TargetSmart numbers: Could Hillary be heading toward a landslide Florida?

254348-jpg-resize-710x399Last week, TargetSmart released a poll stating that 28% of Republicans in early voting are voting for Trump. Overall, this gave Clinton a 55%-37% advantage in the state. Of course, the sample size was only 311. But what if we applied the percentages that TargetSmat came up with to the actual early vote numbers? Would the same apply?

As of right now, 5.69 million people have voted in Florida, with 39.66% Democrats, 39.53% Republicans, and 20.18% NPA/Other. If we take the TargetSmart percentages by party and apply them to the actual numbers, will there be a change? If the TargetSmart numbers deviate from the actual numbers when we apply their percentages, that means that they over- or under-sampled a specific group of people. So, if Hillary received 55% in the poll, but when you apply the numbers she only received 50%, then TargetSmart got something wrong.

So, how did they do? Here is TargetSmart’s numbers from their poll:

Clinton: 55%, 90% Dems, 28% Rep, 40% Oth
Trump: 37%, 6% Dems, 66% Rep, 43% Oth
Johnson: 3%, 0% Dems, 4% Rep, 4% Oth
Stein: 3%, 3% Dems, 0% Rep, 8% Oth
Other: 2%, 1% Dems, 2% Rep, 4% Oth

If we apply those percentages to the vote total (with those who identify as Democrats in the poll being synonymous with registered Democrats), here is what you have:

Candidate DEM REP OTH Total %
Clinton 2041798 (90%) 610573 (28%) 476130 (40%) 3128501 55.02%
Trump 136120 (6%) 1492513 (66%) 511839 (43%) 2140472 37.65%
Johnson 0 (0%) 90455 (4%) 47613 (4%) 138068 2.43%
Stein 68060 (3%) 0 (0%) 95226 (8%) 163286 2.87%
Other 22687 (1%) 45228 (2%) 47613 (4%) 115527 2.03%
2268664 2238769 1178421 5685854

The numbers show that TargetSmart is dead on when it comes to this. If, let’s say, that the amount of Democrats that turned out to vote was only 2.1 million (which is only .7% less of the Democratic vote), yet we still applied the 90% number to Clinton’s total, her percentage would drop down to 53.95%. So even the slightest inaccuracies would be noticed in this application. Therefore, it seems as if TargetSmart is “on target” (pun intended).

Over the last four elections, the voter turnout has been between 70% and 75%. Therefore, let’s conservatively figure that overall turnout would be around 72%, which means that 27% of registered voters still have not voted. If this holds true, Trump would need to win 65.9% of the vote to catch up to Clinton and have an Election 2000 scenario.

With this being the case, the places to find these votes are starting to narrow. This is especially the case if we look at the large Republican counties. In Collier, 66% of Republicans already voted. In Sumter, 74% of Republicans already voted. Some counties in the panhandle that favor Republicans are under the average, but overall, Republican votes are starting to run out. However, the same cannot be said for Democrats. In Broward County, only 44% of Democrats have turned out. The number is 46% in Osceola County, 40% in Alachua, 38% in Leon, 44% in Orange, 37% in Palm Beach, 48% in Miami-Dade, and 43% in Duval.

Basically, the likelihood is that Democrats might hold both a pre-Election Day advantage and a post-Election Day advantage. If that is the case, Clinton could win the state by over 10%…easily.

2 thoughts on “TargetSmart numbers: Could Hillary be heading toward a landslide Florida?

  1. A survey with a sample size of 311 is meaningless, because of the huge margin of error. Further, even if you had a sample of 1million, that doesn’t tell you how Reps and NPAs will vote on election day. If, as I suspect, they are more conservative, then the final results will be close.

  2. It was just part of a larger sample, around 800 I think. But TargetSmart was more accurate in regards to the percentage of early/vmb numbers compared to the other polls.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s