(This was a late-night thought, the basic premise of which is that Wadi el-Hol might more squarely fit the after-Sinaitic and before-Arabian writing period - placing it anywhere from possibly around 1400 to 1200 BC).
I think, given what I feel to be the high consistency of my translation of Wadi el-Hol both internal to Semitic alphabetic scripts but also to our knowledge of the specific use of that area [of Wadi el-Hol] that the alphabetic inscriptions there are likely from a range closer to 1400-1200 BC rather than 1800 BC.
In fact, the presence of this distinct ḥ, arguably continuous with the Arabian (12th-9th centuries) ḥ probably should have indicated this possibility far earlier on. Whereas I have argued, and will continue to do so, that a fish-ḥ is present in Sinai, this isn't found outside and neither is another (or arguably a) distinct ḥ. However, that same ḥ does occur in Sinai, Proto-Canaanite, Phoenician [and Greek and Latin E] as an h. The confusion of an apparently less distinct Northern Semitic ḥ in contrast to the continuing distinction in Ethiopic, Arabian etcetera is one possible mode of transmission (a sort of Shibboleth confusion probably of similar names - hawy and ḥawy, perhaps).
Nevertheless, they may suggest a key source of continuity between Wadi el-Hol and Arabia. And that seems most likely if Wadi el-Hol is closer to the 2nd millennium attested origin of Arabian scripts. Additionally, the Wadi el-Hol h, possibly a precursor of North Arabian but not South Arabian variants, a one-arm-up and one-arm-down person, might be attested in an anomalous Yemeni inscription called Jamme 863 - which also attests the Arabian worshiping ḥ but upside down (for Epigraphic South Arabian). Jamme and Albright felt it was anomalous and probably archaic (I believe they or Albright singled out the patina) - and it is on a rock wall, though it is not clear if that might have been a structure.
Moreover, the same one-arm-up and one-arm-down person may also have become the Ugaritic alphabet's ḥ, its h the more commonly post-Sinaitic h. The Ugaritic alphabet evolved sometime between 1200 and 1150 BC in Coastal Syria. It also possesses at least one partial abecedary (I think two) in the South Semitic [Arabian] order. This connection is unusual. One possibility is that there may have been some influence from Ugarit toward Arabia. The obvious connection is the Syro-Arabian desert; and caravans would have traveled between Syria and Arabia.
However, it is also possible that the post-Sinaitic developments in South Egypt (Wadi el-Hol > Theban inscriptions), may have been transmitted to the borderline of North Arabian and South Arabian territory earlier than thought (Jamme 863). It is possible that Ugaritic was actually the result of a circuitous transmission from Arabia, which was then influenced by the prevailing Proto-Canaanite trend. In support of that is both the cuneiform (not just in the letters, but in the vocabulary and scribal training methods) and the apparent lack of importance of Ugaritic's 'second S', which is basically Canaanite samekh. The primary presence of the Wadi-Theban-Arabian S1 (~shin) and the common S2 (~sin) also suggests the possibility of significant Arabian influence, however seemingly unlikely. Though it also could suggest the inverse, except that some of that paleography does originate at Wadi el-Hol and not in cuneiform.
The core question is whether the evidence for the dating of Wadi el-Hol to 1800 (or 1800-1700 or whatever) BC is strong - particularly given it is based on an unknown language and no agreed-upon translation, nor any physical evidence. The original brief report by Darnell et al. (2005 (?)) made this determination, to my memory, on flawed paleography and the presumption that traveling that path was common around 1800 BC. It is significant they mention this was likely the cite of a Hathoric procession - and that rock spur likely the spot of or near a temple or religious place for a particular ritual. To my mind, inferring from the several other Darnell papers I've read on this area (the area around Wadi el-Hol) and Hathoric celebration, if the inscriptions are related to that sort of 'best known' period, then the inscriptions are more likely from the New Kingdom or just before (given their anomalous presence).
In effect I see these inscriptions as integrally related to a large number of the Egyptian inscriptions on that rock - Hathoric worship, probably in specific reference to a specific myth. Based on my translation, I had always been perplexed at how the Wandering Goddess story could have been present by 1800 BC. However at the beginning of the New Kingdom, there is now a philologically consistent story.
It may be the case that with the collapse of the Hyksos, random scribes were pushed primarily across Sinai into Palestine. However, some no doubt remained in Egypt and/or as the Syro-Palestinian royal and aristocratic entourages returned to regular tutelage, it spread that way [possibly also]. Hiring Canaanite shamans was not unknown, so some scribes from the Sinaitic period might have found purpose there. Two of the Petrie Theban inscriptions are likely invocations - one of which is a curse; another, I have argued with increasing personal skepticism may be a star chart. The third is very unclear but is possibly an invocation to Set.
Having said all that, the Wadi el-Hol inscriptions are not a hoax. However, they are likely not as old as has been commonly thought. And they are probably not the earliest alphabets - which is inherent also from the lexical usage in the vertical inscription and the name of S2 in the horizontal inscription. They do, however, represent one of the most important discoveries in the history of the study of the alphabet - in that they likely do reflect the earliest definitive common link between Sinai and Arabia.
They were clearly carved in transit - at the site of a Hathoric shrine or temple on a road between Hu and Thebes. The later timing also explains why Athtar's name is written with an `ayin that is, effectively, a Ugaritic `ayn - they aren't separated by 600 years, they're separated by perhaps 200-250 years. That leaves fewer interceding missing links.*
*Some of that interim can actually be found in poorly known cases: i.e. of 3 inscriptions from a Deir `Alla palace destroyed by earthquake in 1200 BC (not the 'Deir Alla Inscription'), and at least one presumably non-'workers' graffiti' inscription at Deir el-Medina published (only?) in an archaeological team's report. The last is not definitive, in my mind. The Deir Alla inscriptions - at least the first and possibly the third are arguably somewhat solid (the first I think is). These also potentially provide another oddity - a point of transmission to Greek chi. Greek itself, in its earliest forms, had also been transmitted by this point, and there have been some suggestions it does not match Proto-Canaanite perfectly.