I really enjoyed SMTM's "A Chemical Hunger" series[1]. But recently they've gone into weird fad-diet territory with their potato obsession.

Like, they crow that the potato mono-diet "works," in that the people who successfully followed it lost weight. Well, sure -- all of those 70s fad diets "worked" in that sense! Grapefruit and popcorn? Sure, you can lose weight on that!

But their own numbers show that people regain the weight after they start eating other foods again: "On average, people gained back most of the weight they lost."[2]

People who successfully follow very restrictive diets will lose weight... as long as they follow it. And these "riffs" in the OP where it's potatoes and bacon, or potatoes and gummi worms, or whatever, won't change that basic observation.

[1]: http://achemicalhunger.com/

[2]: https://slimemoldtimemold.com/2023/01/26/smtm-potato-diet-co...

Beautiful. We are in a golden era of citizen science, where access to knowledge, tools to connect, tools to process data, and the ability to communicate this is at an all time high. The kind of stuff you see on Youtube is amazing: people like AppliedScience achieving incredibly things in the garage, or recently NileRed took a nature paper [0] 1 step further and published it on youtube [1]).

From a chemist/material scientist perspective: Whether the results of the Riff trial may ever have a p value suitable for nature/science, likely not. When it comes to the human body and our biology, a mass trial like this may even be more useful than traditional studies, where pre-existing biases in data collection may weed out the most useful 'Riff'. Better than that, the information collected by mold_time is regularly released and discussed, in the open, on twitter/x [2].

[0] https://www.nature.com/articles/nature25476

[1] https://www.youtube.com/watch?v=CglNRNrMFGM

[2] https://twitter.com/mold_time

I think this is a more compelling idea than people in the comments are giving it credit for

> The problem is that you can easily come up with 100 different hypotheses for what’s going on. Ok, so you run 100 different studies to test each one. But studies take a long time to run — let’s say 6 months per study. Congratulations, you’ve just locked yourself into 50 years

This is a major problem with science whenever you have less of a theoretical foundation. Compared to physics or chemistry, we know very little about nutrition or sports science. Because of this, the search space is very large. One could argue that given the number of surprising results (and difficulty reproducing those results), medicine and psychology also fall into this category.

> A riff trial takes advantage of the power of parallel search. Some riffs will work better than others (or at least differently), and parallel search helps you find these differences faster, especially if the differences are big.

What if we did more to encourage people to track and report their personal experiments? If even 10% of everyone on a diet (any diet) just tracked what they ate, what exercise they did, and how much weight they lost, and reported it to a centralized database, scientists could then look for patterns in that data and do formal studies based on suspected patterns.

We could do similar things with longevity/happiness. Look at the "Harvard Study of Adult Development" but imagine it was spread out over 10s of thousands of diverse people instead of just 300 upper-class American men? The data quality wouldn't matter much if all you are doing is searching for patterns to do follow-up studies.

I remember Kevin Smith talking about his first stab at dieting, following the instructions of the "guru" that got Penn Jillette lean: basically, he ate nothing but potatoes for something like 3 months.

The point wasn't that potatoes were particularly good for you, but to "reset" one's attitude towards food, eating only when hungry rather than for pleasure or to deal with stress - because after a few weeks, one is so tired of potatoes that they'll forego unnecessary eating.

In this sense, the potato diet is probably useless in itself, but might work in that "reset" role.

Does anyone remember a meta study that showed ANY even slightly reasonable restrictive diet works because they almost by definition dramatically reduced sugar and HFCS in the diet, and if you just controlled for that, the individual types of restriction didn't matter?

I can't find it on PubMed or in my bookmarks.

If memory serves - high/low fat, high/low carb, high/low protein, etc - didn't matter as long as the restriction is stopping sugar/HFCS.

And it also explained the rebound effect - e.g. after the extreme restriction, the participants start re-introducing sugar and HFCS back into the diet, and since that's the real culprit, weight goes back up.

No taking away from this super cool citizen science - deep kudos on testing things like this out! I'm tempted to participate in something like this. Self-experimentation is a lost art.

