I'll say it again for about the 20th time on here, you need more information to be able to form that assumption. Something like 30-40% of people report that they change the party they vote for from one election to the next so say for instance Party A got 10,000 votes last time and 10,000 votes this time, you are just guessing if you form the assumption that these must be the same people. Now if you have some kind of data around who they voted for last time, you have some kind of chance although that is self reported and people have a habit of lying if they voted for somebody they now don't like.
Any pollster that says any different is not worth listening to because they are just trying to find something to say to sound interesting.