I appreciate the careful statistical comparison between Shai Gilgeous-Alexander and Michael Jordan, but it's not one that would have occurred to me from watching the two play. The analysis touches on this with discussion of efficiency and wasted possessions. Those things are more important to the feel of the game than perhaps to team wins.
I think of Michael Jordan in a category with baseball player Rickey Henderson. Both were all-time great players, of course, but both could have been greater if they had done less. Rickey Henderson's teams won more games when he got on base and didn't try to steal than when he got on base and tried to steal. It's not just the loss from when he was caught, it's that he attempted steals when the game situation--score, inning, outs, batter at the plate and pitch count--made it a negative expected value move. You can see that in the statistics, but you feel it more strongly as a fan--when Henderson got on base you felt he was thinking of the record book, not the game.
Similarly, Jordan missed too many shots. Threes, of course, and free throws, but also bad two opportunities. He played too many minutes a game, and too many games a season. He was the last Bull to touch the ball on too many wasted possessions. And he benefited more than any player I've ever seen from the NBA-referee star-favoritism. I'm not sure he would have had any steals or blocks if fouls had been called on him the way they were called on players guarding him.
Of course Jordan was one of the greatest NBA players ever, but I don't think he's head-and-shoulders above the rest of the top 20 or so, and he's a type of player I don't enjoy watching. I think his reputation is enhanced by his team successes--which were not all due to him--and his gaudy totals. Admittedly, I saw him in person mainly in New York, which seemed to bring out his hot-dogging more than other places.
Shai Gilgeous-Alexander has an almost opposite feel. When he's on the court you feel like he's playing for the team to win, not for the applause or the stats. He rarely takes a bad shot, or plays tired, or wastes a possession. He makes the players around him better, he doesn't steal their opportunities. He may not be the physical talent Michael Jordan was, and he probably won't have nearly as many peak years as Jordan, but he's the kind of player I pay to watch.
Interesting comment Aaron. And it's possible you're right about Jordan taking too many inefficient shots, although I wonder how much that's a function of Jordan playing in the pre-analytics era and the fact that the Bulls looked to Jordan to bail them out when possessions bogged down versus being Michael's "fault."
What I can tell you, and obviously this is just my vibes, is that as someone who generally rooted against Jordan, he was far and away the player I was least confident of beating. You just always felt that he would figure out a way to get the win. And no one was particularly close in that regard. It's for that reason that I do think there is significant daylight between him and the other super elite players.
With Jordan, I don't think the driver would have been analytics. If he played in an era where other players played like they do today, his pathological drive to be better than everybody would have driven him play differently.
There's also a question of whether he "played down" to the level of his competition. This is a guy who got bored with winning titles and switched sports. If Lebron came into the league and he and Jordan had a Magic-Bird type rivalry, I'd bet that MJ's 3-point %, efficiency, etc. would have been better.
This is all counterfactual and impossible to prove. But I think the question is: were Jordan's limitations fundamental to him - he was physically or mentally incapable of changing his game - or primarily a function of his environment.
Agreed it's all impossible to prove. I don't care about the moral issue--is a guy bad for being selfish?--but the fun of watching the game. If a guy always prefers a shot to a pass, and to pass up pretty good passes or shots in the hopes of better ones resulting in too many poor last-second attempts; he's not much fun to watch.
I’m not expert, but believe that this is a theory that’s never had any public evidence to support it. Jordan always cited both having lost the desire to play basketball and the death of his father.
You could well be right, I'm talking about feel rather than analytics. I felt that Jordan took too many bad shots with good passes available, and waited too late hoping for perfect shots rather than taking a pretty good shot or pass. Even before modern analytics, coaches understood these tendencies and tried to discipline their players, but not all players responded.
I agree that Jordan seemed supernaturally good at winning games. But I suspect some of that is circular. If you always take the last shot, and your team wins a lot of games, you'll have a lot of game-winning shots. Maybe there was some special Jordan-magic that elevated his teams, but I didn't feel it.
