It started, as things tend to do, at the beginning.

One of the first states to report results on Tuesday evening was Florida, where Democrats had two candidates vying at the statewide level that embodied the cycle: a long-term senator defending his seat and a progressive black gubernatorial candidate marking a new wave of Democratic politicians. The early vote in the state was massive, suggesting, perhaps, a surge in enthusiasm. All this talk about the blue wave: Would we see it crash over Florida in the first few hours of the night?

Well, no. In both the Senate and governor’s races, there was no runaway lead for either Democrat. At the end of the night, Sen. Bill Nelson (D) was trailing Gov. Rick Scott (R), and Tallahassee Mayor Andrew Gillum (D) had conceded to Rep. Ron DeSantis (R). Where was this blue wave? Was this 2016 all over again?

Well, sort of. In 2016, the early vote suggested that Democrats might have a good night, only to see voters enthusiastic about President Trump’s message turn out on Election Day to give him the victory. But, unlike in 2016, much of the rest of the 2018 midterm election was good news for Democrats.

But that might not be how they look at it.

Florida. Florida is one reason. A quick Gillum victory in particular would have seemed like a mark of Democratic dominance for the night. His race was also about tied in polling, but optimists figured that the surge of Democrats would carry him to victory. (See also: O’Rourke, Beto.) Instead, Gillum trailed (and ultimately lost), dampening the optimism from the outset, turning Dem energy into yet another cycle of Dem depression. That feeling lingered for hours, even as the party slowly started to accrue the sort of gains in the House that had been predicted.

What happened was what was expected. Those polls were another reason the Democrats were bummed. They’d heard about a Democratic wave and probably saw projections like those at FiveThirtyEight that predicted that the Democrats would pick up more than 30 seats in the House. For many, those expectations — of a big shift in the House — became the baseline, the expectation.

Many more Republican-held seats were considered vulnerable this year. See live U.S. House results

Meeting an expectation doesn’t seem like a wave. A wave feels energetic, like a constant push of energy. Picking up one House seat after another, like clockwork, doesn’t have the same feel as suddenly winning contests such as the 5th Congressional District in Oklahoma.

People focused on Senate candidates. That’s in part because we associate politics with politicians. It’s much easier to pick out senatorial and gubernatorial candidates than House candidates, in part because they have bigger constituencies, and in part because those seats hold more power than House seats. Before Election Day, attention was on a handful of Senate seats that might make the difference in determining control of that chamber. The aforementioned Rep. Beto O’Rourke (D-Tex.), Sen. Heidi Heitkamp (D-N.D.), the close fight for the Senate seat in Arizona. There were a few House races that burbled up to national attention, maybe, but generally eyes were on the people running for other seats.

Other seats like those ones in the Senate — where Democrats didn’t do all that well. The map was always stacked against the party in the Senate, as a result of the party’s success in 2012, when Barack Obama was reelected. Six years later, a lot of incumbents were in vulnerable territory. But Democrats hoped for a miracle, hoped that the projections about the House (overwhelmingly showing Democrats taking control) were correct and projections about the Senate (showing Republicans holding it or even gaining seats) were wrong.

The projections were not wrong. Republicans had a good night in the Senate, in large part because of a good Senate map that had Democrats playing defense. That Democrats probably held the seats in Montana and West Virginia is, in that context, remarkable. But Democrats expected something beyond expectations.

California. When Hillary Clinton conceded the 2016 presidential election, her lead in the popular vote was fairly small, not much more than a tie.

But California kept counting ballots. Two weeks after the election, there were still millions to be counted. As those came in, Clinton’s lead grew and grew. We now regularly say that she won the popular vote by nearly 3 million ballots, but that perception came well after Election Day.

There are two reasons California takes a long time to count ballots. The first is that people can return their absentee ballots the day of the election, including having them postmarked that day. The other, of course, is that almost 1 out of every 8 Americans lives in California. That’s a lot of votes to count in general, much less with them still trickling in even as I write.

What it means is that a number of the close races in California won’t be settled for weeks. The full scale of the Democratic shift in the House won’t be clear until, what, December? It’s on track to match expectations of about 35 more seats gained. But we won’t know for a while.

Imagine if, when polls closed Tuesday, the results were known immediately. Democrats gained 35 seats in addition to hundreds of legislative seats at the state level and a slew of gubernatorial races. They lost a few red-state Senate incumbents, and the high-profile candidates that had earned so much attention didn’t get the boost they needed to win.

But, had that happened, how might the night have been seen?