South Dakota and California: What One Recent Poll Can and Cannot Tell You
Dated editorial analysis; figures in this article are not recalculated automatically. Visit the linked race pages for current data.
Not every race gets polled every week. In some states, the only recent survey is also the only one in our tracker for the past month. This piece looks at three contests where exactly one verified poll ended within 14 days of our October 28 data date: the South Dakota Senate and governor races, and the California governor's race. All three polls are from Emerson College Polling, which gives us a consistent method to compare, though not a polling average.
Why a single poll is not an average
Our averages use the mean of verified polls ending in a 14-day window. With one poll, the "average" is just that poll, and it carries that poll's quirks: its sample, its question wording and its definition of a likely voter. Margin of error also matters. A survey of about 800 people has a sampling margin around plus or minus three and a half points on each candidate's share, and the margin on the gap between two candidates is roughly double that. Our guide to reading polls and margins of error covers the arithmetic.
South Dakota
Emerson surveyed 803 likely voters from October 19 to 21, using the initial ballot without pushing undecided respondents toward a candidate.
- Senate: Sen. Mike Rounds (R) 50%, Brian Bengs, listed on the ballot as an independent, 42%. That is an eight-point gap with the remainder undecided or choosing others.
- Governor: Gov. Larry Rhoden (R) 56%, Dan Ahlers (D), a former state senator, 38%. That is an eighteen-point gap.
The two races sit in the same poll, which makes the difference between them informative. Rounds is seeking another Senate term and Rhoden is seeking his first full term as governor after taking office in 2025. The same respondents gave the Republican candidate a larger share in the governor's race than in the Senate race. Why is not something the poll can say. It could reflect how respondents view the individual candidates, differences in name recognition, or simple sampling variation. Both Cook Political Report and Sabato's Crystal Ball currently rate both contests safely or solidly Republican, which describes a rating, not a vote count.
California
Emerson surveyed 1,000 likely voters from October 17 to 19 for the California governor's race. Xavier Becerra (D), a former U.S. Health Secretary, was at 60%, and Steve Hilton (R), a former TV host, was at 33%, with 7% undecided.
California's primary is a top-two contest: the two candidates with the most votes advance regardless of party. In the June primary, Becerra and Hilton finished first and second, and reports at the time showed the order changing as late mail ballots were counted. That history is a reminder that California counts slowly. Ballots postmarked by Election Day can arrive afterward, and counties continue tabulating for days or weeks, so a state-level result can shift after election night. Our guide to how results are counted goes into the general mechanics.
Why a large gap is not the same as a certain result
Wide gaps are less sensitive to polling error than narrow ones, but they are not immune. Polling error varies by state, by year and by pollster, and in lopsided states there are often fewer polls to cross-check one another. Other things a single poll leaves open include whether the sample of likely voters is the right mix of the eventual electorate and whether undecided voters break evenly. In South Dakota, where 803 respondents cover a small statewide population, subgroup results such as by region are particularly rough.
How to use a poll like this
- Treat it as one reading of the race on its fielding dates, not as a forecast.
- Check whether it reports undecided voters and how it handled them.
- Compare with earlier polls in the state if any exist; in these three races, none ended in the 14-day window besides this one.
- Wait for additional surveys before drawing conclusions about movement.
As more polls are verified and added to the tracker, the race pages update with each pollster's dates, sample sizes and sources.
Poll data as of October 28, 2026. Only verified polls ending within 14 days of that date are counted in averages; each race page lists the sources.
Relevant race pages: South Dakota Senate · California Governor · South Dakota Governor