Previous Election Cycles

Return to the Michigan Homepage Here.

My analysis covers data from the 2022 midterms, the most recent Senate general election, and the 2026 Senate Democratic primary. Because Republican nominee Mike Rogers ran uncontested, I will not analyze his race results.

However, the previous race was during a Presidential election cycle, not a midterm, which means turnout—particularly for younger voters—will decrease, making the data less useful for comparison. There has not been a midterm with a Senate election in Michigan since 2018. Furthermore, Elissa Slotkin, a moderate, likely attracted a different demographic compared to the group Abdul El-Sayed will attract in this midterm cycle. Because of these differences, information from previous polls will not be as representative for El-Sayed and Rogers' races.

For past polling aggregates, I will reference polling averages from 270 to Win. Click here to visit their page.

Understanding Data From Previous Election Cycles

Taken together, these elections support comparing how the outcome changes under lower, typical, and higher participation from younger voters. Because previous election cycles do not provide an accurate representation of likely turnout for different age demographics, I will use this cycle's polling to predict the most likely outcome but provide alternative likely outcomes based on demographics.

Previous elections provide useful information, but none directly match 2026. The 2024 Senate election occurred during a presidential cycle, while midterms generally attract lower turnout, particularly among younger voters. Slotkin and El-Sayed may also draw different coalitions due to their differing political views. The 2022 midterms can suggest how turnout might look in the 2026 midterms, but Michigan had no Senate contest that year, leading to a different turnout group. These differences limit how confidently I can predict age-based turnout in 2026.
Predictions based on demographics proved accurate. In the 2026 Democratic primary, Ruth Igielnik of The New York Times found that "polls did seem to understand the underlying dynamics of the [primary]," accurately capturing El-Sayed's success with young voters and Stevens' strength with Black voters and suburban communities. This suggests that demographic patterns are accurate predictors even when polls overestimate a candidate’s lead, meaning my age-based analysis of polling data can prove useful
Polling historically underestimates Republican support. In the 2024 general election, the 270toWin polling average accurately predicted Elissa Slotkin’s support but underestimated support for Mike Rogers (who is also the Republican nominee in 2026) by nearly four percentage points, turning an apparent four-point lead into a very narrow victory. In the midterms, Rogers may end up with a higher vote share than is expected in polling. Importantly, Elissa Slotkin is more politically moderate than El-Sayed, meaning these polling inaccuracies may not reflect the outcomes in 2026.
Michigan has relatively strong youth participation. Michigan led the country in youth turnout in the 2022 midterms, with roughly 37% of eligible 18- to 29-year-olds voting. Nevertheless, the state’s report found that 59.2% of active registered voters in that age group did not cast a ballot. Strong support among younger voters therefore represents potential electoral strength whose impact depends on participation.

2024 Senate General Election

In the 2024 general election, while polling accurately predicted Ellissa Slotkin's vote share, it underestimated Mike Rogers' vote share by nearly four points, suggesting that undecided voters ultimately sided with Rogers.

What This Means

Should the same polling outcome hold true in the 2026 Senate election, polling for El-Sayed may remain accurate, while polling for Mike Rogers will end up underestimating his vote share. However, Slotkin's narrow victory with less than 50% of the vote share suggests that—as a result of third-party candidates—El-Sayed's lead in polling can still translate to an overall victory. However, as I mentioned previously, the significant differences between Slotkin and El-Sayed, as well as the differences between midterm and Presidential cycle elections, make this data less useful in this particular outcome.

2022 Midterms – 18-29 Year-Old Turnout

While there was not a Senate election during the 2022 midterms, Michigan nonetheless held statewide elections due to their gubernational election. During the midterm cycle, Michigan had the highest voter turnout for 18- to 29-year-olds in the country at 37%. 556,173 people aged 18-29 cast a ballot, representing approximately 12 percent of the total number of votes cast. This higher turnout rate will likely prove beneficial for El-Sayed, who performs better with voters under 50.

2026 Primary Election Results

Primaries overestimated El-Sayed’s margin in his primary by approximately twelve points. This is considered typical for primary elections for U.S. Senate and governor, where the typical aggregate poll is off by around 13 percentage points because voting patterns become much less consistent.

Why Was Polling Off?

Per the New York Times:

Conducting surveys in partisan primaries has long presented a challenge to pollsters, as it requires them to make assumptions about who will actually turn out to vote. Without mandatory party registration in Michigan, pollsters must guess who will show up on Election Day.

Many surveys assumed political independents would make up 30 to 40 percent of the electorate. These polls showed Dr. El-Sayed winning independents by a wide margin. While there is little authoritative detail available on the actual electorate, that assumption could have contributed to the challenge in polling the race, as voters who decline to identify with a political party tend to vote less frequently.

…

Polls did seem to understand the underlying dynamics of the race. In pre-election polls, Dr. El-Sayed did particularly well with younger voters; on election night, he performed particularly well in areas with a high concentration of young voters. Polls also accurately captured Ms. Stevens’s strength with Black voters and in suburban communities.

What We Can Learn

Even though the margins for voter turnout were off, data analyzing popularity among specific demographics was accurate. The 2026 Democratic primary midterms show that polling predictions on El-Sayed and Rogers' strength with voters in various communities (those with a high concentration of young voters, black voters, or other demographics) will be a fairly accurate predictor of the real election. This general accuracy suggests that, by separating data specifically by age (as this poll does), we can predict different possible outcomes based on age-based turnout.

Other States

Maine
Texas
Alaska
Florida
Georgia
North Carolina
Nebraska