Diets seriously need to be considered through the lens of Occam's razor.

First, radical dietary changes cause rapid fluctuations in weight regularly. Not fat, weight.

Second, if you eat only one thing or are generally restrictive... You will probably lose a lot of appetite.

It's no good going off on some theory about pottasium. Occam's razor suggests that this also works for the been soup diet, fruit diet, etc.

It is, potentially, useful to do these kinds of things as an isolation diet (gradually add back foods and pay attention to effects). It's also useful to just break bad habits by doing weird stuff sometimes.

I'm Irish. I love potatoes. They do not have magic dietary powers. The potato doesn't explain anything. You can get similar results with toast, bacon or bananas. The results do not mean anything specific for longer term fat loss, health, etc.

It just proves that bodyweight fluctuates in response to radical diet change. That is known.

Same for a lot of the "water tricks" that thankfully have started to die down. They use bodybuilder tricks to eventually dehydrate themselves for a perfect look on stage day. It's sold as a weight loss trick.

I'm so disappointed by SMTM's trajectory. One of their key initial insights was that you don't need control groups for diet trials because essentially no one loses more than 10-20 lbs on a diet. You can just run the treatment group and, if you get lots of people losing more than that, you know you've got something.

The flip side, of course, is that losing 10-20 lbs from a diet shouldn't be taken as proof that the diet does anything special. People can do that with almost any diet.

SMTM's potato diet study found exactly that - 10 to 20 lbs of weight loss for most participants. This should be strong evidence that it's not a silver bullet. SMTM is pretending otherwise.

I feel like the only reason people are losing weight is because eating 20 potatoes in a day is insane and no one on this diet is hitting their BMR. Of course they will lose weight and they will with any variation that keeps them under their BMR.

For people saying any diet works, I think the interest here is the rules of "eat as many potatoes as you want" are very easy to follow, it's unambiguous if you followed it properly, simple to buy the ingredients anywhere and when eating out, it's cheap, is meant to work fast, and you aren't going to feel hungry which helps a lot. Most fad diets don't tick as many boxes.

Feels obvious to me that it works via “calories in, calories out” though. 2kg of potatoes a day is about 1500kcal so it's hard to overeat.

For the participants it didn't work on, surely the most likely cause that should be controlled for is how many potatoes they ate or how much oil/butter (some of the most calorific ingredients we use) they had on top?

And not if participants avoided tomatoes ("Tomatoes are our top bet, but other possible blockers might be: wheat, bread, grains more generally, maybe meat.")?

So people lost weight and these people have “no idea why?”

Well, ok, it’s calories, but why even bother to have any sort of study about this?

Of course people who don’t know about calories would think they invented a whole new idea and methodology on scientific research

I just read thousands of words about a potato-based diet on that website. Didn't think my Tuesday was gonna start that way, but here we are. HN is amazing.

One possible riff: cook, chill and reheat all the potatoes. That is known to increase resistant starch.

This concept of a Riff Trial is new to me. Sounds like it would serve as a great pre-trial method to hone in on designs for more serious trials, one with controls and blinding. Is this already the case? Or are riff trials generally only performed in citizen research?

As Vincent Van Gogh once said, "I'd rather die of potassium than of boron."

I love the half-mad, half-hilarious, Ben Franklinesque quality of this enterprise: “Some people think the potato diet causes weight loss because it is bland. We think this is wrong too. First of all, potatoes are delicious. Second of all, this doesn’t make any sense. Why would that happen.”

I'd participate if I hadn't already tried the potato diet before.

My guess is that it probably works, at least to an extent. The reason I used the word "probably" is I didn't stick with it long enough to lose over 10 lbs. This is because it's probably the least pleasant weight loss technique I've tried. No joke, I'd rather eat nothing the whole time. Eating potatoes seems awesome at first, but I began to really hate the taste and texture after just a few days. This is of course my opinion. Some people have had great results on it. It's just not for me. I know the author of this article doesn't agree on this at all, but I beg to differ. Almost every other technique is more comfortable for me. Though I'm sure it'd be fine if I added lots of fat and ketchup. I just wasn't going to do that and risk gaining a bunch of fat.