FWIW, as long as we're sharing "feel," it felt 100% the opposite to me. As someone usually rooting against the Bulls, it *did* feel like there was some special Jordan-magic. I think Gordon Strause's comment is spot-on: as a fan of the opposing team, there was no one who seemed more certain to find a way to beat you. In any sport. I've made the same point to my (college-age) kids, who desperately want to believe LeBron is the GOAT. As great as he is, facing him never scared anyone the way facing Jordan scared them. From what I've seen, players who overlapped with both of them seem to agree - I recently saw an interview with Ray Allen along these lines.
This is all a strange line of discussion for an analytics-oriented site, and there's good reason to be skeptical about intangible claims like this - I still have PTSD from the magical powers attributed to Derek Jeter and his .838 career postseason OPS. But it was hard to watch Jordan through the 90s and not come away thinking that his maniacal commitment to finding ways to win was a real thing over and above his generational abilities as a basketball player.
Finally, a couple of asides: I think your point about always taking the last shot cuts the other way. At any level of basketball, scoring at the buzzer when the other team knows you're the go-to guy is particularly difficult. Jordan didn't just hit a lot of those shots, he hit an unusually high percentage. E.g., https://undisputedgoat.medium.com/jordan-in-the-clutch-30f6e7ed4c43 And as for not passing, he *did* have two pretty prominent last-10-seconds assists in the Finals: Kerr in 1997 Game 6 and Luc Longley to send 1998 Game 1 to overtime.
I certainly see your point, and for what it's worth, I'm used to being in the minority about Jordan, especially among people who went to NBA games while he was playing. No one can deny his spectacular success and statistics, not to mention talent, and most people give him credit for playing 40 minutes per game for 82 games per year with 24 shots per game.
When SGA win 6 championships while putting up those numbers AND in an era with different defensive rules AND while facing defenses specifically designed against him (ain’t heard no “SGA Rules” yet), then AND ONLY THEN the comparison might make sense.
This coming from someone who was a Knick fan in the 90s, BTW.
No argument that Jordan had a far more successful career than Shai Gilgeous-Alexander had, or is likely to have. The point of the column is that the last few seasons for SGA is more similar to Jordan at his peak than you might think.
I don't disagree. Perhaps the way Michael Jordan played was the best way for the Bulls to win championships in the 1990s. It's just not the kind of basketball I care to watch. Yes, Jordan delivered marvelous exhibitions of talent, but it bothered me that I thought he could produce more if he tried to do less. I much prefer to watch SGA with his efficient teamwork.
"Similarly, Jordan missed too many shots. Threes, of course, and free throws, but also bad two opportunities."
I think one thing that's missing from this sort of discussion is era adjustment. A lot of the 90s was an ugly, offensive challenged era. How efficient was Jordan compared to the rest of his peers?
My opinion was based on feel. Yes, offenses were uglier in the 1990s, but I'm comparing Michael Jordan to other players I watched. He stood out for me as one of the most inefficient players in the game at the time--while Shai Gilgeous-Alexander strikes me as more efficient than average among players today.
Moreover most inefficient players are stars for bad teams where it probably helps fan interest more to have a league-leading player in some major category than to win 25 games instead of 20.
Seems like you're describing Kobe more than Michael, Aaron. What made Jordan so great was his combination of volume AND efficiency. In fact, one of the stats everyone is citing about SGA is the fact that he is the only guard other than Jordan ever to average more than 30 points and a shooting percentage of over 50% more than once.
Not going to dispute how good SGA has been this year, and maybe you're right that Jordan would have been even better if he had passed more. But the crazy thing about Jordan is he made those difficult shots at such a high percentage.
And, of course, as others have cited, in today's era of 3 point spacing, it's both easier to find good looks for both open teammates and yourself.
Well I agree that Kobe Bryant was more Michael-Jordan-like than Shai Gilgeous-Alexander-like, but I'd say Bryant was considerably more efficient that Jordan, although perhaps about the same relative to era-average.
No argument that Jordan was spectacular on both volume and percentage--but that's not what I mean by efficiency.
