Better to go with J Rhodes than Garden Betty, whose feed recipes reveal an utter lack of knowledge about chicken nutrition. Not sure where she populated the info in her calculator from either - it doesn't agree with other sources. If the numbers haven't been "fudged", they've been cherry picked.
Best practice to cook beans, peas, and legume "seeds", or otherwise heat treat them. In small quantities, the toxins function (mostly) to block other nutrient absorbtion, making the overall feed less effective, but as quantity increases as a proportion of overall feed, the effects become increasingly pronounced. Except phytotoxins, those are light reactive toxins straight up, but the dosage is again a key function of the severity of the effect.
There is a 2,000+ year old wisdom, "the dosage is the poison". Still true.
follow up, 'cause there's no reason to place any trust in an anonymous internet voice (mine or Garden Betty's, though she has slick Youtube videos and a nice web site). Using Feedipedia, I built (partially) a calculator I'm still working on, using generally accepted average feed values. It needs more work, particularly an "as fed" correction (which will reduce the end value of any feed I put into it to account for moisture content), and some improved tracking of certain vitamins and other trace nutrients.
I've tested it with representative recipes from old feed books,
like this one. I've tested it with feed recipes offered by the governments of developing countries. The numbers it spits out are pretty close to the expected result - close enough to be accounted for by small disagreements over the nutritional value of the inputs.
Commercial feeds, and the research, generally places feed desires for economical production of healthy chickens for commercial purposes between 16-20% protein, around 3.5% fiber, around 3.5% fat. (There are some pretty broad ranges on fiber and fat, depending on breed, season, and some other factors - but I've not seen recommends over 5% fiber, or 7% fat in any source, that for maximizing CX weight at 8 week processing, NOT for a healthy bird) There are also targets in amino acid profiles, particularly for hatchling birds, which are important.
From my calculator, J Rhodes recipe (or what purports to be J Rhodes recipe) is 20.05% protein, 4.12% fiber, 3.68% fat and has a remarkably good amino acid profile (because of the fish meal - without it, the recipe fails). While those numbers don't represent "as fed", we can use them as a decent base line for comparison.
Garden Betty's feed recipe, using the same calculator, is 15.8% protein, 8.1% fiber, and 14.1% fat. The amino acid profile, something she doesn't discuss, misses accepted targets on three of the fur most critical limiting amino acids - the one it does hit is the one most literature suggests can be ignored, becuse if you've gotten the first three right, its almost impossible to miss the fourth with a typical mix.
How about her no corn, no soy mix? 15.08% protein, 8.72% fiber, 14.75% fat - its someohow even worse than her first recipe. So, marginally, is the amino acid profile.
Do those numbers represent "as fed" conditions? No. Most of the ingredients in these recipes are dried to around 90% dry matter, so "as fed" nutritional values would be expected to be about 10% lower. Are they open to adjustment due to differences in the quality or individual character of the ingredients? Yes. Are they nevertheless useful for comparison? I believe so.
You are free to draw your own conclusions.
