Several things are going to get in the way of attempting to correlate chemical composition and susceptibility to hydrogen embrittlement in MS21250 bolts:
a) As pointed out by israelkk, quench response is not the same for the different steels authorized by the standard, and many manufacturers switch grades somewhere up the diameter scale.
b) One of the preponderant factors in the NASM8831 requirements is the double shear strength requirement - there is a lot less spare strength here than in tensile, which is another reason to switch grades with varying diameter (really a special case of a)
c) By far the greatest influence upon hydrogen embrittlement in my view would be the variation in the cad. plate process. Some proprietary mixes of cad. plate chemicals contain brightening agents which can have a significant effect upon embrittlement
d) MS21250 is a very widely used bolt in the aerospace field, and is a perfect candidate for sale by distributors. Some odd diameters and lengths may have been on their shelves for some time, and it should be noted that the embrittlement relief treatment duration has been changed a few times over the life of the QQ-P-416 cad plate standard.
e) Last but not least, how do you intend quantifying hydrogen embrittlement ? Often times, without any noticeable difference in structure under an electron microscope, fatigue testing will give better results before cad. plate than after.