In a very simple model of basketball, every player on a team should have the same shooting percentage. If someone is shooting at a better clip, the team should get him the ball more, so he attracts more defensive attention and takes some less favorable shots, so his shooting percentage drops to the team average. If someone is shooting worse, the team should get him the ball less, so his defensive attention drops and he only gets the ball in the best situation, and his shooting percentage rises.
Of course, that's not exactly true, mainly because some players are harder to get the ball to in favorable situations than others. A point guard can toss up a long-range shot any time he wants without risking even one pass, but getting the ball to a center near the basket takes time and risk--you'll get no shot some of the time, or waste most of the clock trying and get only a desperation heave at the end. So a point guard should have a lower shooting percentage than a center.
Nevertheless, when watching the game you can see players taking too many shots for their shooting percentage, while higher percentage shooters are starved for the ball. This is inefficient. Similarly, you can see tired players on the floor when a lesser-but-fresher player sits on the bench. This is inefficient.
Aaron: Take a look at the stats. Michael was much, much more efficient than Kobe, without even taking into account their eras, which would tilt the balance even more in Michael's favor.
And your model of getting more shots to the player on the team with the higher shooting percentage until the defense pays them more attention makes sense. But Jordan almost always led the Bulls in shooting percentage. You're actually making an argument that Michael took too few shots not too many!
(Note: the one exception to this is Steve Kerr, who consistently did shoot a higher percentage than Michael did when they played together. So arguably the Bulls would have been better off getting Steve more shots. That said, with the exception on one year with the Cavs, Kerr shot a much, much higher percentage when he played with Michael than at any other time in his career. And I'm guessing Steve would be the first to tell you that was a function of playing with Michael in Phil Jackson's offense.)
I fear we're drifting from discussion to argument. I don't claim I'm 100% certain of what I feel and you certainly make good points. I don't want to just reflexively defend my positions. However I do think there is a genuine difference of opinion here around what I mean by efficiency.
It's not easy to measure what I'm calling efficiency from the statistics. Efficiency is seen at the margin--the worst shots a player takes and the worst minutes he plays versus the alternative at the time.
Jordan was such a great player that he piled up great statistics on most possessions and most minutes--but in my opinion from watching him, there were a lot of shots he took and minutes be played that would have been better given to another. These inflate Jordan's career totals, but reduce his value to the team.
Similarly while Jordan had a top shooting percentage among the Bulls he played with, I think that's due to his overall talent. His good shots were so good that he could take more than his share of bad shots and still have a top shooting percentage.
When making these sort of comparison, it would help top use some version of stats like baseball WRC+, that is that are better normed to circumstances and era. I've got a sneaking suspicion that 31 PPG and 6 RPG from a guard were more of an outlier in 1993 than they are in 2025.
Is there any version of on/off that also controls for the on/off of the other players on the court? (Obviously this is an annoying and recursive problem).
I am no Tatum stan, but I think his on/off is depressed this year because the Celtics tend to use him to bolster the second unit while the other starters rest, while coaches in the past were more likely to rest or play all their starters simultaneously leading to inflated on/offs for their starters.
Intuitively, I'd think that score-first guards/wings would get used in the "carry the second unit" role more often, while pass-first star players would maximize value alongside the the starters.
In defense on ant, as a wolves homer, his usage is clearly the lowest across the board making the raw totals suffer some. His on off numbers are also brutalized by playing with a bizarre wolves starting lineup and an elite bench unit. I’m not suggesting that he’s near these others class yet, but there may yet be hope with more growth. Awesome article, thanks for all you do! Silver/ Stephen A. Smith 2028.
Thanks for a great article. Certainly wouldn’t have arrived at this comparison on my own. Is it possible there are some numbers off in the SGA column in table 2 (2p% 3p% and total FG%)? Adding together his 2s and 3s, his overall FG percentage is the best by an even slightly greater margin.
I appreciate the careful statistical comparison between Shai Gilgeous-Alexander and Michael Jordan, but it's not one that would have occurred to me from watching the two play. The analysis touches on this with discussion of efficiency and wasted possessions. Those things are more important to the feel of the game than perhaps to team wins.
