To turn my comments into a semi-reliable answer (meaning I have not much expertise in this): it seems the frequency data is being cited from Wikipedia, but the data is patchy and appears to have been (more) cherry picked Xi-wise in the Reddit comments. Before we delve into data, it's worth noting that one Romanization (like Xi) corresponds to several actual Chinese names in their ideography. Not only is the data patchy regarding the actual Chinese names writing-wise, but it's rather even more difficult to find combined statistics corresponding to the grouping Romanizations. E.g. what Wikipedia says in the two separate surname articles:
Mu:
Surname Mù (穆) It is the 98th name on the Hundred Family Surnames poem. 188th name in 2013 shared by 0.048% of the population or 640,000 people with the province with the most being Guizhou. [...]
Surname Mǔ (母) 344 母 0.010% 13.50 Sichuan.
(It's not immediately clear idea what what the last line means, but it appears copied from a table that we'll discuss a bit later.)
Xi:
[Surname Xi] 奚 It is the 45th name in the Hundred Family Surnames poem.
[Surname Xi] 席 It is the 133rd name in the Hundred Family Surnames poem.
[Surname Xi] 郗 It is 234th in the Hundred Family Surnames poem.
[Surname Xi] 習/习 It is 332nd in the Hundred Family Surnames poem. However it is not among the 400 most common surnames, occupying 0.01% of the Han population. [It's the one used by Xi Jinping.]
Note that the poem data (which from a millennium ago) is probably not ordered in frequency of occurrence, even at the time.
A separate article/table in Wikipedia provides a slightly more complete picture, according to a 2013 survey (by a certain Fuxi Institution), which only has data for the top 400 names though:
Rank |
Name |
% of pop. |
No.(10,000's) |
Province with most |
188 |
穆 (Mu) |
0.048% |
64.00 |
Guizhou |
344 |
母 (Mu) |
0.010% |
13.50 |
Sichuan |
213 |
席 (Xi) |
0.036% |
48.00 |
Henan |
278 |
奚 (Xi) |
0.017% |
23.10 |
Jiangsu |
13 |
馬/马 (Ma) |
1.290% |
1720.00 |
Henan |
99 |
莫 (Mo) |
0.180% |
233.00 |
Guangxi |
373 |
茅 (Mao) |
0.0076% |
10.10 |
Jiangsu |
182 |
解 (Xie) |
0.053% |
71.00 |
Shandong |
67 |
夏 (Xia) |
0.320% |
426.00 |
Jiangsu |
(The other Xi's like 郗 and 習/习 indeed don't appear in that 2013 top-400 survey list.)
According to this 2013 survey, there were at least 711,000 Xi-Romanized surnamed people (23.1 奚 + 48 席 = 71.1 Xi); and at least 775,000 Mu-Romanized surnamed people (13.5 母 + 64 穆 = 77.5 Mo). So the numbers are bit closer than Reddit claims, although Mu does have slight edge in this comparison (which does exclude the >400 ranked names though, so it's not a complete data set.)
If you actually count/include the close enough (IMHO) Romanization-wise Xie (解), there are some additional 710,000 thousands of these alone; never mind the Xia's (夏), of which there are some 4.26 million, which would make the Xi+Xie+Xia twice as numerous than the Mu's even when combined with Mo's (莫) -- as there only some 2.33 million Mo's. (Strikingly, there's not that many Mao's 茅; that ranks at position 373.) On the other hand, there's the "hidden dragon" Ma (馬/马) with 17.2 million, far outranking the rest of those I've mentioned.
It's a slippery slope at this point, I guess. One could even bother with the consonant-ending Meng (孟), Xing (邢), Xiong (熊), and Xiang (向 and 相) all fairly common (孟 ranks 73 with 3.84 million, 熊 ranks 72 with 3.94 million, 邢 at 111 with 1.92 million, 向 is at 102 with 2.26 million and 相 at 320) etc.
Rank |
Name |
% of pop. |
No.(10,000's) |
Province with most |
73 |
孟 (Meng) |
0.290% |
383.00 |
Shandong |
72 |
熊 (Xiong) |
0.290% |
384.00 |
Jiangxi |
111 |
邢 (Xing) |
0.140% |
192.00 |
Shandong |
102 |
向 (Xiang) |
0.170% |
226.00 |
Hunan |
320 |
相 (Xiang) |
0.012% |
16.20 |
Shandong |
Note that Mo (莫) is sometimes Romanized as Mok (esp. in Cantonese--apparently due to different pronunciation in the latter), so the issue with consonants-ending is not that straightforward.
Adding all the Xi-prefixed names certainly beats the Mu's, but probably not the Ma's.
And I'm totally torn about Xu's (許/许 and 徐). In which (larger) bin should these go in the Xi vs Mu debate?
Rank |
Name |
% of pop. |
No.(10,000's) |
Province with most |
11 |
徐 (Xu) |
1.450% |
1930.00 |
Jiangsu |
26 |
許/许 (Xu) |
0.660% |
881.00 |
Guangdong |
And if you're asking my hunch, it's probably the case that (regardless of actual frequency in China) the average Westerner doesn't know (can't indicate) any individual named Mu (or even the far more common Ma or Xu), but for Xi there's the obvious presidential designator.