The Evolution of the Usage of the Three Point Shot, Part IICharles Barkley recently made some very derogatory comments about the Golden State Warriors and he reiterated his contention that jump-shooting teams are not built to win championships. As the father of a wonderful young daughter, I echo Rachel Nichols' eloquent statement that "girlie" or "girl" should never be used as a synonym for, as she put it, something that is "weak" or "lame."
Moving past Barkley's ill-advised method of delivering his message, is there any truth to Barkley's assertion about jump-shooting teams purely from the standpoint of strategically analyzing basketball? I have discussed this subject a number of times in a variety of contexts. For instance, I wrote The Evolution of the Usage of the Three Point Shot for NBCSports.com nearly 10 years ago and then a few months after publication (during which time NBCSports.com was torn down and then rebuilt without maintaining links to prior content) I updated the article and posted it at 20 Second Timeout.
To best understand my take on the strengths and limitations of the three point shot as a basketball weapon, it would be helpful to provide some historical background, both in terms of the sport of basketball in general and also in terms of how my involvement in the sport as a recreational player informed my perspective about the three point shot.
Let's start with "advanced basketball statistics." Newer visitors to 20 Second Timeout--and perhaps even some of my veteran readers--may wonder why I refer to high profile advocates of "advanced basketball statistics" as "stat gurus." I borrowed that phrase from a line that Mike Lupica delivered a few times many years ago on the "Sports Reporters"; he would refer to someone who is supposed to be an expert at something as a "guru" and then he would say, "It is time for the guru to start 'guruing.'" I no longer recall who specifically Lupica was talking about but that wry sense of "OK, smart guy, it's about time that you actually prove that you know what you are doing" perfectly captures how I feel about many of the pompous analytics acolytes who act like they know more about basketball than people who have been playing and/or coaching the sport at the highest level for decades. Is there some value in trying to analyze the sport objectively? Of course there is but there should be a sense of humility and an understanding that the numbers don't capture everything about a sport that is played (and coached and officiated) by human beings, not robots.
Dean Oliver and Dan Rosenbaum are two of a handful of people whose basketball statistical work I respect--but far too many "stat gurus" are less concerned with objective truth than they are with making a name for themselves by proving that their pet theories are some kind of basketball gospel. Those self-serving, self-interested and self-satisfied "stat gurus" have had a pernicious effect on the way that the game is played, analyzed and discussed. Those "stat gurus" act like no one effectively utilized basketball statistics until they came along. I've got news for them: Dean Smith was using plus/minus back in the 1970s, if not earlier; Hubie Brown was hired by the Kentucky Colonels--who he led to the 1975 ABA title--because he impressed team management with his intricate understanding of basketball statistics and how those statistics can be used as tools to build a winning team.
"Advanced basketball statistics" are only as meaningful and relevant as (1) the box score numbers upon which they are based and (2) the accuracy of the formulas that are used to slice and dice those box score numbers. I have already demonstrated that assist statistics cannot be fully trusted, which means that any "advanced" formula that places a high value on assists will artificially inflate the value of players who accumulate many assists. There is good reason to question the accuracy/reliability of other box score numbers as well: subjectivity impacts the scoring of rebounds, steals, blocked shots and turnovers.
Hall of Famer/Top 50 player Rick Barry once told me that the only statistic he completely trusted was free throw percentage. Perhaps that seems self-serving of Barry because he owns one of the best career free throw percentages in pro basketball history but his point is that a player can inflate his field goal percentage by only shooting easy shots, rebounds can be padded by tips/taps that are not consistently recorded the same way and so forth. Then there is the reality--acknowledged by Rosenbaum but glossed over by many "stat gurus"--that "advanced basketball statistics" do not accurately measure individual defense. It should be obvious that subjective--but critically important--factors such as leadership, attitude and willingness/ability to play through injuries cannot be measured analytically at all. This is why I roll my eyes when a "stat guru" confidently declares that he can accurately rank every player in the NBA to the second decimal point of his proprietary "advance basketball statistic."
Many "stat gurus" assert that post play and the midrange game are inefficient, while the most efficient basketball shots are three pointers and free throws. The three point shot has evolved from a late game situational shot to a supplementary weapon to the mainstay of many teams' offenses. It should be remembered that even before "advanced basketball statistics" gained widespread influence, teams like the 1990s Houston Rockets and 1990s Orlando Magic utilized a "1 in, 4 out" offense featuring a dominant center surrounded by three point marksmen who punished opponents for double teaming the post. At that time, the three point shot was used to open up the game for the dominant post player but that dominant post player was still the focal point.
