Roughly 10 years ago, Phil Jackson supplied a pearl of wisdom which Marc Stein added into his weekly power rankings column of “40 before 20”. The implication being that the Lakers were finally on track that season to be a contending team by winning 40 games before losing 20.
Now obviously, “40 before 20” is hardly anything groundbreaking. Teams that win a lot more games than they lose are generally considered to be the elite teams in any given season. Rephrased another way, “40 before 20” means a team with a 75% win rate or higher through the first part of the regular season. However, can such a simple rule of thumb actually be used to identify the teams most likely to win an NBA title?
Nearly every player, coach, and executive has a soundbite buried somewhere on the internet about how the goal is to win a championship. However, no matter what happens, only a single team out of 30 in the association can achieve this goal each season. Does that mean that every season 29 teams will inevitably fail? Maybe so in the arena of hyper-competitiveness and sports punditry, but teams seeking sustained success opt for a more nuanced approach. Listen to the infamous Sam Hinkie (former GM of the Philadelphia 76ers) during the 2018 MIT Sloan Sports Analytics Conference:
“To Griff’s point you don’t you’re not going to win the title every year. No one’s going to win them all so that’s challenging and sometimes too high of a bar, there’s a lot of luck involved even in that…so often people would lower the bar a little bit. The thing we would talk about a lot is trying to put yourself in contention where you had a reasonable shot. You can define that in lots of different ways…Often a decent place was that you were kind of a Final Four team.”
Championship or bust is simply not practical way to evaluate a team from the front office perspective. Even the best teams need luck to break the right way to win a title. An injury to a key player, unexpected off the court circumstances, or even a single missed basket can derail a team’s championship aspirations.
When taking the long view, the most sensible approach for executives is to evaluate the team at the end of the season to decide if the team had a realistic shot of winning the title. This is where Phil Jackson’s rule of thumb can come back into play. If we assume that the champion will be one of the teams that fulfills the criteria to be a contender (40 before 20) and that all contenders have a chance of winning the championship, then the goal of a team should be to be a contender every year.
In this post, I will test Phil Jackson’s “40 before 20” against the actual playoff results to find how often his rule of thumb can be used to predict the contending teams, and how often it misses a contender.
First of all, how often is the champion a team that won 40 games before losing 20 games? Here is a list of the champions starting from the 1988-89 season:
|Season||Team||Final Record||Interim Record||40 Before 20|
Phil Jackson’s rule of thumb positively flagged 26 out of 29 (~90%) of the eventual champions over the past 30 seasons (I am counting the 1998-99 Spurs since they were well on pace to qualify for 40 before 20 in the shortened season). This seems pretty good when taken at face value, but the goal of this rule of thumb is not to only predict the eventual champion and instead to predict the elite contending teams of that particular season. Thus, it is important to dive deeper into the data.
Historical Conference Champions
What about the teams that made it to the finals and lost? Surely any team that makes it to the finals should be classified as a contender even if that team ultimately loses. After all, these teams were one 7-game series away from the title. Even an overmatched team should have some chance of winning with a couple lucky breaks. How does Phil Jackson’s rule of thumb hold up in this case? Again here is the list of teams reaching the NBA finals over the past 30 seasons along with their win loss record:
|Season||Team||Final Record||Interim Record||40 Before 20|
When we include both teams that ultimately made the finals in this analysis, we find that Phil Jackson’s rule of thumb begins losing some of its predictive power. Now, only 47 out of the 58 (81%) finals teams managed to win 40 games before losing 20.
Four teams in the 2017-18 season have managed to achieve the 40 before 20:
|Season||Team||Final Record||Interim Record|
Popular consensus entering the playoffs has penciled the Warriors and the Rockets in for a date in the Western Conference Finals. As of this moment neither team has given any indication to the contrary with both teams pulling ahead of their respective first round matchups with 2-0 series leads. Meanwhile in the Eastern Conference, the Raptors have just finished their best regular season in franchise history and have seemingly exorcised some of their historical playoff woes also taking a commanding 2-0 lead in their series against the Wizards. The Celtics also have a 2-0 lead over the Bucks but with Kyrie Irving and Gordon Hayward out for the remainder of the playoffs (reportedly), are not expected to make the Eastern Conference Finals.
The Western Conference champion will almost surely be either the Warriors or the Rockets and in either case will add another data point to the list of conference champions achieving 40 before 20. The Eastern Conference on the other hand features a couple teams not flagged with Phil Jackson’s rule of thumb which may serve as strong obstacles in the Raptors’ path to the finals in the Cavaliers and the Sixers.
Conclusion and Next Steps
Based on this initial analysis it appears that Phil Jackson’s rule of thumb has already experienced issues in identifying contenders. Nearly 1 in 5 finals teams did not achieve 40 wins before 20 losses. In part 2 of this post I will look to find reasons why some of these finals teams did not appear to achieve notable regular season success. I will also explore teams that did achieve 40 before 20 and did not end up emerging from the playoffs as a conference champion.
If you are interested in how I collected the data for this analysis I have included a short explainer below. Otherwise, thanks for actually reading this far through my initial post and feel free to skip this section.
ccagarwal’s nbaTools package for R – This package was a great help for scraping data from NBA.com for both game logs of individual seasons as well as historical data all the way back to the 1946-47 season. I ended up including only the data from the 1988-89 season onwards for the sake of simplicity.
Custom R Scripts and Functions – I wrote a couple simple scripts and functions to help me cycle through game log data as well as generate the tables shown in the post.
Microsoft Excel – I used excel to perform some simple data cleaning as well as a way to store the scraped data. I actually initially ran into an issue where the scripts I was writing based on the nbaTools package would time out because I tried to pull the data for 30 teams across 30 seasons.