Crops are also not perfectly efficient, but it's simply replacing two chained instances of inefficiencies with one.

Instead of

Crops -> Animal feed -> Cattle -> Meat -> Eat

you have

Crops -> some processing -> Eat

Then there's the 92% of grazing land that doesn't grow crops. That's a clear win for cattle.

According to the UN's Food and Agriculture Organization [1] "just less than half the world's usable surface is covered by grazing systems" and "Grazing systems supply about 9 percent of the world's production of beef and about 30 percent of the world's production of sheep and goat meat."

Doesn't seem like such a clear win for cattle. They do note that "For an estimated 100 million people in arid areas, and probably a similar number in other zones, grazing livestock is the only possible source of livelihood.", so not everyone might be able to afford a vegetarian diet, but 91% of beef seems replaceable.

[1] http://www.fao.org/docrep/X5303E/x5303e05.htm#chapter%202:%2...

No it's not. Only a tiny fraction of cattle are solely pastured. Each beef cow, even if pastured for the first year (which is typical) requires literally tons of high-calorie corn and or soy feed to bring it up to slaughter weight. Guess where that feed is grown.

Besides, the range land you're talking about isn't needed for crops anyway. We have more than enough arable land to grow crops for direct human consumption.

Beef is fattened on feedlots before slaughter, the exception is gourmet beef, which I doubt many people eat regularly.

https://www.theatlantic.com/health/archive/2011/12/feedlots-...

Are you talking about all the land that used to be rainforest?

Here is a really great article that covers a lot of ground in terms of what food is most efficient for protein both in terms of land & CO2.

https://ourworldindata.org/yields-and-land-use-in-agricultur...

EDIT: sorry here is the better link:

https://ourworldindata.org/meat-and-seafood-production-consu...

I don't know where you're getting your information, but it's not correct.