Teams that are run by--or at least influenced by--"stat gurus" deemphasize the value of the traditional big man in favor of deploying small lineups, shooting a lot of three pointers and (ideally) drawing a lot of fouls (typically via dribble penetration).
As for my personal involvement in the sport as a player, I have played rec league basketball since I was a kid in the 1970s. The strong suit of my game was always my jump shot (which really is not a traditional jump shot but some hybrid of Michael Cooper's set shot and Larry Bird's two hands behind his head delivery) but I spent my formative years playing without the three point shot. In 1987-88, the 19-9 three point shot was introduced into high school and college basketball. That wonderful arc soon appeared on the courts where I played both pickup ball and rec league ball. In the pickup games, we played to 12 by ones and twos. Under that scoring system, a good three point shooter is more valuable than a good post player; if I made even just 3 of 10 three pointers I scored six points, which is half the amount needed to win the game and a total that a post player could only match by shooting 60% from the field.
I really enjoyed playing 1 on 1 to 12 under that scoring system; I could hit six threes and take out a bigger, stronger opponent who was trying to wear me down one point at a time.
Of course, in league play when threes were not worth twice as much as twos, the math was not quite as favorable for a three point shooter but I did often mention to skeptical teammates that if I could make just one third of my three pointers I was as efficient as a post player who was making half of his shots in the paint. I was often the youngest participant in these games and the older guys who disdained long jumpers (and did not bother to do the simple math) got on my case for shooting so many three pointers. I told them rather bluntly that unless they were shooting 50% in the paint, then my three pointer was a higher percentage shot if I was connecting 33% of the time (and I probably was shooting closer to 40%), without even taking into account that the offensive team was more likely to rebound a long jumper than a missed shot in the paint. In addition, team defense is poorly organized in most rec leagues, so a smart and patient team can easily obtain wide open three point shots, particularly if the league does not use a shot clock.
So, from a young age I understood that in short games scored by ones and twos and in rec league games where post play was erratic and team defense was disorganized the three point shot was a very potent weapon.
Similarly, in the 40 minute college game featuring a short season followed by a one and done playoff system, it is possible to have a lot of success by constructing a good three point shooting team with well conditioned athletes who could beat the defense up and down the floor. Such teams generally play at a fast pace and thrive on shooting three pointers in transition. Rick Pitino has had success with this approach at multiple NCAA Division I schools and Paul Westhead enjoyed success both in NCAA Division I basketball and in the WNBA, where he led the Phoenix Mercury to the 2007 title. Westhead is the only coach to win a championship in both the WNBA and the NBA (1980 L.A. Lakers), though ironically Magic Johnson ran Westhead out of L.A. because Johnson felt that Westhead's offense at that time was too slow and methodical. However, NBA basketball consists of an 82 game regular season followed by four rounds of best of seven playoff series. Each regulation game lasts 48 minutes, not 40. Strategies that work in the college game are not necessarily applicable to the NBA game.
What does all of this have to do with Charles Barkley's comment about jump shooting teams? The NBA does not keep score by ones and twos and NBA defense is infinitely better than rec league defense. The notion that you can win big in the NBA just by shooting a large number of three pointers may make mathematical sense but it does not make real world sense. In that regard, there is some truth to what Barkley said. Take, for example, the Houston Rockets, who are now coached by Mike D'Antoni, whose Phoenix teams led the NBA in three point field goals made for three consecutive seasons (2005-07). The Rockets have a potent offense and are on pace to shatter many three point shooting records but they rank 22nd in defensive field goal percentage after finishing 19th in that category (under Kevin McHale and J.B. Bickerstaff) last season. That is a recipe for elimination early in the playoffs.
The nuance that Barkley is missing when he singles out the Warriors is that the Warriors differentiated themselves from other three-point happy teams (such as D'Antoni's Suns and Rockets) by playing organized, intense and effective defense. In their 2015 championship season, the Warriors led the league in defensive field goal percentage; they ranked third in that category last season and they currently rank second this season despite the loss of Andrew Bogut's rim protection.
Championship teams typically rank in the top five or 10 in defensive field goal percentage. The 2000-02 Shaquille O'Neal-Kobe Bryant Lakers are an interesting example. The Lakers led the league in defensive field goal percentage in 2000 and in 2002. In 2001 they slipped to 11th but still won the title. What happened? Unlike Bryant, O'Neal did not always fully exert himself during the regular season--but O'Neal was capable of playing championship level defense when he felt like it and after coasting through the 2001 regular season he ramped up his effort during the 2001 playoffs, when the Lakers ranked second in defensive field goal percentage. O'Neal decided to "heal on company time" during the 2002-03 season as the Lakers plummeted to 21st in defensive field goal percentage and this time O'Neal was not willing or able to turn it up in the playoffs: the Lakers ranked 15th out of 16 playoff teams in defensive field goal percentage and they lost in the second round to the eventual champion San Antonio Spurs, who ranked second in defensive field goal percentage during the 2003 playoffs and during the 2003 regular season.
