Comparing the '96 Bulls and '17 Warriors (6/6/17)

By Frank Brank @realfrankbrank

Over the past week, there have been several comparisons made between this season's Golden State Warriors squad and the '96 Chicago Bulls. Comparing teams from different eras for random, highly speculative games is very difficult to do, not only in the way of remembering the talent levels from two decades ago but making comparisons to today's style of play. However, a number of advanced metrics and finding comparisons to today's current players will allow us to come close.

Earlier this week, ESPN Chalk published an article in which they spoke with several Las Vegas bookmakers and asked them to set the line in
a game and series between the two powerhouses.

Per ESPN: “Five of the six (bookmakers) had the Warriors favored, from as high as -8 at the MGM (per Jay Rood, vice president of race and sports at MGM Resorts) to as low as -2 at William Hill US (per Nick Bogdanovich, director of trading, William Hill US).”

Our immediate reaction was that there is no chance in hell the Warriors would be eight-point favorites on a neutral court against the '96 Bulls, as this would imply that this year's Cavs team would also be favored over the '96 Bulls. It's not uncommon to see bookmakers make up numbers out of thin air in these hypothetical matchups, as there's really no point in putting in work to determine a true spread in a game that cannot actually garner any wagers.

Before we proceed further, it is important to note that there is no absolute way of breaking down this matchup. We discussed a handful of different ways that we could compare these two teams before settling on the method that we used, which is outlined in detail below. There are certainly some flaws, but overall, we feel like this is a pretty accurate representation of both teams.

First, we assembled the Bulls and Warriors roster, and assigned expected minutes played to both teams. Since we are assuming this is an NBA Finals matchup, both teams' starters get the bulk of the minutes, and neither team works deep into their bench. We can then gather each Bulls player's offensive and defensive box plus-minus (BPM), from Basketball Reference, and make a comparison to a player from this era. By making this comparison, we make up a team of current NBA players and use ESPN's real plus-minus (RPM) (the single best advanced predictor of point spreads in the NBA), and assign an expected value of team RPM based on minutes played.

For example, if Michael Jordan is expected to play 40 minutes of the 240 minutes available (48 total game minutes for five players), we would multiply his comp's RPM by 16.7% (40 minutes played divided 240 total minutes) to find his contribution to the Bulls' RPM. Once we assemble each team's RPM by summing the contribution of each player, we can use our regression analysis to find the relationship of team RPM to point spread. We've already completed the regression analysis and will not lay out all of the details in this article.

The method is imperfect and should be recognized to have some variance or error when converting one metric to another as they each have different inputs; however, this method should get us close to see how we can rank this Warriors team amongst one of the greatest teams of all-time.

The '96 Bulls would be represented with a top six of LeBron James, DeMarcus Cousins, Karl-Anthony Towns, Nerlens Noel, Bradley Beal, and Paul Millsap. This would be considered a super team, by any measure. In fact, while doing the comparisons, we were amazed at the results considering how truly great that artificial team would be perceived in today's NBA. Though the bench may not seem as a strong point for the '96 Bulls, their minutes share would remain pretty insignificant in a Finals matchup.

It's important to note that positions don't matter for this exercise. Obviously, Karl-Anthony Towns and Toni Kukoc are very different types of players, but what we are looking for here is a player, regardless of position, who best represents Kukoc's offensive and defensive abilities on the floor.

Along with the recognition of the Bulls' comp roster, we must recognize this current Warriors team. Their top six minutes' getters rank within the top 46 players in the NBA, per RPM. Four of those players are in the top 23, three players (Curry, Durant, and Green) are within the top 11 in all of the NBA.

The last thing we must do to finalize our artificial game is adjust for the Bulls' pace. In that era, the average pace for the NBA was around 90 possessions. That number has risen quite a bit to more than 96 possessions this season. Hence, the best comparison we can make is the '14-'15 Utah Jazz, the slowest playing team of that year, for the Bulls' pace. No team after the '14-'15 season has played close to as slow as the '96 Bulls, which played at an average offensive pace at that time.

After gathering all player minutes, assigning expected values to each player for their contribution to the game, and summing those expected values/contributions, we find the '96 Bulls team RPM to be 2.9 with 59% of their RPM being contributed from offense. Per those values, we anticipate the Bulls comp team to beat a current average NBA team by 11.4 points. The '17 Warriors, as currently constructed, have an expected RPM of 3.4, 62% of which comes from offense, and would beat an average NBA team by 13.0 points.

After looking at the comparison team of the '96 Bulls, we were somewhat shocked to see that the current Warriors team still ranks higher. After adjusting for the slower pace of the Bulls' era, we find that the Warriors would beat the Bulls by an average of 0.8 points with a final score of 108.1 to 107.3. Again, these results should not be seen as an absolute, there is certainly some error within the methodology; however, the error may just as likely favor the Warriors as it does the Bulls.

For reference, the Bulls won by an average margin of 12.3 points in the '95-'96 season, which is 0.9 points better than our comparison team against the average NBA team. If we consider that margin to be a guideline for the error within the methodology, we would anticipate the two teams to be essentially even. Even with the Warriors being favored by 0.8 points, the expectation would be to win 51% of the time, essentially a coin flip. Nonetheless, if the gauge or measuring stick for all-time great teams is the '96 Bulls, we must consider the '17 Warriors, in the least, to be bumping right up next to them.

Here is a look at a full series between the '96 Bulls and '17 Warriors, simulated on a neutral court 50,000 times: