They're under 50%, but Democrats should feel good about their Senate chances in Pennsylvania
Posted June 15, 2018 11:07 p.m. EDT
(CNN) — First things first: The theme song of the week is The Facts of Life from the television show The Facts of Life.
The poll of the week: A new Franklin and Marshall poll that puts Democratic Sen. Bob Casey ahead of his Republican opponent Lou Barletta 44% to 27%. It's the latest poll that has Casey with a double-digit lead over Barletta.
What's the big idea: The new Franklin and Marshall poll is bad news for Republicans with hopes of picking up a Senate seat in Pennsylvania, where President Donald Trump won by less than a percentage point.
The race pits Barletta, who was an early backer of Trump's and is someone who sees eye-to-eye with the President on the issue of immigration, against Casey, a staunch critic of the President.
Barletta's campaign put some interesting spin on the poll numbers last week, arguing that it reflects poorly on Casey that he couldn't crack 50% in the poll and that a majority of voters are either for Barletta or are undecided.
"There's a lot to be optimistic about in this poll for someone challenging a two-term incumbent senator who has been in statewide office for two decades," his campaign spokesman David Jackson said.
However, in Senate general elections, it's the size of the lead that matters most, not how many voters are still left undecided.
As FiveThirtyEight's Nate Silver showed using earlier data, a look at polling over the last 12 years suggests it's far more important to look at the margin by which the leading candidate candidate is ahead as opposed to the percentage of the vote they are garnering. Further, there really isn't a reason to make a distinction between whether the person ahead is an incumbent or not.
Since 2006, there have been 45 Senate incumbents who led in the polls in a January to June average in the year of the election but had less than 50% of the vote. They won a little more than 75% of the time.
Now, winning 75% of the time is less than the average re-election rate for incumbents, which tends to hover between 80% and 95%, depending on the year. Still, that's not a major difference.
The big statistic you should look at is the size of the margin that those under 50% have over their challengers, not the fact that they are below 50%.
Since 2006, there have been 48 incumbents polling at greater than 50% of the vote. All of them had leads of at least 13 percentage points over their opponents. All but George Allen of Virginia in 2006 won re-election. Allen, of course, was embroiled in controversy for referring to a campaign tracker as "macaca", a term seen as offensive.
Now let's look at the incumbents who were polling below 50% and were up by at least 13 percentage points like Casey currently is in Pennsylvania. There were 9 nine of these incumbents. All 9 nine of them won. John Cornyn and Cory Booker were two of them in 2014.
You can even expand your search to those incumbents who were polling below 50% and were up by 10 points or more. Senators in this situation were 15-for-15 in their re-election bids.
Indeed, a 10-point lead in early Senate polling seems be a sort of magic cutoff for judging whether a lead is relatively safe or not, regardless of whether it was an incumbent, challenger or a candidate in a race without an incumbent. In the 87 races where Senate candidates had leads of at least 10 percentage points, they have won 86 times. This includes 21 elections of 21 elections where the leader was polling below 50%.
Obviously, the past isn't always prologue in American politics. But when you look at all 112 incumbents for whom we have early polling, it suggests that someone in Casey's polling position should win about 97% of the time. That's a pretty safe lead, even if Casey is under 50%.