UPDATE: 5/2/2017 12:26PM
New poll released by Gravis moments ago shows Trump up by +17 over Cruz. More proof the outlier Downs Poll is garbage. Poll dates 4/28 and 4/29.
Tomorrow, May 3, 2016 is a huge day for the GOP Presidential Primary races. It is a make or break day for the candidates. Should Donald Trump win tomorrow, the race is over. Momentum and a sense of inevitability will set in with a Trump victory. Going forward, a Trump win will dry up funds for Ted Cruz and we will see the masses move to Trump. And a loss for Cruz would essentially end his campaign.
So it is potentially a “make” day for Trump and a potential “break” day for Cruz.
And we have seen some good and not so good polling this primary season.
Taking a look at the Real Clear Politics Averages we see:
As you can see, the last 6 polls have Trump with a lead, and the single outlier is the IPFW/Downs Center poll that has Cruz with a 16 point advantage.
When I see these wild swings in a poll I usually check out the source and break down the poll myself to determine the methods and potential validity.
The Downs Poll, right off the bat shows that they conducted the poll between 4/13/2016 and 4/27/2016. A two week period, which means that they called less than 30 people a day. the rest of the polls show their polls conducted during a 1 to 4 day period. Two weeks is an long time in a primary election. And things happen fast and public opinion moves, sometimes quickly. And the internal numbers in the polls are not available. But one interesting bit of information was: the poll was not weighted. Which means that the poll results are raw numbers.
What is weighting? Precision Polling defines the process:
It is frequently the case that the people who answered your poll are not fully representative of the region you were polling over. Weighting is a technique to adjust answers to account for over- and under-represented groups.
For example, suppose the population in an area is 55% female; but 70% of my survey answers come from women. When reporting my survey results, I’d want to to boost the importance of men’s answers a bit, and reduce women’s answers. Weighting is the process to do this.
During the first week of the polling – April 13-19 – Ted Cruz had some momentum with wins in Utah and Wisconsin the previous week. The next contest was Trump’s home state – New York, but not until the 19th. And by the time the polls closed and the results were known, it was past 9 PM Eastern so the calls stopped before Trump’s huge win was announced. Now everyone expected Trump to win his home state, but the margin was more than expected. The RCP average had Trump winning 53% of the vote, but he won over 60%. So half of this poll was taken prior to Trump’s huge win in NY.
And fast forward to a week later when Trump won 5 states in major fashion. He outperformed every poll and the momentum he gained from that is just starting to be measured. And the Downs Poll stopped their poll the day after Trump’s big momentum shifting win the previous day. So 13 of the 14 days during which this poll was conducted were prior to the 5 state sweep in which Cruz came in last place in 4 of the 5 states.
This poll ended the day Cruz names Carly Fiorina as his VP pick, a decision about which NBC News said, “Given that must-win situation, they’re trying every trick in the book (early VP pick, alliance) to win. But when you throw that “Hail Mary” and the pass falls incomplete, the game is usually over.”
And then there was the Kasich – Cruz pact/alliance to stop Donald Trump. Another Hail Mary pass that fell incomplete according to 58% of Indiana voters who disapprove if this move.
And while Ted Cruz was pandering to an Indiana group gathered in a gym the day before the Downs poll ended, Cruz did the unthinkable in the state where basketball is king. He called the basketball hoop a “ring”.
And of course, one of the biggest announcements if you are a Hoosier was Bobby Knight “The General” endorsing Donald Trump on April 28, the day after the Downs Poll ended.