There is a so-called "red shift", but it's caused by unreliable exit polls leaning blue, not fraud causing the actual results to shift red.
If you had asked whether polls in general showed a Democratic lean compared to the results, you would be getting a very different answer. But since you specifically asked about exit polls, this is all about them.
This article, which dates from shortly before the 2008 election, discusses 10 reasons that exit polls aren't reliable. They may be informative, but you can't draw any statistical conclusions from them, even for something as basic as who won. I'll quote sections that specifically address the "red shift" issue below.
1. Exit polls have a much larger intrinsic margin for error than regular polls.
This is because of what are known as cluster sampling techniques. Exit polls are not conducted at all precincts, but only at some fraction thereof. Although these precincts are selected at random and are supposed to be reflective of their states as a whole, this introduces another opportunity for error to occur (say, for instance, that a particular precinct has been canvassed especially heavily by one of the campaigns). This makes the margins for error somewhere between 50-90% higher than they would be for comparable telephone surveys.
It's a fact of life that almost no media coverage of poll results ever discusses the error margins. But that's where almost all the fluctuation in polling comes in. If one poll predicts a candidate will get 47% of the vote +/- 1%, and another predicts that candidate will get 52% of the vote +/- 5%, then when the second poll is released all the media coverage will be about how much the candidate's position has improved, even though the error margins are so high that (in the absence of any other information), the candidate is really getting closer to 48% of the vote (and is thus likely losing). Because exit polls have even higher error margins than normal polls, their numbers are that much less reliable.
Error bars on statistical results are of huge importance. I can say right now, over a year out from the next presidential election, that the Republican candidate will receive 51% of the vote, +/- 49%. If the media were reporting on my prediction, they'd say that I said the as-yet-unnamed Republican was ahead... but I really said that it was way too uncertain to say anything at all (other than that the Republican will get votes, because my floor was 2% total).
2. Exit polls have consistently overstated the Democratic share of the vote.
Many of you will recall this happening in 2004, when leaked exit polls suggested that John Kerry would have a much better day than he actually had. But this phenomenon was hardly unique to 2004. In 2000, for instance, exit polls had Al Gore winning states like Alabama and Georgia (!). If you go back and watch The War Room, you’ll find George Stephanopolous and James Carville gloating over exit polls showing Bill Clinton winning states like Indiana and Texas, which of course he did not win.
This builds on the first point. Exit polls are unreliable, and (for reasons discussed below) lean Democratic. Thus if the unreliable numbers are skewed one way, then the actual results will appear to be skewed the other in comparison. Thus what's really happening is not that the votes are suddenly going more red than they should, but the exit polls are bluer than they should be.
4. Exit polls challenge the definition of a random sample.
Although the exit polls have theoretically established procedures to collect a random sample — essentially, having the interviewer approach every nth person who leaves the polling place — in practice this is hard to execute at a busy polling place, particularly when the pollster may be standing many yards away from the polling place itself because of electioneering laws.
5. Democrats may be more likely to participate in exit polls.
Related to items #1 and #4 above, Scott Rasmussen has found that Democrats supporters are more likely to agree to participate in exit polls, probably because they are more enthusiastic about this election.
These two mostly stand by themselves, but together they sortof explain why exit polls lean Democratic. In a busy polling place, even one where there's an even mix of people, if someone makes themselves more approachable by the pollster, they're more likely to be polled. If more Democrats than Republicans are willing to be polled (the link to the proof of which is dead in the article, and the reasons are not discussed), then the results will obviously skew Democratic.
6. Exit polls may have problems calibrating results from early voting.
Contrary to the conventional wisdom, exit polls will attempt account for people who voted before election day in most (although not all) states by means of a random telephone sample of such voters. However, this requires the polling firms to guess at the ratio of early voters to regular ones, and sometimes they do not guess correctly. In Florida in 2000, [absentee votes were substantially Republican], leading to an overestimation of Al Gore’s share of the vote, and contributing to the infamous miscall of the state.
7. Exit polls may also miss late voters.
By “late” voters I mean persons who come to their polling place in the last couple of hours of the day, after the exit polls are out of the field. ... this adds another way in which the sample may be nonrandom, particularly in precincts with long lines or extended voting hours.
So both people who voted before election day itself and those who come close to closing time aren't necessarily well represented. I couldn't find any statistics on how these groups tend to vote overall, but they're certainly enough to introduce even more error into the exit poll results (even if they balance out from one year to another).
In short, any claims that are based off of the difference between exit poll results and the actual results of an election are only showing how unreliable exit polls are.
The article I quoted from finishes with a good summary:
10. You’ll know the actual results soon enough anyway.
Have patience, my friends, and consider yourselves lucky: in France, it is illegal to conduct a poll of any kind within 48 hours of the election. But exit polls are really more trouble than they’re worth, at least as a predictive tool. An independent panel created by CNN in the wake of the Florida disaster in 2000 recommended that the network completely ignore exit polls when calling particular states. I suggest that you do the same.
