How much do the Republican Party’s chances in the general election depend on whether Trump or DeSantis is the nominee? Let’s consult the people who have put money on it. Combining two separate election-betting markets — one about who will win the Republican nomination, and the other about who will win the general election — yields the following probabilities of becoming president if nominated: Ron DeSantis enjoys a 60 percent chance of winning if nominated, vs. 37 percent for Donald Trump.

Betting markets are not omniscient, of course, but they do produce a convenient distillation of the conventional wisdom among people who are not pure partisans. So let’s take these estimates seriously. The market is telling us that DeSantis as the nominee has a six in ten chance of becoming president, while Trump as the nominee has less than a four in ten chance.

That gap is remarkably large. Traditionally, historians of presidential-election outcomes have emphasized the “fundamentals” — economic growth, foreign affairs, incumbency advantage, etc. — and put relatively less weight on candidate quality. For example, Allan Lichtman’s 13 “Keys to the White House” contain just two keys about candidate quality, and in his estimation, only two recent candidates (Reagan and Obama) have garnered significant advantage from their charisma. The point is not that there is no such thing as a bad candidate; rather, it’s that the two major parties tend to avoid nominating them. When both parties’ candidates are acceptable to the median voter, it’s mainly the fundamentals that determine the result.

Under the fundamentals model, the electability of mainstream contenders within the same party should be similar. They should be expected to perform in the general election roughly as if they are “generic” members of their party. Obviously, however, bettors do not see the GOP race as a battle between two generic Republicans. In the absence of historical data, I cannot say definitively how anomalous the perceived Trump–DeSantis electability gap is at this juncture. Nevertheless, the market clearly does not view Donald Trump as a candidate with broad enough appeal to win a general election in many environments in which Ron DeSantis would.

Republican voters will have to decide: Either the market is skewed by ignorant bettors unaware that elections are about the fundamentals, or Trump as the nominee would be so unpopular with centrists that he breaks the historians’ models.