By: Rik Mukherjee
Every year when the first round of the NBA Playoffs come around, basketball fans rejoice, as the regular season ends and the “important” part of the season begins. During the playoffs, stars are made or broken, narratives are built, drama is felt. It is meant to be the highest level of basketball imaginable.
Logically, however, the playoffs are an extension of the regular season. The teams with the best records in the league are going to be looked at as the title contenders before the playoffs begin, while a team that barely squeaks in as the eighth seed will be looked at as a team who is simply happy just to be there. For this reason, we would expect the regular season series between two teams to dictate a playoff series between them. In the NBA, teams play the other 14 teams in their conference, either 3 or 4 times per season (except for shortened seasons in 2019-2020 and 2020-21). Based on how well Team A plays against another Team B during the season, we would expect Team A, who won the season series against Team B, to also win a playoff series against Team B if the two teams were to be matched up in the playoffs. For the most part, this is exactly how playoff series go. But for series where this is not the case, what causes the difference between the season series outcome and the first round playoff series outcome? Let’s take a look at how this phenomenon has taken shape since 2014-15, which will be the earliest year of data used for the purposes of this study.
This pie chart shows that the team who wins their season series will win their first round playoff series over 53% of the time. This certainly lines up with the assumption of what I expected to see. However, there are a sizable chunk of season series that were tied, which means that the two teams had a 2-2 record against each other during the regular season.
If we remove the season series ties from this chart, which do not favor any specific team when it comes to playoff series, we now see a far more overwhelming advantage in winning first round playoff series when a team wins their season series.
So now that it has been established that winning a playoff series against a team is pretty strongly related to winning the season series against that same team, it will be interesting to see whether winning a season series relates to winning a first round playoff series as an underdog team.
Surprisingly, among first round playoff series upset wins, winning a season series against a higher seeded team does not seem to strongly relate to the eventual result. Perhaps it could be said that lower seeded teams winning or tying their season series against the higher seeded team correlates with a playoff series win almost 64% of the time, but even then, this means that approximately 36% of the time, a team that upsets a higher seed in the playoffs still lost their season series. This creates a really interesting dichotomy between the fact that winning your season series strongly relates to winning your first round playoff series, but does not strongly relate to a first round playoff upset if the team that wins the season series is the lower seeded team. This could perhaps be attributed to the fact that higher seeded teams will usually win both the regular season series and the playoff series.
As in the case of higher seeded teams, it is far more common that higher seeded teams who won their first rounds playoff series also won or tied their season series (almost 87% of the time). Even then, a majority of season series winners who are the higher seed win their first round playoff series.
All that being said, the only extremely clear and persistent trend is that winning the season series most usually end up leading to winning the playoff series.
Therefore, we can say as our general rule, if two teams play each other during the regular season and later meet up in the first round of the playoffs, the team that wins the season series is more likely to win the playoff series.
Since 2015, only 11 series have had a contradiction to this general rule of thumb, so let us go through the two most contradictory series to our rule of thumb to see if there are any specific trends that affect these series and lead to these differences from what we would expect to occur.
Sports Illustrated: https://www.si.com/nba/2015/04/27/paul-pierce-wizards-raptors-sweep-game-4-john-wall-kyle-lowry
This sweep is perplexing to me, and completely contradicts our general rule of thumb. Let us see what was so drastically different between the regular season and playoff matchups between these two teams.
This is a scatterplot of how the Wizards players in 2014-15 played against the Toronto Raptors in regards to their points per game (PPG) in comparison to what their season average was. I used t-formulas, which would give me an idea of how far off the PPG averages of the three game sample size against the Raptors during the regular season compares with the way the Wizards played on average during the regular season as a whole. Overall, we can see that there were some players playing above average, and others playing below average, but there’s nothing too shocking in the sense that Wizards star and leading scorer John Wall didn’t play wildly above nor below expectations.
There’s two caveats with this: Bradley Beal only played one of the three games against the Raptors during this 2014-15 regular season, and since he was the team’s second leading scorer, that clearly played a role in his team not playing well against the Raptors in the regular season. Secondly, Ramon Sessions was added as a trade deadline acquisition after the season series against Toronto had already concluded, so he is not a factor in this set of data, even though he played in and affected the outcome of the playoff series against the Raptors with 6.5 PPG and a +/- of +25 throughout the span of the series as a whole.
Ramon Sessions and Will Bynum were two mid-season acquisitions for the Wizards who did not play against the Raptors in the regular season, so their contributions helped the Wizards offense increase their points per game. On top of that, Bradley Beal and Marcin Gortat had a much higher scoring average in the postseason than they did in the regular season series against the Raptors, and as the team’s 2nd and 3rd leading scorers respectively, this was certainly a big help in the Wizards winning their playoff series against the Raptors.
As far as the Raptors side of things, their stars and two leading scorers DeMar Derozan and Kyle Lowry did not play particularly better or worse than expected when compared to their regular season points per game during the season series against the Wizards. Jonas Valanciunas, however, had an abysmal season series against the Wizards in comparison to his season average for points per game, and was a major outlier from the rest of his team.
*Patrick Patterson, not Peterson
Here is where we kind of see that the Kyle Lowry underperforming in the playoffs narrative (aside from 2019) may be warranted. In this series against the Wizards, as his team’s second leading scorer during the regular season, he put up about 4 points per game less than he did in the regular season series against the Wizards. Lou Williams also turned in a subpar playoff series against the Wizards, where despite having the best t-statistic in points scored against the Wizards compared to expected in the regular season, he massively underperformed in the playoffs.
