What I don't understand is how recipes high in olive oil are any good then.
That's because soap profile numbers are just that … numbers. They're quite simplistic sums of some fatty acids, with the best intentions, but they just cannot reflect all details of the very intricate properties of oil blends, let alone in a quantitative manner.
High-olive recipes (castile & friends) do get
hard (note how the
hardness number of calulators underestimates the actual hardness), but they're not famous for being very long-lasting – in this respect, the calculator numbers are even helpful to not rely too much on the apparent hardness that well-cured olive soap brings.
Still, the longevity/hardness numbers don't distinguish between, say, olive and soybean oil – which is about as weird as it sounds. They are of some use, but they won't tell you by themselves about their limitations.
On top of that, they're not complete. For one, the superfat thingie. Then, oils like cupuaçu or sal butter bring “minor” FAs that contribute to hardness, but aren't recognised by most
soap calculators. The worst is soy wax (vegetable oils hydrogenated to a degree that the manufacturers usually aren't eager to disclose) – you have to guess/tinker around until you've got an understanding of its behaviour.
Then there are general process-dependent issues. CP/HP? Gelling/not gelling? Water content? EOs/FOs? Dual lye? Sugar/sorbitol/goat milk/coconut milk/rice? I've found that soaps with palm as the only hard oil will
usually harden up nicely, but sometimes not; but even if they don't, they
may or may not have a more satisfactory longevity than expected from the softish touch.
I personally (with my
usual notorious choice of hard oils) try to aim at a longevity number below 35, since above that, I've witnessed false trace once too often
. “Level-up” from the calc numbers is a deeper understanding of the actual FA profile. My (current) ideal soap is about 7–11% stearic, and about twice the palmitic.
In your case 60% lard, 25% olive, 10% almond, 5% castor. (Yes, it's not a mistake I've left out coconut. Soapcalc numbers really do what they can to trick people into above-than-average levels of coconut, but not me
.)