Off the top of my head, here are all the possible ways that a monodiet of potatoes can result in weight loss:

- Potatoes are a highly-satiating food, one of the highest in satiety

- Potatoes get boring without lots of added fat, salt, and spices.

- Potatoes may contain some "resistant" starch, though I've seen a few people self-experiment with resistant starch and conclude that it's no different than consuming glucose.

- A potato diet mostly engages the metabolism of glucose and not so much with fat (unless you're adding a ton of fat), so the Randle cycle is engaged much less, hypothetically. Though this is supposedly contradicted by people eating potatoes with heavy cream.

- Potatoes only cause a small increase of uric acid in the blood in contrast with other foods, uric acid having a correlation with fat mass. Then again, there are people who eat diets that cause a higher increase in uric acid and remain lean and muscular (carnivore diet is an example).

- Sarcopenia from a lack of dietary protein can cause some net weight loss.

Most other ideas I've seen are pure hypotheses that either aren't that plausible, or are untested, or are mechanisms only demonstrated in vitro.

Some things I'd like to note about this crowd study and what I'm reading in this proposal:

- Allowing participants to choose their own adventure is a poor design. There's definitely value in testing different combinations of adjunct foods along with the potato diet, but this should be controlled based on how many participants sign up. The cohort might otherwise become lopsided towards a certain preparation, adding sour cream, etc. Instead, candidates should be assigned an adjunct and be allowed to accept or reject the challenge.

- The part about "If you can’t get potatoes, eat something else rather than go hungry, and pick up the potatoes again when you can" really reduces the potential value of this experiment. This may add too much noise. It's already bad enough that people are bad at self-reporting, but now you're giving people permission to just do whatever.

- Participants should record their physical activity. This will of course be full of statistical noise, but it might as well be recorded in case it's helpful. Have them record their activity for at least a month prior to the trial and record it during the trial. The reason I think this is a good idea is that, when dieting, people may have a tendency to put themselves into an "I'm getting healthy" mindset which encourages them to also get exercise, which is a confounding factor here.

- Day-by-day body weight data is mostly worthless. I would explicitly encourage participants to not record their weight at all except at the beginning and end of the trial. Daily weight checks can have a psychological effect that may encourage the participant to perform actions that confound the trial, such as walking more or eating less than they otherwise would in order to make sure they get the intended result. This can happen unconsciously.

Otherwise, I think this idea is awesome and that more crowd "riff trials" should be done.

Does anyone know of a site dedicated to riff trials? If there isn't one, it should definitely exist.

Let's try potato + semaglutide.

I love this idea.

Does anyone know of a similar riff trial related to irritable bowel syndrome and gut health?

I imagine anyone who is researching diets this heavily and is willing to sign up for a weird one may have lost weight regardless of if they used the potato diet or not - they committed to serious lifestyle changes that caused them to lose weight. Maybe it’s potatoes, maybes it’s everything the person changed in their and potatoes didn’t offset those.

Only works if the potatoes are McDonald’s fries.

One suggestion: give participants an "IDGAF LOL" option where the experimenters tell them what protocol to follow, and randomly assign some leading contenders.

That way people can do what they want, but we might also get some well-identified estimates :)

We already know you can rapidly lose weight eating nothing but Twinkies. Why bother with this?

I like how they are really going to hash this all out in one go.

> The diet worked — people lost 10.6 lbs on average over only four weeks — and we had basically no idea why

I do fasting a lot. Doing diet for weight loss is a huge red flag. Human body gains and loses water very easily. Change of 10 lbs (in any direction) is rounding error, after changing a diet.

Major reason to do potato diet are health benefits. It decreases inflammation, and gives your gut chance to heal. It may improve sugar digestion, liver and so on. Basically any junk and toxins you eat normally, go away on potato only diet.

AFTER you become healthy, you may try to lose weight.