Pew has an interesting tool for this; from that it look like Nevada is the biggest loser:
http://www.pewsocialtrends.org/2009/03/11/magnet-or-sticky/
Just having a glance at Florida, using some of the slightly outdated numbers I've found here: http://www.bebr.ufl.edu/files/FloridaPop2005_0.pdf around 9,716,500 of the Florida population was born out of state while around 4,313,100 have left. Putting those numbers into the 2000 census (without accounting for how many of those who left have died) you get a population of 10,578,987 in 2000. I'm sure I've done something wrong somewhere, but I'd doubt its TOO far off.
Interesting. Here's the full "magnetic" and "sticky" list:
http://www.pewsocialtrends.org/2009/03/11/sticky-states/Using CA as an example.....
38.0% of its residents were born in another state. But that 62.0% of residents who were born there only account for 69.0% of all people born in CA.
So, .62/.69 = .90
The # of people born in CA is then only 90% of its current population.
Likewise, that would be 88% for Texas, 146% for New York, 45% for Florida, 125% for Illinois, 124% for Pennsylvania, 115% for Ohio.....
And so if you believe those #s, then no, Florida is not one of the top 5 states by # of people currently alive who were born there, despite now being 3rd in population overall.
The top 5 states by # of people born there would be:
1) CA
2) NY
3) TX
4) IL
5) PA
Of course, that's probably wrong, because they say these migration #s only apply to adults, and I'm also not sure how they deal with international migration, as opposed to internal migration within the US.