Romney will not get to 1,144 delegates (a majority) by the end of the primaries and caucuses assuming he keeps accumulating delegates at his current pace.
Now, good academic that he is, Seth immediately assures us that this trendline is not a prediction. Why?
There's no real reason to believe Romney will continue to acquire delegates at this same pace this year. Romney looks likely to do well in Illinois tomorrow, as well as in some large winner-take-all states like New Jersey and California* later on. (And I like him for the winner-take-all Utah primary.) [*Note: California's primary is winner-take-all by state and congressional district. . . .]
So my expectation is still that Romney will have this thing wrapped up before the primaries and caucuses are over. But it may take some time -- NJ and CA aren't until June 5th.
In an update later in his post, he notes that Illinois didn't make much of a dent to the trendline.
Seth also mentions that having multiple candidates, particularly Gingrich, stay in the race has extended the time it takes for any one candidate to win a majority. But, he fails to make the critical follow-up point to that musing:
If Gingrich had left the race, Romney would not have nearly the lead that he does right now. And, when and if Gingrich leaves the race, Santorum's position in the remaining primaries and caucuses will be strengthened.
Gingrich's effect on this race has been to split the conservative vote which would otherwise be great enough to defeat Romney. If Gingrich gets out, most of his supporters will back Santorum unless Romney looks completely inevitable (e.g. if Romney already has enough delegates to win the race). Santorum doesn't have to play to win. He simply has to play to deny Romney a majority in Tampa at the first ballot. If Santorum succeeds in that task, the Republicans will have a brokered convention and all bets are off.
I'm not sure that I think that there will be a brokered convention in Tampa either. But, it is quite a bit more likely than Seth suggests.