The last thing I want to mention about this season series is that even though the Raptors swept the Wizards, the last two of their regular season games against each other were won by an average margin of 3 points, so the Raptors won this season series quite closely overall. So even though we should have expected a more competitive series in the playoffs, the Wizards going on to sweep the Raptors is truly shocking. Definitely an anomaly to our big general rule.
2019: #5 Portland Trail Blazers defeat #4 Oklahoma City Thunder 4-1, (OKC won regular season series 4-0)
Blazer’s Edge: https://www.blazersedge.com/2021/4/23/22399689/damian-lillard-shot-thunder-2019-nba-playoffs-game-5-buzzer-beater-video
This series is perhaps just as surprising as the Wizards sweep of the Raptors in 2015. The Thunder swept the season series against the Portland Trail Blazers, and I remember specifically when this matchup was set on the last day of the regular season, a majority of NBA fans said this was a horrible matchup for the Trail Blazers, and the regular season track record between the two teams certainly must have played a role in those predictions. So what went so wrong for the Thunder to have their hearts ripped out by Damian Lillard and the Blazers in only five games?
Initially, things may look sort of odd on this graph. Not included in this graph is Enes Kanter, who was a deadline day acquisition so he only played one game against OKC as a member of the Trail Blazers in the 2018-19 regular season, and that gave him an unrealistically low t-statistic, so I did not include him in this chart. The worst performer compared to his average was Mo Harkless, whose t-statistic hovered around -5, which meant he played quite a bit worse than expected in the regular season series against the Thunder.
In the playoffs, Jusuf Nurkic was not available for the Blazers after suffering a lower body injury, so it would be assumed that the complimentary pieces of the Blazers stepped up in his absence to provide the scoring load that would be necessary to defeat the Thunder.
So it can be seen that losing Jusuf Nurkic results in 15.6 points that need to be made up by the rest of the team, and they very nearly are able to up their production accordingly. Mo Harkless played far better in the playoffs than he did in the regular season series against the OKC Thunder, and Enes Kanter and Rodney Hood played above their expectations. CJ McCollum also contributed more than expected when compared to his regular season series average against the Thunder, where he had two above-average scoring outputs and two below-average scoring outputs. In the playoffs, however, he was able to be far more consistent and find his scoring output increase by almost 5 points a game. When summing all these changes in points per game together, we get -0.7 points, which is quite close to the 0 points that would have meant the team perfectly made up for all the points lost from the Nurkic injury. So if the Blazers didn’t necessarily increase their offensive output in the playoffs, how did they defeat the Thunder who swept them in the regular season?
When looking at this chart, we can see that Paul George and Russell Westbrook, the Thunder’s stars and two leading scorers, both played quite a ways better than they normally did when playing the Blazers during the 2018-19 regular season. Along with this, the Thunder got an 18 point performance out of 2.6 PPG scorer Deonte Burton in one of their games against the Trail Blazers and also a 15 point game out of Raymond Felton who averaged 4.3 PPG across the entirety of the regular season. It is safe to say that those games, as well as the increased points per game of Paul George and Russell Westbrook, helped the Thunder in their sweep of the Blazers during the regular season.
In the playoffs, both Paul George and Russell Westbrook came down to earth on their scoring averages against the Portland Trail Blazers. When all these changes in points scored are summed up, the Thunder ended up with a -11.65 change in points per game output compared to their regular season series against the Portland Trail Blazers.
So clearly this difference in the amount of points that the Thunder’s stars were scoring ended up making the playoff series between the Blazers and Thunder go towards the Blazers favor. When looking at the point differential of the regular season games between the two teams, all four games ended with a single-digit point differential. So similar to how the Wizards played the Raptors very tough in their regular season series, the Trail Blazers did the same, and then drop-offs in points scored led to the Blazers being able to make up for the difference. Still, it is shocking that this series ended in a convincing five games and in the Blazers’ favor.
Conclusion
Through these two series, we saw the four seed 2014-15 Raptors and 2018-19 Thunder sweep their opponents, the five seed Wizards and Trail Blazers respectively, but fail to beat them in the playoff series that followed. We observe that during the regular season series between the teams, each game was very close, with both season series of 2014-15 Raptors vs. Wizards and 2018-19 Thunder vs. Trail Blazers had single digit average point differentials across all three or four games.
Another similarity we can observe is that one of or both of the leading scorers of the four seed teams (Kyle Lowry for the Toronto Raptors and Paul George & Russell Westbrook for the Oklahoma City Thunder), were not able to put up the same scoring output in the playoffs as they were able to in the regular season series against their five seed opponents of the Wizards and the Trail Blazers. In order to find out why this would be the case, we would end up having to deviate from strict statistical analysis, and go into film study on adjustments the coaches made to change the defense shown to these star players and cause them to have a tougher time scoring at the same output they did in the regular season series against the same opponent.
To address potential flaws of the study, it is clear that points per game is not always the best indicator of a player playing better or worse from one game to the next. Pure points per game without the use of efficiency stats like field goal percentage (or more advanced effective field goal percentage) can give a skewed perception on just how well a player played in a game. If I were to continue work on this study in the future to greater flesh it out, I would probably add more factors in rather than just points per game, or find a more encompassing advanced stat that can combine more aspects of a player’s game, and use that advanced stat to compare between playoff performance and regular season performance.