Primary debates are usually a letdown. Viewers may tune into the debate expecting a WWE-style brawl complete with suplexes and clotheslines, but they usually get a bunch of white men in their 50s talking over one another. And after a couple of hours, viewers and journalists can usually only remember a couple genuinely interesting, unexpected interactions — moments that often get lost in the next few news cycles and fail to really change public opinion.

But debates aren’t always disappointing. In the fluid stages of the campaign, debates can produce a media narrative that gives a long-shot candidate a temporary bump or, maybe more important, contribute to the rise or fall of a front-runner. The 2020 Democratic race is in that early, fluid stage: Media narratives aren’t set, and voters aren’t familiar with the field yet. That makes the debate the perfect chance for a couple of candidates — but probably not more than that — to at least temporarily change their trajectory.

The average primary debate doesn’t usually change much about the dynamics of a primary. I collected the RealClearPolitics polling averages for the 2012 and 2016 Republican primaries, the most recent big primary brawls of the social media era, and calculated how much each candidate’s support shifted between each debate and a week after the event.

The result: The average candidate’s standing shifted by a grand total of 1.4 percentage points. Many candidates followed the path of Bobby Jindal, who seemed like a smart, capable, accomplished person but couldn’t use debates, combative tweets or anything else to heave himself out of obscurity. Some candidates broke that pattern: Carly Fiorina had a short surge in 2015 after earning media attention for a solid debate performance, and Rick Perry’s standing in the polls started to slide in 2011 when he emotionally defended the Texas DREAM Act in a debate. In retrospect, maybe that should have been a warning about the role immigration would play in the 2016 Republican primary. But for the most part, candidates who seemed hopeless in the early going weren’t able to change their fortunes via debate.

But the situation for Democrats now might be a little bit different. Unlike in 2012, when the second and third tiers of the field were mostly filled with improbable also-rans, the 2020 Democratic field has a lot of quality candidates waiting in the wings. Julián Castro and Sens. Michael F. Bennet (Colo.), Kirsten Gillibrand (N.Y.) and Amy Klobuchar (Minn.) have profiles that ought to make them serious competitors, but they’re each registering 1 percent or less support in the RealClearPolitics 2020 Democratic national poll average. Rank-and-file Democrats still don’t know much about these or many of the other candidates, and it’s easy to imagine some qualified Democrats both surging and having some staying power after a good debate performance.

The media dynamics of 2020 might also make the post-debate landscape a bit less predictable than it has been in the recent past. Most voters come into general elections knowing which party they prefer, and usually they’ve watched each major-party candidate either fight through a tough primary or govern for four years. But voters don’t have the North Star of partisanship in a race where everyone is a Democrat, and they are unfamiliar with many of the major candidates. They rely on media outlets to inform them about their choices.

So far, former vice president Joe Biden has gotten more media attention than other candidates. Despite some of the coverage he’s attracted for controversial statements, that has probably helped him: He’s ahead in the polls, and voters are hearing more about him than either challengers for his specific lane of the party, like Klobuchar, or Biden’s progressive opponents, like Sen. Bernie Sanders (I-Vt.). But if the debates refocus media coverage on some other candidate, that person could steal the media spotlight from Biden, introduce themselves to voters and make considerable gains.

But the debates will change only so much.

While examining the post debate polling from 2012 and 2016, I took a quick swing through the archives at Memeorandum, a news aggregator that roughly tracks the conventional wisdom in the news media by finding articles that other news sites link to. After debates, journalists tend to come up with a couple of clear story lines relatively quickly. For example, right after the September 2015 Republican primary debate in Simi Valley, Calif., journalists at a variety of outlets quickly decided that Carly Fiorina had a good night but Jeb Bush and Donald Trump did not. Immediately afterward, Fiorina surged and Trump dropped. The polls didn’t really move for most of the other candidates. The same pattern tended to repeat itself: A post-debate story that involved only a couple of candidates got traction, those candidates saw their numbers move a bit, and the candidates who weren’t part of that story didn’t gain or lose much support either way.

So these first debates could change something about the current shape of the race, but they probably won’t change everything. And subsequent debates probably won’t turn this race completely upside down, either. According to the Economist’s calculations, the candidate who led polls a year before the convention (mid-July this cycle) has won the nomination in roughly half of modern presidential primaries. By July 2015, Trump had begun to surge, Jeb Bush and Scott Walker were showing signs of weakness, Ben Carson would soon start to climb and savvy observers were watching Sens. Marco Rubio (Fla.) and Ted Cruz (Tex.).

It’s early, but it’s not that early. Candidates who are polling poorly may enter the top tier as time goes on, but the list of candidates who make it through the early states will probably bear some resemblance to the current top tier.

Even if they don’t decide the Democratic nomination, much less the 2020 general election, the debates definitely matter: They’re the first opportunity for 20 of the Democratic candidates to have a real conversation about both policy and where the party should go politically. More than that, they’re the first big campaign event, outside of candidate announcements, that voters will pay attention to and really learn from. If history is a reliable guide (and that’s a genuinely big if), they won’t fully restructure the race. But they don’t need to transform this race into a Royal Rumble between former Colorado governor John Hickenlooper, New York City Mayor Bill de Blasio, Rep. Tulsi Gabbard (Hawaii) and Rep. Eric Swalwell (Calif.) to tell us a lot about the state of the Democratic Party and the nation.

Read more:

Donna F. Edwards: What I’m looking for in the Democratic debates

E.J. Dionne Jr.: Thursday will be a big deal for both Biden and Sanders

Jennifer Rubin: Who’s rising, who’s falling and who’s invisible

Jennifer Rubin: Want to win your presidential debate? Follow these 10 rules.

Paul Waldman: The 2020 Democrats will debate soon. Here’s why it will be awful.