The Republican presidential campaign has already provided plenty of drama and entertainment, but things are about to get even more interesting with the start of actual voting. The first event, the Iowa caucuses, take place in the depth of a mid-western winter on January 3.

The key thing to understand about Iowa is that its intrinsic importance is negligible — only a small number of delegates are at stake, Iowa is untypical of the Republican electorate at large, and the idiosyncratic voting system makes in more unrepresentative still. Instead, it’s mostly about perception: what a candidate’s performance, relative to expectations, does for their momentum, including media attention and fund-raising ability.

Current expectations are that Newt Gingrich will score a clear but not overwhelming victory, that Mitt Romney and Ron Paul will be roughly equal second, and that Rick Perry and Michele Bachmann will be the best of the others some distance back. For more detail, see the polling averages at RealClearPolitics and the forecast modelling by Nate Silver at FiveThirtyEight.

Things often move quickly in primary contests, so the landscape could change over the next two weeks. But if those expectations are borne out — and for what it’s worth, they seem reasonable to me — it’s hard to see how anyone other than Gingrich or Romney can win the nomination.

Perry depends on the Gingrich balloon deflating so that he can take his place as the anti-Romney candidate. To make a go of it, he would need to get into the top three in Iowa, be well clear of Bachmann, and have Gingrich do significantly worse than expected. For Bachmann, who appeals to much the same sort of voter, the task is much the same, but her lack of establishment support makes it even harder. Whichever of them loses to the other in Iowa is probably doomed.

Paul is relatively stronger in Iowa but even more disliked by the establishment. He needs to win Iowa or at least dead-heat with Gingrich, and preferably have Gingrich and Romney perform below expectations; that would allow him to hope for maybe a second-place finish in a crowded New Hampshire field, after which the media might start taking him seriously.

John Huntsman, who is not a serious contender in Iowa, depends on Romney being badly beaten there to give him traction in New Hampshire. If Romney can be fatally weakened in the early contests, respectable Republican opinion may rally to Huntsman as the only sane candidate remaining.

At the moment, none of these scenarios looks very likely. More probably, Romney will go on to win New Hampshire on January 10, Gingrich will win South Carolina on January 21, and both will stay in the running at least until super Tuesday on March 6.

Gingrich, however, is not looking as strong as he was a week or two ago.

It may be that, like Perry, Bachmann and Herman Cain before him, he has peaked too early. Silver last week suggested that “the race remains quite fluid, but that fluidity may no longer be working to Gingrich’s benefit”, and that his lack of money and of high-profile endorsements limits his ability to respond to his opponents’ attacks now that he is a more visible target.

However, as Silver also points out, none of this seems to have translated into stronger support for Romney, whose national numbers remain stubbornly stuck below 25%. So while Romney deservedly remains the favourite overall, it’s unlikely to be plain sailing for him — even with what would otherwise seem the inestimable advantage of having Newt Gingrich as his prime opponent.