I like this scatter plot. If you really want to get freaky with it, you should take into account the “protein digestibility-corrected amino acid scores.” Things like eggs and whey are 100%, beans are usually in the high 70’s, and peanuts are actually down at near 50%.
So for nutrition’s sake, not all protein sources are created equal, and it makes sense that if you are trying to get adequate protein at the lowest price, you also want to get sources where you can eat the least of it to satisfy the protein requirements of your body.
This data seems way off - steak has 24-30g of protein per 100g, for example
deleted by creator
This would be cool with ratio of protein to calories as well, in the same format.
Yeah protein vs calories would be way more useful than protein vs weight.
Great post!
I wanted to add that this isn’t quite how proteins work. Those protein-rich legumes aren’t what you would call ‘complete proteins.’ There’s a number of amino acids our bodies use as proteins and while legumes are a good source of many of them, there’s a couple proteins you won’t get enough of from just the beans. Fortunately, brown rice- while not as rich in protein- gives you the proteins that the beans are lacking. That’s why beans and rice are a match made in heaven.
Herbivorous animals are just better at metabolising proteins from plants and of course they’re capable of eating much more than us. That’s why they’re able to live off of grass.
This just stuff I read up on a few years ago so if I’ve gotten something wrong please say so
I want this chart, but add the third Z axis of “environmental cost” whether it be just CO2 emissions or a “total” impact score.
I imagine those legumes get even stronger, while the meats lose ranking.
Curious where tofu would be at on this chart.
Probably somewhere around the legume cluster. They’re really pulling their weight there, as expected, though peanuts are quite the dark horse.
Pricing and product availability is regional and variable, but some napkin math using my local Walmarts pricing puts it at:
- 5.33g protein per 100g food
- $3.04 for 30g protein
That puts it in the green veggie cluster if I’m not getting the axes confused.
the problem is there are a ton of varieties of tofu, and they’re all generally around the same price. it seems the silken tofu have around 5g of protein, but some of the extra firm varieties have over 15g protein per 100 tofu, putting it in a much more respectable spot in the bottom middle with the grains and such
Is that extra soft tofu? It usually has more protein than that. A pack of extra-soft I have is 8g / 100g, and some other varieties seem to be 10-15 from online sources.
c/theydidthemath?
Thanks so much, that’s very helpful and actually a little disappointing.
No problem! Turns out my Walmart only sells a single package type of tofu so take it with a grain of salt. It’s still a cheap and good protein source, but not as dense as an animal or legume protein source.
Take it with way more than a grain of salt — add some nutritional yeast and MSG, anything to give that tofu flavor!
Comparing liquid to dry foods on the same chart is completely disingenuous. Also look at any label. Cows milk always has more protein than soy milk.
I don’t see what you mean. The left axis is a measurement of cost per gram of protein. The bottom gives you a measurement of density. So anything lower on the chart is cheap for what you get and then the further right you go the smaller the portion required to consume to get that amount of protein.
How do liquids differ from dry foods in terms of protein stuff? (Waves hands vaguely).
I was mildly surprised that milk is way down in the bottom left quadrant.
its mostly water
This may be helpful from a cost / gram of protein but its a bit misleading on the grams protein/ 100 g axis for beans - those are the dry bean numbers.
It would be nice to see cholesterol counts on here as well. Good work though.
Why? I thought dietary cholesterol had little to nothing to do with blood cholesterol levels? That’s what it seems to me like this source is saying, but I’m not an expert by any means.
You’re correct. Saturated fat intake increases LDL (atherogenic kind, e.g. the “bad”) cholesterol, there’s even a nice formula for it: https://pubmed.ncbi.nlm.nih.gov/25286466/
I got a bag of almonds for $11. 85 grams of protein. Holy fuck. Also delicious!
Explains how elephants are able to get so jacked without consistent income, they just poppin’ them peanuts for days!
Sirloin steak needs to be added to this. It’s not only cheaper than other steak but it’s significantly higher in protein too at 27g per 100g.
What, no grubs?
how legumes have more protein per weight than ground beef?
Most likely dried legumes vs uncooked meat that is full of water.
Y axis should be reversed. Since 90% of graphs are trying to find the sweet spot, would y the lowest-cost, highest protein be in the upper right? Unless I’m reading this correctly it’s the lower right, which is hardly /dataisbeautiful
Course I’m shit at math and could be wrong…
You might notice Pork Belly and Ribeye Steak belong way higher than the red dots representing them. In your proposition, I don’t think there exists an elegant way to represent those data points.
If you did grams per USD, then the Ribeye would be 0.06, Pork Belly 0.10. The next worst would have been 0.25, so I think it would clearly show the relatively poor cost per protein.
Of course, I don’t think anyone is deluding themselves to think that those foods are the ones to choose if you just want “some source of protein”.