3. Shorter early-round series
This is something that’s been tried in the past. For years, the first round series was a best-of-five. In bygone eras, the NBA Playoffs opened with best-of-three miniseries. Eventually, owners realized that they were punting revenue away and expanded to the current format. It’s why this proposal would never happen. Frankly, the NBA would likely go to best-of-nine before it would go to best-of-three or even best-of-five.
But we’re not dealing exclusively in reality here. Our goal is to make the playoffs fun.
This would do that.
The first round would be a best-of-three. The first and third games of each series would be on the home floor of the higher-seeded team, while the second would be played on the lower-seeded team’s home court.
The exceptions would come in the two 1 vs. 8 series. We’d give the one seeds a choice. Option 1, play the aforementioned home/road format. Option 2, host the first two games. That would give the top seeds a chance to win without ever having to leave their home arenas. But it would also mean a potential decisive Game 3 would be on the road.
The second round would then be a best-of-five, with all series played in the 2-2-1 format.
This would be more fun for a few reasons.
One, upsets are significantly more likely when dealing with a shorter series. Over an extended period of time, the best team will prevail far more often than not. But in a short series, anything can happen.
Two, if nothing else, it would end some of the more lopsided series faster. Quite frankly, that can’t be a bad thing.
At this point, we’d suggest playing both the Conference Finals and NBA Finals with the same best-of-seven format (2-2-1-1-1) that we all know today. We could say that the Conference Finals are best-of-seven while the NBA Finals are best-of-nine. But while some recent NBA Finals have been classics (2013 and 2016), others have been relative duds (2014 and to a lesser extent, 2012 and 2015), and really, dragging those on for an extra game or two would not have been a good thing.