I think of Michael Jordan in a category with baseball player Rickey Henderson. Both were all-time great players, of course, but both could have been greater if they had done less. Rickey Henderson's teams won more games when he got on base and didn't try to steal than when he got on base and tried to steal. It's not just the loss from when he was caught, it's that he attempted steals when the game situation--score, inning, outs, batter at the plate and pitch count--made it a negative expected value move. You can see that in the statistics, but you feel it more strongly as a fan--when Henderson got on base you felt he was thinking of the record book, not the game.
Similarly, Jordan missed too many shots. Threes, of course, and free throws, but also bad two opportunities. He played too many minutes a game, and too many games a season. He was the last Bull to touch the ball on too many wasted possessions. And he benefited more than any player I've ever seen from the NBA-referee star-favoritism. I'm not sure he would have had any steals or blocks if fouls had been called on him the way they were called on players guarding him.
Of course Jordan was one of the greatest NBA players ever, but I don't think he's head-and-shoulders above the rest of the top 20 or so, and he's a type of player I don't enjoy watching. I think his reputation is enhanced by his team successes--which were not all due to him--and his gaudy totals. Admittedly, I saw him in person mainly in New York, which seemed to bring out his hot-dogging more than other places.
Shai Gilgeous-Alexander has an almost opposite feel. When he's on the court you feel like he's playing for the team to win, not for the applause or the stats. He rarely takes a bad shot, or plays tired, or wastes a possession. He makes the players around him better, he doesn't steal their opportunities. He may not be the physical talent Michael Jordan was, and he probably won't have nearly as many peak years as Jordan, but he's the kind of player I pay to watch.
Interesting comment Aaron. And it's possible you're right about Jordan taking too many inefficient shots, although I wonder how much that's a function of Jordan playing in the pre-analytics era and the fact that the Bulls looked to Jordan to bail them out when possessions bogged down versus being Michael's "fault."
What I can tell you, and obviously this is just my vibes, is that as someone who generally rooted against Jordan, he was far and away the player I was least confident of beating. You just always felt that he would figure out a way to get the win. And no one was particularly close in that regard. It's for that reason that I do think there is significant daylight between him and the other super elite players.
With Jordan, I don't think the driver would have been analytics. If he played in an era where other players played like they do today, his pathological drive to be better than everybody would have driven him play differently.
There's also a question of whether he "played down" to the level of his competition. This is a guy who got bored with winning titles and switched sports. If Lebron came into the league and he and Jordan had a Magic-Bird type rivalry, I'd bet that MJ's 3-point %, efficiency, etc. would have been better.
This is all counterfactual and impossible to prove. But I think the question is: were Jordan's limitations fundamental to him - he was physically or mentally incapable of changing his game - or primarily a function of his environment.
Agreed it's all impossible to prove. I don't care about the moral issue--is a guy bad for being selfish?--but the fun of watching the game. If a guy always prefers a shot to a pass, and to pass up pretty good passes or shots in the hopes of better ones resulting in too many poor last-second attempts; he's not much fun to watch.
I agree with that. Beautiful passes leading to great shots are the most satisfying thing to watch in basketball, only rivalled by spectacular blocks.
Didn’t he switch sports for gambling-related reasons?
I’m not expert, but believe that this is a theory that’s never had any public evidence to support it. Jordan always cited both having lost the desire to play basketball and the death of his father.
Yes, and I presume there was information that the general public was never privy to.
You could well be right, I'm talking about feel rather than analytics. I felt that Jordan took too many bad shots with good passes available, and waited too late hoping for perfect shots rather than taking a pretty good shot or pass. Even before modern analytics, coaches understood these tendencies and tried to discipline their players, but not all players responded.
I agree that Jordan seemed supernaturally good at winning games. But I suspect some of that is circular. If you always take the last shot, and your team wins a lot of games, you'll have a lot of game-winning shots. Maybe there was some special Jordan-magic that elevated his teams, but I didn't feel it.
