Polling average nailed the Markey/Gomez Massachusetts Senate result

On Tuesday, June 25, Congressman Ed Markey defeated financier Gabriel Gomez with a 54.8% share of the vote to Gomez's 44.8% share. A normalized polling average of the 17 independent polls of the race had Markey with 55.5% and Gomez with 44.5%, less than a percentage point difference.

While it is easy, in hindsight, to look back and pick the individual polls that were close to the final result, the result being close to the average of all polls is significant. Averaging polls smooths out differences in polling methodology and sampling variability and error, and gives a result over a much larger pool of voters. Using the average prevents us from needing a crystal ball to know which poll is significant, and which one is off base.

Political reporters or pundits will often quote the most recent poll, or sometimes cherry pick a poll that tells the story they are trying to tell. Many quoted the June 19 Herald poll with Markey up 20 points as a sign that Markey was on his way to a blowout. It is important to look at outliers like these in the context of the other polls, and polling averages give us a tool to better understand all of the data points.

In the following graph the yellow and green lines represent the final average of the normalized polling numbers and the black squares show the final results, just inside the average lines.

There is a great deal more analysis that can be done of the Markey/Gomez results, but it is clear that polling of the race and the resulting averages gave us a pretty clear picture of the state of the race before yesterday's voting.

A comprehensive pre-election overview of the Markey/Gomez polling

There have been 16 independent polls of the match-up to replace John Kerry in U.S. Senate between Congressman Ed Markey and financier Gabriel Gomez spanning the beginning of May through June 20th. Markey has led every poll and average of all polls gives Markey 49% to Gomez's 39% (a margin of 10 points) with 13% undecided. Normalizing the results by evenly distributing undecideds in a poll-by-poll fashion gives Markey an average of 56% to Gomez's 44%.

The results of the polls have been remarkably stable and the candidates' normalized share has never been more than 5 points above or below the final average. It only takes one look at the charts below with the polls numbers slightly above or below the marked average line to get a feel for the stability of the polling.

The demographic numbers in the polls suggest that Markey and Gomez are receiving equal support from men, while Markey will take the majority of women voters, with about a 20 point advantage. Gabriel Gomez would need to win the votes of many Democratic voters and also take Independents by a very large margin in order to win the election. However, the polling suggests that Gomez will win Independents by just over 10 points.

Enough with the words—here are the numbers and the graphs that tell the story.

The normalized version of the polls that split the undecided vote can give us some idea what the final vote share of each candidate might be. Averages suggest Markey receiving 56% of the total vote, with Markey getting 44%. It is significant that only the UMass Lowell/Boston Herald Poll of June 19 is more than 3 points away from the final average. The small sample of only 312 likely voters may explain the Herald outlier.

Another remarkably stable graph:

The polls have also been quite consistent relative to gender and party affiliation. Markey and Gomez are splitting the male vote, while Markey has held a consistent 20 point lead with women.

With respect to party affiliation, Markey is winning 84% of Democrats, while Gomez is getting 88% of Republicans (the actual result will probably be higher as there seems to be more consolidation of the Republican vote as the election nears). The polls show Gomez up about 11 points with Independents, not nearly enough to overcome the Democratic over Republican registration advantage.

Given the consistency of the polling, a win by Gabriel Gomez on Tuesday, June 25 would be an incredible upset, and an indication that something was systematically wrong with the polling. All signs point to a sizeable Markey win.

Polling Data

• Emerson College Polling Society - 5/2/2013
• Public Policy Polling/League of Conservation Voters - 5/3/2013
• MassINC/WBUR - 5/6/2013
• Suffolk University - 5/7/2013
• Public Policy Polling/League of Conservation Voters - 5/15/2013
• Emerson College Polling Society - 5/22/2013
• New England College - 6/2/2013
• Public Policy Polling/League of Conservation Voters - 6/4/2013
• YouGov/University of Massachusetts - 6/5/2013
• Suffolk University - 6/9/2013
• MassINC/WBUR - 6/9/2013
• Harper/Conservative Intelligence - 6/11/2013
• Globe/UNH - 6/14/2013
• UMass Lowell/Boston Herald - 6/19/2013
• Emerson College Polling Society - 6/20/2013
• Western New England University/Masslive.com - 6/20/2013

A Gomez win over Markey very unlikely even if GOP targets are reached

An article written by Alexandra Joffe in The Hill congressional newspaper reports that the Republican Party sees a narrow path to victory for Gabriel Gomez against Ed Markey in the Massachusetts special election to replace John Kerry. However, the numbers don't add up—a Gomez win is very unlikely, even if he achieves the GOP targets described in the article.

The current aggregated polling averages for the Markey/Gomez race predict that Gabriel Gomez will win 16% of the Democratic vote, 87% of the Republican vote, and 54% of the Independent/Unenrolled vote as indicated in the table below. Combined with a likely turnout (also based on the polling averages) of 38%/14%/47% for D/R/I puts Gomez with a likely result of 44% of the vote on June 25.

The Joffe article says "Republicans familiar with the race say they’re aiming to win 55-60 percent of independents and 15-18 percent of Democrats to take the seat."

