Another week, another polling disaster.
Progressive candidates in two big-name Democrat primaries – one in Wisconsin, one in Michigan's U.S. Senate race – both walked into Election Day thinking they had it in the bag.
The polls told them so.
Democratic socialist Francesca Hong was supposedly up by roughly 20 points in Wisconsin heading into the primary.
Abdul El-Sayed was riding a similar wave of good press in Michigan.
Then the votes got counted.
Suddenly those “commanding leads” evaporated, the establishment candidates closed the gap or won outright, and the political press corps was left scrambling to explain what happened.
Here's what happened: The polling industry is broken. Has been for years.
Nevadans don't need me to tell them that – we've watched it happen cycle after cycle.
Even the pollsters' own defenders can't spin their way out of this one.
One Democratic pollster who worked the Michigan race admitted El-Sayed's supporters were simply more eager to answer pollster phone calls than the other guy's voters were – which skewed the whole sample.
Translation: The polls didn't measure the electorate. They measured who likes talking to pollsters.
That's not a minor technical glitch. That's the entire product not working.
A veteran Wisconsin political reporter put it bluntly on social media, saying primary polls shouldn't be treated like breaking news because Wisconsin primaries are brutally hard to poll.
Fine. But nobody tells readers that up front.
Every poll gets reported as gospel truth the moment it drops, complete with a horse-race headline and a candidate anointed the front-runner.
Campaigns get covered based on the polling snapshot, not the fundamentals. Donors make decisions based on it. Reporters build entire narratives around it.
And then, when the numbers turn out to be garbage, everybody shrugs and says “well, it was just a snapshot in time.”
A snapshot in time that just so happened to shape the whole campaign.
This isn't a new problem for those of us who've been in the trenches of Nevada politics for three decades.
We've seen “unbeatable” leads vanish and “sure losers” pull off upsets more times than I can count.
Polling has always had a shelf life shorter than a Vegas buffet shrimp cocktail.
What's changed is how fast bad numbers spread now.
Social media aggregators grab a poll, slap a dramatic headline on it, and it's everywhere within minutes – regardless of sample size, methodology or how stale the data already is by the time it's published.
Even Democrats – who love nothing more than a good poll showing their candidate crushing it – are starting to admit the emperor has no clothes.
One Democratic strategist and candidate recruiter says the party needs to reckon with how broken primary polling has become before the wide-open 2028 presidential primaries arrive.
She's right. But it's not just primary polling. It's polling, period.
Response rates have collapsed. Nobody answers unknown numbers anymore.
The people who do answer skew toward a certain kind of voter – often not representative of the electorate that actually shows up.
Add in late-breaking news, shifting campaign dynamics, and pollsters guessing at turnout models, and you've got a discipline that's more art-project guesswork than science.
So here's my advice heading into Nevada's 2026 cycle:
Ignore the polls. Ignore the horse-race headlines. Watch what campaigns are actually doing – who's raising real money, who's building an actual field operation to turn out low-propensity voters.
That's where elections get won. Not in some pollster's spreadsheet.
The polls will keep telling stories. Just don't be surprised when the ending doesn't match the trailer.
The opinions expressed by contributors are their own and do not necessarily represent the views of Nevada News & Views. Digital technology may have been used in the research, writing, or production of this article. Please verify information and consult additional sources as needed.
