Beyond Turnout: Understanding Where Voters Roll Off the Ballot
By Lindsay Veazey and Sara Stephenson
Getting a voter to the polls is only part of the challenge. The next issue is whether that voter makes it all the way downballot.
In elections across the country, some voters cast ballots for high-profile races but skip contests farther down the ticket. For state parties, legislative campaigns, and other political organizations, understanding where that happens can provide a valuable layer of intelligence for targeting and GOTV strategy.
TargetSmart’s new Roll-off Score is designed to help identify where voters are more likely to exhibit that behavior.
What is Voter Roll-off?
For this analysis, TargetSmart defines roll-off as the share of voters who cast a ballot but did not cast a vote in a downballot state legislative race.
The analysis uses election results at the census-block level and averages roll-off across valid election years for each state’s legislative chambers.
Put simply: Where are voters participating in an election, but skipping the state legislative race?
The national numbers show just how significant that question can be.
TargetSmart’s analysis found an average state legislative roll-off rate of 16.44% nationally. That means that on average, about 16 out of every 100 ballots cast did not include a vote in the state legislative race analyzed. There is also considerable variation between states. New Jersey had an average roll-off rate of just 5.6%, while Nebraska and South Carolina were among the states with the highest non-anomalous averages at 21.7% and 20.7%, respectively.
That variation exists within states as well.
Turning Election Results into Actionable Scores
Rather than treating roll-off as simply a statewide statistic, TargetSmart calculated roll-off propensity at the census-block level and then assigned voting-age individuals a decile score from 1-10 based on their block.
The scores are relative within each state. Deciles 1–3 represent census blocks with higher-than-average roll-off, while voters in deciles 7–10 live in blocks with comparatively lower roll-off. In Minnesota, for example, we can see how dramatically those patterns can differ.
In heavily Democratic HD-59B, 80.1% of individuals fell within the three highest roll-off deciles. By comparison, just 3.2%of individuals in swing HD-3B fell into those same deciles, while 77.6% were in the four lowest roll-off deciles.
That kind of geographic variation can give campaigns a more sophisticated way to think about downballot participation.
From Turnout to Downballot Participation
The Roll-off Score becomes especially useful when combined with other voter intelligence.
For example, a campaign could combine roll-off scores with partisanship and turnout scores to identify likely supporters who are also expected to vote but live in areas with historically high levels of downballot roll-off. TargetSmart’s example combines high partisanship scores of 80+, turnout scores of 70+, and roll-off deciles below 3 to develop a potential GOTV audience.
Instead of asking only, ‘Who is likely to turn out?’ campaigns can begin asking another question:
Who is likely to vote, but may need additional encouragement to continue downballot?
That distinction can help campaigns and state parties build more precise targeting strategies, particularly when downballot races could be decided at the margins.
And roll-off is only one part of the picture.
Next week, we’ll look at TargetSmart’s Democratic Ticket-splitter Score and how historical voting patterns can help identify areas where Democratic support at the top of the ticket doesn’t necessarily carry through to races farther downballot.
Beyond Turnout: Understanding Where Voters Roll Off the Ballot
By Lindsay Veazey and Sara Stephenson
Getting a voter to the polls is only part of the challenge. The next issue is whether that voter makes it all the way downballot.
In elections across the country, some voters cast ballots for high-profile races but skip contests farther down the ticket. For state parties, legislative campaigns, and other political organizations, understanding where that happens can provide a valuable layer of intelligence for targeting and GOTV strategy.
TargetSmart’s new Roll-off Score is designed to help identify where voters are more likely to exhibit that behavior.
What is Voter Roll-off?
For this analysis, TargetSmart defines roll-off as the share of voters who cast a ballot but did not cast a vote in a downballot state legislative race.
The analysis uses election results at the census-block level and averages roll-off across valid election years for each state’s legislative chambers.
Put simply: Where are voters participating in an election, but skipping the state legislative race?
The national numbers show just how significant that question can be.
TargetSmart’s analysis found an average state legislative roll-off rate of 16.44% nationally. That means that on average, about 16 out of every 100 ballots cast did not include a vote in the state legislative race analyzed. There is also considerable variation between states. New Jersey had an average roll-off rate of just 5.6%, while Nebraska and South Carolina were among the states with the highest non-anomalous averages at 21.7% and 20.7%, respectively.
That variation exists within states as well.
Turning Election Results into Actionable Scores
Rather than treating roll-off as simply a statewide statistic, TargetSmart calculated roll-off propensity at the census-block level and then assigned voting-age individuals a decile score from 1-10 based on their block.
The scores are relative within each state. Deciles 1–3 represent census blocks with higher-than-average roll-off, while voters in deciles 7–10 live in blocks with comparatively lower roll-off. In Minnesota, for example, we can see how dramatically those patterns can differ.
In heavily Democratic HD-59B, 80.1% of individuals fell within the three highest roll-off deciles. By comparison, just 3.2%of individuals in swing HD-3B fell into those same deciles, while 77.6% were in the four lowest roll-off deciles.
That kind of geographic variation can give campaigns a more sophisticated way to think about downballot participation.
From Turnout to Downballot Participation
The Roll-off Score becomes especially useful when combined with other voter intelligence.
For example, a campaign could combine roll-off scores with partisanship and turnout scores to identify likely supporters who are also expected to vote but live in areas with historically high levels of downballot roll-off. TargetSmart’s example combines high partisanship scores of 80+, turnout scores of 70+, and roll-off deciles below 3 to develop a potential GOTV audience.
Instead of asking only, ‘Who is likely to turn out?’ campaigns can begin asking another question:
Who is likely to vote, but may need additional encouragement to continue downballot?
That distinction can help campaigns and state parties build more precise targeting strategies, particularly when downballot races could be decided at the margins.
And roll-off is only one part of the picture.
Next week, we’ll look at TargetSmart’s Democratic Ticket-splitter Score and how historical voting patterns can help identify areas where Democratic support at the top of the ticket doesn’t necessarily carry through to races farther downballot.
