10% may be a magnitude too high at least if we expect the explanation to be correct. For example in a sample of statisticians/epidemiologists only 12.5% got two basic features of p-values correct [1]. And there are a lot of studies showing similar levels of misunderstanding among people using p-values professionally.

I use p-values daily. I can and do compute them using various methods, including by hand. But I'd still not be confident in my two sentence explanation. P-values are very unintuitive and very easy to get subtly wrong.

[1] https://pubmed.ncbi.nlm.nih.gov/35991465/

P values are tricky, and I think you can get a lot out of knowing what they are without being able to compute them and use them reliably.

But statisticians not knowing the basics of the Null Hypothesis is absolutely terrifying!