Let's suppose that Gabriel Gomez hits the high end of those targets, winning 18% of the Democratic vote and 60% of the Independent vote, giving Markey 82% of Democrats and 40% of Independents. These Democratic and Independent margins would still only multiply out to 48% of the final vote for Gomez and a winning 52% of the vote for Markey.

However, the polling has been remarkably consistent throughout the race, and it is more likely that the Gomez will win Independents by around 10 points. There is very little time for Gomez to make up his large gaps with Independent women, and Markey has a very high probability of winning the special election on June 25.

Latest Markey/Gomez polls show very slight tightening of race

Markey still holds strong lead with two weeks left

The latest polls from Suffolk University and WBUR/MassINC of the Massachusetts special election race between Ed Markey and Gabriel Gomez provide some evidence of a tightening race. A comparison between a simple average of all of the polls, with a weighted average giving stronger weighting to the most recent polls, shows Markey moving from about a 9 point lead over Gomez, to an 8 point lead. However, this is hardly good news for Gomez as candidates rarely overcome an 8 point deficit with two weeks until an election.

If you have been following the Mass. Numbers Markey/Gomez polling average page, you have seen that there have been very few surprises as new polling results are released.

In addition to calculating the simple average of the polls, I have also given weighted average, based on the number of days since the last day the poll was conducted. This weighted average shifts the margin from about 9 points to about 8 points. While this does indicate a slight tightening of the race, the polling has been remarkably stable, with all of the polls moving very close above and below the average.

The fact that Congressman Markey continues to hang close to even with men, and to be winning women by about 17 points, combined with Gomez's inability to increase his margin with Independent and Unenrolled voters, means that a Gomez win continues to be a very, very long shot.

Latest Markey/Gomez poll is more of the same: evidence of a solidifying Markey lead

Yesterday, New England College released another poll of the Ed Markey/Gabriel Gomez special senate election race and it does little to change the evidence that Congressman Markey has a sizable lead. The party and gender breakdowns from the NEC poll also align well with previous surveys of the race. Financier Gabriel Gomez is probably hoping for a game changing performance in the upcoming debates, but time is running out and all signs point towards a Markey win on June 25.

Averaging the polls together evens out inconsistencies in polling methods and gives us a much larger sample of voters. The average is largely unchanged by the poll, although the number of undecided voters in the NEC poll is smaller than all of the previous polls, most likely because of the polling question and methodology. The Markey advantage continues to hover around 10 points.

We can normalize the polls to get a better feel for what an actual election result might be like by choosing a method to divide up the undecided voters between the two candidates. I have chosen to divide them up equally. Normalizing the polls also gives us some more consistency between the polls. We again end up with about a 10-point lead for Markey.

The new poll is also consistent with our averages in terms of Markey's margin among men and women. The NEC poll shows Markey even with men (the average is Markey +1) and has Markey +20 with women (our average was +16). As I wrote in a previous post, Gomez will have to completely change these margins in order to have a chance on June 25.

The NEC poll is also reasonably consistent with the averages in terms of party breakdown. The higher percentages for both Markey and Gomez in each category primarily come from the undecided percentages, rather than cutting into each other's vote share. I wrote in a previous post that Gomez cannot win by turning out more Independent voters, but rather needs to increase his margins with both Democrats and Independents to stand a chance.

The latest poll brings more of the same with a solidifying lead for Ed Markey that will be very hard for Gabriel Gomez to overcome in the short time before the June 25 special election.

Gomez will have a hard time overcoming Markey's 16 point lead with women voters

Markey has consistent lead in the numbers and on the issues

There have been six public polls of the Massachusetts Senate special election between Congressman Ed Markey and financier Gabriel Gomez, and an averaging of the polls shows Markey with a one point lead among men and a whopping 16 point lead among women. Markey holds positions on women's reproductive health and gun control issues that are more popular with the majority of Massachusetts women than those held by Gomez, making it unlikely that Markey will relinquish that lead.

A recent New York Times article detailed Markey's advantage with women and his efforts to solidify and increase his lead among Democratic and Independent women by emphasizing Gomez's unpopular positions on women's health issues and assault weapons. It is clear from the polling of the race that Gomez cannot win without moving his margin with men into the positive double-digits and significantly cutting into Markey's advantage with women, an unlikely scenario given Gomez's stances.

Over the time period of the six polls, Markey has moved from a small disadvantage with men in the first three polls, to a small advantage. He has held a significant double-digit lead with women in every single poll of the race.

Martha Coakley, the Democrat vying against Scott Brown in the 2010 special senate election, also had an advantage among women, according to the last several polls of the race. However, her advantage was about two points, while Brown held a large double-digit 13 point lead over Coakley with men. There were no exit polls of the January 19, 2010 vote.

Gabriel Gomez would need to come close to these sorts of margins to stand a chance on June 25, but the numbers seem to be moving in the opposite direction. Ed Markey's margin with women is solid in the double-digits, and his standing with men has improved, not worsened. Gomez would most likely need a game changer to pull out a win, while Markey has the easier task of making sure likely women voters understand Gomez's positions on the issues.