FWIW, as long as we're sharing "feel," it felt 100% the opposite to me. As someone usually rooting against the Bulls, it *did* feel like there was some special Jordan-magic. I think Gordon Strause's comment is spot-on: as a fan of the opposing team, there was no one who seemed more certain to find a way to beat you. In any sport. I've made the same point to my (college-age) kids, who desperately want to believe LeBron is the GOAT. As great as he is, facing him never scared anyone the way facing Jordan scared them. From what I've seen, players who overlapped with both of them seem to agree - I recently saw an interview with Ray Allen along these lines.
This is all a strange line of discussion for an analytics-oriented site, and there's good reason to be skeptical about intangible claims like this - I still have PTSD from the magical powers attributed to Derek Jeter and his .838 career postseason OPS. But it was hard to watch Jordan through the 90s and not come away thinking that his maniacal commitment to finding ways to win was a real thing over and above his generational abilities as a basketball player.
Finally, a couple of asides: I think your point about always taking the last shot cuts the other way. At any level of basketball, scoring at the buzzer when the other team knows you're the go-to guy is particularly difficult. Jordan didn't just hit a lot of those shots, he hit an unusually high percentage. E.g., https://undisputedgoat.medium.com/jordan-in-the-clutch-30f6e7ed4c43 And as for not passing, he *did* have two pretty prominent last-10-seconds assists in the Finals: Kerr in 1997 Game 6 and Luc Longley to send 1998 Game 1 to overtime.
I certainly see your point, and for what it's worth, I'm used to being in the minority about Jordan, especially among people who went to NBA games while he was playing. No one can deny his spectacular success and statistics, not to mention talent, and most people give him credit for playing 40 minutes per game for 82 games per year with 24 shots per game.
When SGA win 6 championships while putting up those numbers AND in an era with different defensive rules AND while facing defenses specifically designed against him (ain’t heard no “SGA Rules” yet), then AND ONLY THEN the comparison might make sense.
This coming from someone who was a Knick fan in the 90s, BTW.
No argument that Jordan had a far more successful career than Shai Gilgeous-Alexander had, or is likely to have. The point of the column is that the last few seasons for SGA is more similar to Jordan at his peak than you might think.
But the idea is that a season that looks like Jordan was way more valuable in 1993 than it was in 2025.
Babe Ruth hitting 60 home runs in 1927 is way more impactful than Aaron Judge hitting 63 in 2023.
I don't disagree. Perhaps the way Michael Jordan played was the best way for the Bulls to win championships in the 1990s. It's just not the kind of basketball I care to watch. Yes, Jordan delivered marvelous exhibitions of talent, but it bothered me that I thought he could produce more if he tried to do less. I much prefer to watch SGA with his efficient teamwork.
"Similarly, Jordan missed too many shots. Threes, of course, and free throws, but also bad two opportunities."
I think one thing that's missing from this sort of discussion is era adjustment. A lot of the 90s was an ugly, offensive challenged era. How efficient was Jordan compared to the rest of his peers?
My opinion was based on feel. Yes, offenses were uglier in the 1990s, but I'm comparing Michael Jordan to other players I watched. He stood out for me as one of the most inefficient players in the game at the time--while Shai Gilgeous-Alexander strikes me as more efficient than average among players today.
Moreover most inefficient players are stars for bad teams where it probably helps fan interest more to have a league-leading player in some major category than to win 25 games instead of 20.
Seems like you're describing Kobe more than Michael, Aaron. What made Jordan so great was his combination of volume AND efficiency. In fact, one of the stats everyone is citing about SGA is the fact that he is the only guard other than Jordan ever to average more than 30 points and a shooting percentage of over 50% more than once.
Not going to dispute how good SGA has been this year, and maybe you're right that Jordan would have been even better if he had passed more. But the crazy thing about Jordan is he made those difficult shots at such a high percentage.
And, of course, as others have cited, in today's era of 3 point spacing, it's both easier to find good looks for both open teammates and yourself.
Well I agree that Kobe Bryant was more Michael-Jordan-like than Shai Gilgeous-Alexander-like, but I'd say Bryant was considerably more efficient that Jordan, although perhaps about the same relative to era-average.
No argument that Jordan was spectacular on both volume and percentage--but that's not what I mean by efficiency.