In other words, instead of continually bashing the Warriors, Barkley should focus his attention on the Rockets: that is the jump shooting team that will never win a championship (as opposed to the Warriors, a jump shooting team that plays defense and has already won a championship).
There is an important point implied but not explicitly mentioned in Barkley's statement: the three point shot is a high variance shot, as I discussed in The Score, the Key Stat, the Bottom Line: A New 20 Second Timeout Feature, noting that three point field goal percentage can be a misleading statistic: if a player shoots 6-9 from three point range in one playoff game and 1-9 from three point range in the next playoff game that works out to a very good .389 3FG% but his team will likely lose the second game (and may not win the first game, either, if the team is deficient defensively and/or on the boards). In that article, I argued that Gilbert Arenas' three point field goal percentage did not justify his style of play and that a team was not likely to have much playoff success when led by a point guard who is indifferent defensively and who has poor shot selection--even if his overall shooting percentage is not bad, the high variance inherent in shooting a lot of shots from three point range (many of which were bad shots, even if they sometimes connected) is a recipe for disaster for that point guard and his team.
James Harden is similar to Arenas in many ways, though Harden is a better and more consistent player than Arenas was. My prediction for a Harden-led team is the same as my accurate prediction regarding an Arenas-led team: I would be surprised if such a team advances past the second round of the playoffs.
The Rockets' organization has been led for nearly a decade by Daryl Morey, who has been widely praised in the mainstream media as a visionary despite the fact that during his first eight years in Houston the Rockets missed the playoffs three times and only won three playoff series. Last season, the Rockets barely qualified for the playoffs but instead of revising their offense happy/defense optional approach Morey doubled down by hiring D'Antoni. The Rockets might win 50 games this season, as some of D'Antoni's Phoenix teams did. Harden might win the MVP over more deserving players, as D'Antoni's Phoenix point guard Steve Nash did twice. However, it will be surprising if the Rockets advance very far in the playoffs, barring a defensive turnaround similar to the 2001 Lakers (which is highly improbable, to say the least).
Successful NBA teams have judiciously incorporated "advanced basketball statistics" into their programs but teams (like Houston) that have gone all in have not reached the highest level. This is a valid point that Barkley could (and should) make during TNT telecasts.
The Philadelphia 76ers are the poster children for a "stat guru" run amuck or, as I put it a couple years ago, The 76ers Are the Waterloo for "Stat Gurus." Sam Hinkie, a protege of Morey's, instituted an infamous "process" in Philadelphia that came close to ruining a once storied franchise. Hinkie's media buddies still try to paint him as some kind of visionary but he actually is the 21st century Ted Stepien with a much better sense of public relations. In the early 1980s, Ted Stepien tore down the Cleveland Cavaliers to the point that the league had to essentially wrest control of the franchise from him, which is what happened (in a more subtle way) to Hinkie as well. It took Wayne Embry years to undo the damage that Stepien did and by the time Embry turned the Cavaliers into contenders there were no traces left of Stepien's handiwork; it will similarly take years for Bryan Colangelo to salvage the 76ers and by the time they become good again Hinkie's "process" will be a distant, dark memory.
Since the "process" began, the 76ers have never had a winning percentage above .250 and, despite some recent misguided media hype about Hinkie being vindicated, they are currently 5-18 (.217 winning percentage).
One of the many aspects of pro sports that Hinkie fails to understand is that a loser's mentality is contagious and must be purged from an organization if that organization is going to enjoy any success. You cannot turn your franchise into a perennial loser, stockpile some talent and then suddenly become a championship team; the Cleveland Browns' front office has been infiltrated with Hinkie-like thinking and they have turned the Browns into a de facto expansion team on par with the ultimate NFL laughingstock, the 1976 Tampa Bay Buccaneers.
When Mike Ditka was hired as the head coach of the Chicago Bears in 1982, he told his players that he had good news and bad news: the good news was that the Bears would win the Super Bowl within three years but the bad news was that many of them would no longer be on the team by that time. Both predictions came true. No "advanced basketball statistic" blueprint takes the place of a basic understanding of how to build a team that has a championship mindset and championship habits, two traits that are missing in both Houston and Philadelphia.
