You've often said this, but does it really matter?
When somebody feels like a loser, usually somebody else feels like a winner. Was there a distinctive swing of the ethnic German and Hungarian votes away from the Dems in 1920? Only the Germans and Irish seem to matter numerically.
Were ethnic groups so attuned to the details that, for example, did the Dem vote stayed high for South Slavs but Italian-American went heavy GOP when Italy got publicly bitter about the mutilated peace?
If one were to design an approach to a European settlemen to appeal to a majority of American voters with strong ethnoreligious identities and sympathies, I'm sure there's some set of positions an American leader could champion that is clearly on the side of ones with more rather than less voters. It should be as simple as counting.
Was there a disproportionate swing to the GOP among German Americans? Well, let's consider Stearns County, MN (St. Cloud)--a heavily German Catholic county that went three times for Bryan, rejected Wilson in 1916, and as for 1920:
86.3 percent for Harding!
https://en.wikipedia.org/wiki/Stearns_County,_Minnesota
The Irish vote? "In 1920, when, among the Irish, registered Democrats vastly outnumbered Republicans, Warren Harding carried one group and was competitive in the others. Harding received one-third of the vote cast by the lower-class Irish of South Boston, Uphams Comer, Charlestown, and Mission Hill at a time when only 6 percent of those identifying with a party considered themselves Republican. Similarly, in lower-middle-class Brighton, Harding received 53 percent of the vote cast, carrying a population three-quarters of whose party enrollees were registered Democrats..."
https://books.google.com/books?id=FvZsms2NMcYC&pg=PA153
The Italian American vote? In Chicago, it "overwhelmingly chose Harding over Cox and Debs, as returns from representative Italian precincts show."
https://www.google.com/search?hl=en...-ab..0.0.0....0.Ril1tei746w#spf=1570842617759 In Boston, "In 1920, Republican presidential candidate Warren G. Harding received 53.7 percent and 70.1 percent of the vote in the city's "Little Italies" located in East Boston and in the North End."
https://www.google.com/search?hl=en.....4.0.0.0...208.rP_FJ46fPvg#spf=1570843978408 A study of the normally Democratic Italian--American voters in Providence indicates that Fiume was definitely a factor in their swing to Harding.
https://books.google.com/books?id=vtP0KA_L9WAC&pg=PA52
The Polish vote in 1920? It shows that it is not necessarily true that if you alienate ethnic group A you get support from rival group B. Giving Poznan, the Corridor, part of Upper Silesia, etc. to Poland hurt the Democrats with the Germans but didn't help them much with the Poles. Harding got 44.22 percent of the vote in Chicago's Polonia--slightly behind TR's 46.15 percent in the 1904 landslide but Hughes had gotten less than 30 percent in 1916.
https://books.google.com/books?id=dt1hXjPYgxAC&pg=PA243&lpg=PA243 Some Poles resented the minorities provisions of the Versailles Treaty, which they saw as granting "special privileges" for Jews.
https://books.google.com/books?id=dt1hXjPYgxAC&pg=PA118
Well, you might say, if the Poles thought Wilson too pro-Jewish, surely the Jews must have voted Democratic? Wrong again! Cox trailed not only Harding but Debs in heavily Jewish parts of Boston:
https://books.google.com/books?id=7g1jaWlsY24C&pg=PA152
I don't think it should be necessary to go on to show that there was indeed a massive ethnic backlash against the Democrats in 1920 caused at least in part by the Treaty. Maybe Hughes could negotiate a treaty that would avoid such losses, but I doubt it.
(One might add that whichever party got the US into the war would probably do poorly with Scandinavian-American voters, many of whom were pacifists and/or didn't like going to war with German fellow-Lutherans. Cox got 19.4 percent of the vote in MN in 1920, and Scandinavians as well as Germans contributed to that result.)