Philip Klein thinks not (via Andrew):

However close the polls may look in Alabama and Mississippi, evangelical voters are a very active and dependable voting block that is likely to come out in big numbers tomorrow given the closeness of the race, so it’s unlikely that Romney will be able to pull off a victory if the past is prologue.

According to Rasmussen, Romney leads in Mississippi while he appears to be trailing in Alabama. One of the reasons for this is that he does much better among non-evangelical Protestants (45-25%) and Catholics (41-24%) in Mississippi than in Alabama (32-33% and 38-35% respectively), and he wins slightly more evangelical support in Mississippi (27% vs. 25%). Among “very conservative” Mississippians, he receives 31%, but just 21% of “very conservative” Alabamians support him. He runs in second among “very conservative” respondents in Mississippi, but trails badly in third place in Alabama.

While Romney has significant leads among his natural base of supporters of “somewhat conservative” (36-28%) and moderate/liberal voters (44-26%) in Mississippi, his margin among “somewhat conservative” voters in Alabama is smaller (34-30%), and his share of the moderate/liberal vote is smaller (36%). Whatever image of the Mississippi electorate one might have, it’s useful to revisit the 2008 exit poll to see the ideological composition of the primary electorate. According to CNN’s exit poll, 34% of the 2008 Republican voters identified as “very conservative,” 28% as “somewhat conservative,” 28% as moderate, and 10% as liberal. Mississippi appears to be friendlier ideological territory for Romney than Alabama, and tomorrow’s results are likely to reflect that with a Romney win in Mississippi and a narrow loss in Alabama.