went from maybe 35 -40% (already moderate) down to maybe 40 or 50 g/day - about 7%. At that level I was usually mildly positive on urine ketone test strips. I started out with no evidence of any metabolic disorder and weighing 169 lbs at 5' 11''. I had been a consistent runner for about 10 years. Interestingly, and supporting Richard Nikoley's walking experience as well as Gary Taubes assessment of the inefficacy of exercise for weight loss - my peak fat and weight of 169 lbs corresponded exactly to the summer I was running about 25 miles a week. Both before and after that time I weighed less by about 5 lbs.
Early in my reading, after GCBC, I flirted with the idea that ketosis or very minimal carbohydrate intake might have been a consistent or even necessary feature of our evolutionary environment. It was easy to discard this idea pretty quickly, though, as most of the anthropological and ethological evidence argues against it. You would have to think that the exceptional periods of European occupation by hominins during glacial maxima were more important than the other parts of our dietary environmental experience. Or that the Inuit foodway was typical of hunter gatherers and not an exceptional late paleolithic niche.
So I still entertained (and still do now) the idea that ketosis could be a useful therapeutic maneuver, in cases of neurodegenerative diseases which seem to be marked by already acquired defects in glucose metabolism (Alzheimer Dementia, Parkinson's) and I've done quite a bit of reading on ketosis as a CR Mimetic (calorie restriction mimetic) for treatment of cancer or life extension. So far, though, nothing I've seen contradicts my assessment that for normal people deep ketosis is not something that needs to be sought, and I've seen no evidence that failure to maintain ketosis is a cause of diseases of civilization or that ketosis will lead to a preternaturally long lifespan (worms notwithstanding).
For over 2 years, I continued to eat at the border of ketosis because my weight was stable at 151 lbs and I felt fine that way. All the while long time readers will remember that (when this was still an "argument clinic") I was fighting defensive actions against assaults alternating between WAPF fanatics (not you, Chris!) upset about wheat, and the meat- and-water zero-carb hezbollah who could not understand why I could not see that glucose in the diet was a poison.
During the 2 years I was eating less than 10% CHO, and often much less, I ran an average of twice a week 5K and did McGuff style (actually just what I learned in high school football and swimming) weight training about twice a week. The running is always done just before the weight training. I would usually work out fasted, and the only adverse effect of the low carbohydrate intake - that I can identify in retrospect - was mild orthostatic hypotension for an hour or so after the workout. I would feel a bit lightheaded if I stood up quickly from a squatting position. I cannot be certain that it was not the fasting state, though, as now I usually work out in the evenings non-fasted.
The only thing that made me change my carbohydrate intake was a dramatic increase in physical work requirement. In June of 2010, I had to work about 60 hours a week doing real work on some buildings I own. I mean, carpentry, painting, bricklaying, hanging gutters, climbing on ladders. What happened was I started to lose weight. With no increase in appetite. It wasn't that I had difficulty doing the labor, I was just dropping weight. I got down to about 145 lbs with a month of this type of work and decided to increase my carb intake to see if it would increase my appetite. I added a large bowl of generic rice krispies with half and half every morning and made sure I got at least 100g/day of starch (or minimal sugar, lactose etc.) and eat three times a day. That was all it took. I gained back all of the lost weight in about another month and have kept at 100 to 150 g/day of mostly starch since then. I've increased by workout intensity a bit in the past few months, but it's nowhere neat the total work I did for several months of the summer. Weight right now is 153 lbs.
Things to Note:
1) Weight is less than 1% different between VLC diet with 7% carbs and LC at 20% or more
2) Body temperature has been exactly the same at each carb intake level and the same as it was for me in medical records from over 20 years ago when I ate 50% carbs and had much more body fat. By "exactly the same" I mean the thyroid enthusiasts would claim I have a "low metabolism", yet my wife can't understand how I sit around in the winter with only a t-shirt on. 97.5 F orally. Normal labs. Normal ultrasound (I owned the machine)
3) My own n=1 experience doesn't prove anything but does bias me to believe in Stephan's set point idea and not in the idea that the more CHO you eat the more you will store fat. That idea makes no evolutionary sense, either.
3) My own experience argues against the idea that you will "ruin your metabolism, starve, die, get hypthyroid, etc. by eating VLC for years, but does argue that work tolerance and exercise tolerance will be much better if you eat substantially above the level required to just stay out of ketosis. I don't really feel any different with changed CHO intake, but I don't deny that many people are quite sensitive to changed CHO intake.
I've never advocated anyone eat VLC, only described that I often did. It is not necessary to eat VLC to stop the metabolic damage of the SAD. VLC might only be necessary if you have diabetes or need to lose a lot of weight. PaNu is not intrinsically a weight loss prescription. It is not therapy. It is just eating in a non-damaging way.
So in keeping with the mission of the blog, which is just to share what I know, I am letting you know my experience. I am not recommending any particular level of macronutrient intake for anyone. I am open to the idea that macronutrient ratios outside of wide parametric ranges might be important, but I think it is an unproven concept. You can see from recent posts I don't think the very concept makes much sense.
I don't routinely measure my CHO intake or anything else. Everyone will want to know, so I will say that my purely speculative idea is that as long as there are enough animal foods in the diet, a range of 15 to 60% carbohydrates (exact numbers not important) are probably all equally healthy as long as the NAD are avoided.
I intend to keep eating at least 20% starch, which is twice the Kwasniewski number, to give myself a buffer against ketosis and because I speculate that this is about the level at which your glycogen stores start to get optimized. I certainly feel that way. The post-workout orthostasis is completely gone*.
Oh, and Rice Krispies (of whatever brand) have have only a few grams of added sugar in a bowl. It's really just toasted white rice. A good sized bowl is maybe 40 g of starch, a few grams of sugar, plus whatever lactose if you add H/H or milk. For lunch and/or dinner add sweet potatoes, white potatoes and whatever green veggies and dessert fruit you eat and that should do it.
*Interestingly, the orthostasis was independent of my resting blood pressure. It's been steady at about 118/66 all along. Before I first went VLC it was about 125/75 or so.