In a very simple model of basketball, every player on a team should have the same shooting percentage. If someone is shooting at a better clip, the team should get him the ball more, so he attracts more defensive attention and takes some less favorable shots, so his shooting percentage drops to the team average. If someone is shooting worse, the team should get him the ball less, so his defensive attention drops and he only gets the ball in the best situation, and his shooting percentage rises.
Of course, that's not exactly true, mainly because some players are harder to get the ball to in favorable situations than others. A point guard can toss up a long-range shot any time he wants without risking even one pass, but getting the ball to a center near the basket takes time and risk--you'll get no shot some of the time, or waste most of the clock trying and get only a desperation heave at the end. So a point guard should have a lower shooting percentage than a center.
Nevertheless, when watching the game you can see players taking too many shots for their shooting percentage, while higher percentage shooters are starved for the ball. This is inefficient. Similarly, you can see tired players on the floor when a lesser-but-fresher player sits on the bench. This is inefficient.
Aaron: Take a look at the stats. Michael was much, much more efficient than Kobe, without even taking into account their eras, which would tilt the balance even more in Michael's favor.
And your model of getting more shots to the player on the team with the higher shooting percentage until the defense pays them more attention makes sense. But Jordan almost always led the Bulls in shooting percentage. You're actually making an argument that Michael took too few shots not too many!
(Note: the one exception to this is Steve Kerr, who consistently did shoot a higher percentage than Michael did when they played together. So arguably the Bulls would have been better off getting Steve more shots. That said, with the exception on one year with the Cavs, Kerr shot a much, much higher percentage when he played with Michael than at any other time in his career. And I'm guessing Steve would be the first to tell you that was a function of playing with Michael in Phil Jackson's offense.)
I fear we're drifting from discussion to argument. I don't claim I'm 100% certain of what I feel and you certainly make good points. I don't want to just reflexively defend my positions. However I do think there is a genuine difference of opinion here around what I mean by efficiency.
It's not easy to measure what I'm calling efficiency from the statistics. Efficiency is seen at the margin--the worst shots a player takes and the worst minutes he plays versus the alternative at the time.
Jordan was such a great player that he piled up great statistics on most possessions and most minutes--but in my opinion from watching him, there were a lot of shots he took and minutes be played that would have been better given to another. These inflate Jordan's career totals, but reduce his value to the team.
Similarly while Jordan had a top shooting percentage among the Bulls he played with, I think that's due to his overall talent. His good shots were so good that he could take more than his share of bad shots and still have a top shooting percentage.
When making these sort of comparison, it would help top use some version of stats like baseball WRC+, that is that are better normed to circumstances and era. I've got a sneaking suspicion that 31 PPG and 6 RPG from a guard were more of an outlier in 1993 than they are in 2025.
Is there any version of on/off that also controls for the on/off of the other players on the court? (Obviously this is an annoying and recursive problem).
I am no Tatum stan, but I think his on/off is depressed this year because the Celtics tend to use him to bolster the second unit while the other starters rest, while coaches in the past were more likely to rest or play all their starters simultaneously leading to inflated on/offs for their starters.
Intuitively, I'd think that score-first guards/wings would get used in the "carry the second unit" role more often, while pass-first star players would maximize value alongside the the starters.
You're looking for RAPM
In defense on ant, as a wolves homer, his usage is clearly the lowest across the board making the raw totals suffer some. His on off numbers are also brutalized by playing with a bizarre wolves starting lineup and an elite bench unit. I’m not suggesting that he’s near these others class yet, but there may yet be hope with more growth. Awesome article, thanks for all you do! Silver/ Stephen A. Smith 2028.
I think OKC, without SGA, is still a top5 seed in the west. Without Jokic, Denver wins 25 games. Jokic should be this years MVP and it isn’t close.
Thanks for a great article. Certainly wouldn’t have arrived at this comparison on my own. Is it possible there are some numbers off in the SGA column in table 2 (2p% 3p% and total FG%)? Adding together his 2s and 3s, his overall FG percentage is the best by an even slightly greater margin.
Can we get this data for LeBron?