If you Google, “Do great players make great coaches?”, you will get a lot of opinions but few facts. I’m sure there is an academic article out there somewhere with a definitive answer, but I could not find it. Refusing to give in, I contacted a professor at the University of Chicago Graduate School of Business who specializes in sports statistics; unfortunately, even he was not aware of a study that addresses this question. Though I usually write about public speaking and have little interest in sports other than watching my son play soccer, I decided to turn to ‘armchair’ sports statistics to resolve whether great players make great coaches once and for all.
My fixation with this question started with an email that I recently received:
“I’m halfway through your How To Deliver a TED Talk book and I think it will greatly help my public speaking. However, I’m wondering, have you ever given a TED Talk? I’ve studied hundreds of runners and I know what it takes to run a marathon in under 3 hours. But, even though I have run 13 marathons, I have not run one in less than 3 hours. So, I don’t feel authentic giving advice to other runners.”
The individual who emailed me was asking two questions.
The first question was, ‘Have you ever delivered a TED Talk?’ My answer is a little more complex than yes or no. At the time I wrote the book, I had not given a TED talk despite having invested over ten years and 10,000 hours in public speaking and coaching speakers. Since writing the book, I have delivered a TEDx Talk, organized 3 TEDx conferences, and coached many TED and TEDx speakers.
The second question, one with a more ominous subtext, was, “And, if you have never delivered a TED Talk, then can I trust the advice in your book?” Of course, this is just another form of ‘Do you have to have been a great player to be a great coach?” To answer this question, I turned to a free database of coach and player information provided by Basketball-Reference.com.
Round 1: Do coaches who were players win more?
There are several ways to think about what it means to be a great player. The first criteria we will look at whether coaches who were NBA players won more games than coaches who did not play in the NBA. After all, anybody who makes it into the NBA has to have been an exceptional college (or high school) player. As shown in Table 1, owners prefer to hire coaches who were NBA players (57%) versus coaches who did not play (43%)
Similarly, there are at least two ways to think about what it means to be a great coach. One way is based on career regular season winning percentage. Another way is based on a coach’s ability to win championships.
When being a great coach is defined as having a high winning percentage, it does not matter if the coach played in the NBA or not. When being a great coach is defined as winning championships, former players have a slight edge. Given these mixed results, it is too soon to conclude that great players make great coaches.
[table caption=”Table 1″ colalign=”left|center|center|center”]
Coach History,# Coaches,Win %,Championship %
Never played in NBA,100,44.5%,11.0%
Played in NBA,132,44.3%,14.4%
[/table]
Round 2: Do coaches who won championships as players win more?
In Round 1, I defined great players as those good enough to make it to the NBA. But, that may be too loose a definition of what it means to be a great player. Suppose that to be considered a great player, a coach would have needed to have won a championship during his playing days. As Table 2 shows, exactly half of coaches who were former players won at least one championship as a player.
Coaches who won championships as players fared slightly better in average career winning percentage than coaches who did not win titles as players and than coaches who did not play. However, coaches were more likely to win championships (19.7% chance) if they won when they were players than if they did not (9.1% chance). These results support the contention that great players are more likely to make great coaches, but we will explore a better test in Round 3.
[table caption=”Table 2″ colalign=”left|center|center|center”]
Coach History,# Coaches,Win %,Championship %
Never played in NBA,100,44.5%,11.0%
Never won title as player,66,42.9%,9.1%
Won title as player,66,45.6%,19.7%
[/table]
Round 3: Do coaches who were great individual players win more?
Round 2 imposed a more restrictive definition of what it means to be great player than Round 1 did. However, a coach may have simply been lucky enough during his playing days to warm the bench on a team that took home the title. That does not mean he was an exceptional player. Fortunately, John Hollinger came up with the player efficiency rating (PER), a single composite score that factors in many basketball performance measures; it is used to make fair comparisons between all kinds of different players. Table 3 shows coaching achievement broken out by how well coaches performed when they were players.
The better a coach played when he was in the NBA, the higher his career winning percentage. But, only coaches that were incredible players fared better than coaches who never played. When great coaching is defined as winning championships, coaches who were the weakest players did just as well as coaches who were exceptional players.
[table caption=”Table 3″ colalign=”left|center|center|center”]
Coach History,# Coaches,Win %,Championship %
Never played in NBA,100,44.5%,11.0%
PER: unknown,11,40.8%,9.1%
PER: 7.0 to 11.6,30,41.7%,26.7%
PER: 11.7 to 13.7,28,41.6%,3.6%
PER: 13.8 to 16.3,33,44.5%,3.0%
PER: 16.5 to 27.0,30,50.4%,26.7%
[/table]
Bonus Round: Do point guards really make better coaches?
When you Google this topic, you find a lot of people expressing the conventional wisdom that point guards go on to make better coaches than coaches who played other positions in the NBA. The reasoning is that point guards are usually the on-court leaders so they should make natural off-court leaders. As Table 4 shows, the people who hire coaches indeed have a preference for point guards. However, they might do better by hiring former power forwards.
The bottom line is that the position that a coach formerly played has little impact on his ability to be a great coach.
[table caption=”Table 4″ colalign=”left|center|center|center”]
Coach History,# Coaches,Win %,Championship %
Never played in NBA,100,44.5%,11.0%
unknown,9,41.6%,11.1%
Point Guard,50,45.0%,18.0%
Shooting Guard,28,44.3%,10.7%
Small Forward,20,42.0%,5.0%
Power Forward,15,45.4%,26.7%
Center,10,45.9%,10.0%
[/table]
The Final Word
Finally, Table 5 shows what happens to great players when they become coaches. Nearly half (46.7%) become great coaches. The other half do not achieve greatness. Though not shown, 30% of the worst players go on to be great coaches and 22% of those who never played went on to be great coaches.
[table caption=”Table 5″ colalign=”left|center|center|center”]
Coach History,# Coaches,Pct. of Total
Terrible Coach,2,6.7%
Below Avg. Coach,5,16.7%
Above Avg. Coach,9,30.0%
Great Coach,14,46.7%
[/table]
In the end, there are great coaches who were great players (ex: Lenny Wilkins, Doc Rivers, Rudy Tomjanovich). There are great coaches who were terrible players (ex. Pat Riley, Rick Adelman, Larry Brown). There are terrible coaches who were great players (ex: Kiki Vandeweghe, Red Kerr). And, there are terrible coaches who were terrible players (ex: Kenny Natt, Jack McCloskey). I’m of course using ‘terrible’ only in a relative sense since we are talking about some of the most talented athletes and leaders on the planet.
As there are so few championships, I think the better measure of whether someone is a great coach is his regular season win percentage. Defined that way, only the most exceptional players make better coaches than people who never even played.
From all this data, I conclude that you don’t have to have been a great player to be a great coach. Sure, your odds are a little better, but the difference may easily be explained away by other factors such as having a bigger budget to acquire better players to begin with. The proper way to do this is to use a multivariate statistical model but I will leave that to academics and team statisticians.
If you are looking for a coach in any field, rather than asking if he or she was a great player, you should ask:
(1) Does the coach’s advice resonate as intrinsically valuable to you?
(2) Does the coach have a track record of successfully developing great players?
Methodology Notes:
(1) Win percentages are for regular season games unless otherwise indicated
(2) Analysis includes coaches who coached at least 50 NBA games during their careers
(3) Championship percentages indicate the percentage of coaches in each category that coached a team to at least one NBA title
(4) All references to “NBA” are inclusive of the ABA as well unless otherwise indicated