So, there is some truth to what Barkley said but--because he failed to articulate his message clearly and because he used some unfortunate language that overshadowed his larger point--it is easy to dismiss him as an old school curmudgeon. It is true that teams that rely on jump shooting and do not play excellent defense will not win a championship--but it is not correct to lump the Warriors into a category that should be headlined by the Rockets.
Evolution of the Usage of the Three Point Shot
Most Three Pointers Made
Year/League...Team...3 FGM...Player..(team)...3 FGM
1967-68/ABA..Pittsburgh..243..Les Selvage (Anaheim)..147
1968-69/ABA..Kentucky..335..Louie Dampier (Kentucky)..199
1969-70/ABA..Kentucky..330..Louie Dampier (Kentucky)..198
1970-71/ABA..Indiana..306..George Lehmann (Carolina)..154
1971-72/ABA..Indiana..220..Glen Combs (Utah)..103
1972-73/ABA..Indiana..172..Bill Keller (Indiana)..71
1973-74/ABA..San Diego..216..Bo Lamar (San Diego)..69
1974-75/ABA..Indiana..224..Bill Keller (Indiana)..80
1975-76/ABA..Indiana..250..Bill Keller (Indiana)..123
1979-80/NBA..San Diego..177..Brian Taylor (San Diego)..90
1980-81/NBA..San Diego..132..Mike Bratz (Cleveland)..57
1981-82/NBA..Indiana..103..Don Buse (Indiana)..73
1982-83/NBA..San Antonio..94..Mike Dunleavy (San Antonio)..67
1983-84/NBA..Utah..101..Darrell Griffith (Utah)..91
1984-85/NBA..Dallas..152..Darrell Griffith (Utah)..92
1985-86/NBA..Dallas..141..Larry Bird (Boston)..82
1986-87/NBA..Dallas..231..Larry Bird (Boston)..90
1987-88/NBA..Boston..271..Danny Ainge (Boston)..148
1988-89/NBA..New York..386..Michael Adams (Denver)..166
1989-90/NBA..Cleveland...346..Michael Adams (Denver)..158
1990-91/NBA..Portland..341..Vernon Maxwell (Houston)..172
1991-92/NBA..Milwaukee..371..Vernon Maxwell (Houston)..162
1992-93/NBA..Phoenix..398..Dan Majerle (Phoenix)/Reggie Miller (Indiana)..167
1993-94/NBA..Houston..429..Dan Majerle (Phoenix)..192
1994-95/NBA*..Houston..646..John Starks (New York)..217
1995-96/NBA*..Dallas..735..Dennis Scott (Orlando)..267
1996-97/NBA*..Miami..678..Reggie Miller (Indiana)..229
1997-98/NBA..Seattle..621..Wesley Person (Cleveland)..192
1998-99/NBA^..Houston..336..Dee Brown (Toronto)..135
1999-00/NBA..Indiana..583..Gary Payton (Seattle)..177
2000-01/NBA..Boston..592..Antoine Walker (Boston)..221
2001-02/NBA..Boston..699..Ray Allen (Milwaukee)..229
2002-03/NBA..Boston..719..Ray Allen (Milwaukee-Seattle)..201
2003-04/NBA..Seattle..723..Peja Stojakovic (Sacramento)..240
2004-05/NBA..Phoenix..796..Kyle Korver (Philadelphia)/Jason Richardson (Phoenix)..226
2005-06/NBA..Phoenix..837..Ray Allen (Seattle)..269
2006-07/NBA..Phoenix..785...Arenas (Washington)/Bell (Phoenix)..205
2007-08/NBA..Orlando..801...Jason Richardson (Charlotte)...243
2008-09/NBA...New York..823...Rashard Lewis (Orlando)...220
2009-10/NBA...Orlando..841...Aaron Brooks (Houston)...209
2010-11/NBA...Orlando...770...Dorell Wright (Golden State)...194
2011-12/NBA...Orlando...670...Ryan Anderson (Orlando)...166
2012-13/NBA...New York...891...Stephen Curry (Golden State)...272
2013-14/NBA...Houston...779...Stephen Curry (Golden State)...261
2014-15/NBA...Houston...933...Stephen Curry (Golden State)...286
2015-16/NBA...Golden State...1077...Stephen Curry (Golden State)...402
* The NBA shortened the three point arc to a uniform 22 feet (prior to and subsequent to these three seasons the three point arc was 22 feet in the corners and 23 feet nine inches elsewhere).
^ Season shortened to 50 games by a lockout.
Bold indicates an ABA/NBA record.
posted by David Friedman @ 3:21 PM