Like It or Not, It’s Play-In Time

As the NBA playoffs approach, the added “play-in” tournament has been hotly debated over the past few weeks. Opponents argue that the tournament discredits the 72 game season. Why should a team, who played well in many cases and finished as the 7th or 8th seed, be forced into a situation where they are in jeopardy of making the playoffs. (Disclaimer: making the play-in tournament does not mean you have made the playoffs). Proponents contend that the play-in games add much suspense to a part of the season that historically had been bland and boring. Basketball Reference has the Lakers as a 83% chance of being in the play-in tournament as the 7th seed. That undoubtedly would be very exciting. And from the league’s perspective, more excitement equals more revenue. Similar to MLB’s wild card play-in game (which I think is just horrendous, but that is a conversation for another time), any additional teams in the playoff hunt leads to an increase in money coming into the league. It also combats part of the tanking issue.

I thought it would be a good idea to explore how much this play-in tournament matters in the grand scheme of the NBA playoffs. While any cut-off point when looking at data is somewhat arbitrary, I decided to go back to 2007 because many have described this as an inflection point in NBA history where the “modern NBA” was created with the 7 second or less Phoenix Suns.

I think it is important to begin by noting that since 2007, only 4 upsets have happened in the first round of the playoffs and it hasn’t happened since 2011. Any argument that a potential 9th or 10th seed would “ruin” the playoffs is much overstated.

For those who aren’t aware, here is the format for the play-in games. The 7th seed in each conference will host the 8th seed. The winner of this game will be the 7th seed. The 9th seed in each conference will host the 10th seed. The loser of the seven-eight game will host the winner of the nine-ten game and the winner of that game will be the 8th seed in the playoffs for its conference.

To begin the analysis, I looked at the net ratings and overall record for 7th through 10th seed in each conference to understand how each team compared against the others.

Here is the West:

Here is the East:

I then took the average differences in the net ratings and wins between each seed in order to quantify the disparity between the matchups.

In many of the years it is clear from the standings and from the net ratings that the 10th seed is considerably worse than the other 3 teams. It seems rather inequitable that the 10th seed have an opportunity to make the playoffs when in many cases have been far behind their 7th, 8th and sometimes even 9th seed counterpart.

Furthermore, I think the 9th seed getting in over the 8th seed wouldn’t make the 1–8 matchup any worse than it would have been had there been no play-in tournament. The 8–9 seeds since 2007 have been pretty close together both in net ratings and in the standings.

The next question is what are the chances that the 10th seed could upset the 9th seed or the 8th seed could upset the 7th seed? (Or really any other of the combinations after the seven-eight and nine-ten games are played). Since this is a one or two game elimination format (one game for the 9 and 10 seed and two games for the 7 and 8 seed), there is much more randomness than the regular best of 7 series. I thought it would be prudent to try to assess single game outcomes rather than series outcomes.

I looked at every single playoff and Finals game since 2007 and calculated the differences in net rating between the teams playing and then binned those differences. I then calculated how often the team with the higher net rating won for every single game thereby giving me a clearer picture of game to game winning percentage.

For example, if the Cavs had a net rating of 7 and they were playing the Hawks who had a net rating of 2, the difference would be 5 and would be binned in the (4,5] bin. The Cavs are the “better” team in this scenario because they have a higher net rating. So the question became how often does the “better” team win? To illustrate further, when the difference in net ratings between the teams was between 4 and 5, the team with the higher net rating won 66% of the time. The numbers on top of each bar represent the count for how many occurrences are in that bin.

As the difference between net rating increases, the “better team” wins a larger percentage of the games, which makes intuitive sense. However, it is important to keep in mind that there is a steep drop off in occurrences as the the difference in net rating increases. But that doesn’t necessarily mean a drop off in win percentage. For the (11,12] bin, there are only 7 occurrences with the better team winning 57% of the time. Using a smoothing function, that number is closer to 80% with more occurrences.

So where does this leave us? Let’s assume the play-in games will happen as the standings are right now. That means Lakers vs. Warriors and Grizzlies vs. Spurs in the West. And Celtics vs. Hornets and Pacers vs. Wizards in the East. Basketball Reference has this as the mostly likely outcome. I ran a simulation for a 1000 occurrences based on the probabilities from the historical data of the differences in net ratings between teams to determine how the play-in games would unfold.

The 7th and 8th seed in each conference have overwhelmingly better odds to make the playoffs than the 9th and 10th seed but this isn’t absolute. The Pacers making the 8th seed 32% of the time isn’t nothing.

The last question to answer is will any of this have a material impact on the playoffs? LeBron James flies in the face of data to some extent. If the Lakers end up as the 8th seed, and we assume the Suns remain as the 1 seed, there is a 6 point difference in their net ratings. In those games, the better team wins almost 75% of the time. But if LeBron is healthy, it has never mattered what seed he is going into the playoffs. There are seeds in the NBA Playoffs, and then there is wherever LeBron is. The two rarely interact with each other. I’d have a hard time believing the Suns being thrilled with the prospect of playing the Lakers in the first round. Then the question becomes if the Lakers do beat the Suns in the first round, is that really an upset?

Regardless of whether or not the play-in games will have any material impact on the playoffs as a whole, they will be exciting nonetheless. This year in particular, the 7th and 8th seed in the West right now is no cake walk for any opponent. So maybe we will finally see an upset (a real one, at that) in the first round after 10 long years. There is clear excitement around these games (I am writing this article, after-all). Like it or not, I think it’s a fair guess to say that the play-in tournament will be around for awhile.