1990-1992 Michael Jordan Shot Chart

[ShaqAttack3234]

This is great stuff, although I'm questioning the selection of games for this shot chart due to his ridiculous 3pt percentage

Eh, not too surprising. The only one of the 3 seasons he attempted a significant amount was '90 when he attempted 3 per game and shot 37.6%, he then attempted just 1.1 in '91 while shooting 31.2% and just 1.3 attempts in '92 while shooting 27%. Because of this, his percentage over the three years was still 33.8% in those 3 regular seasons combined. The games Dipper 13 used included all of the playoff series from these 3 years as well, and in general, MJ attempted more 3s and shot a better percentage on them in the playoffs than regular season at 35.8% on an average of 2.2 3PA over the 3 playoff runs. So yeah, these were better 3 point shooting games than his average, but not by such a huge disparity, especially since those playoff games are 55 of the 126 games.

[lorak]

This is great stuff, although I'm questioning the selection of games for this shot chart due to his ridiculous 3pt percentage

Yeah, that's unfortunately the biggest problem here. And that's why it would be better to do only playoffs games - as all of them are available, so although sample would be smaller, but at the same time more reliable, without "highlights bias".

[lorak]

This is great stuff, although I'm questioning the selection of games for this shot chart due to his ridiculous 3pt percentage

Eh, not too surprising. The only one of the 3 seasons he attempted a significant amount was '90 when he attempted 3 per game and shot 37.6%, he then attempted just 1.1 in '91 while shooting 31.2% and just 1.3 attempts in '92 while shooting 27%. Because of this, his percentage over the three years was still 33.8% in those 3 regular seasons combined. The games Dipper 13 used included all of the playoff series from these 3 years as well, and in general, MJ attempted more 3s and shot a better percentage on them in the playoffs than regular season at 35.8% on an average of 2.2 3PA over the 3 playoff runs. So yeah, these were better 3 point shooting games than his average, but not by such a huge disparity, especially since those playoff games are 55 of the 126 games.

Even if we include playoffs games from that period and equal (to Dipper's ample) amount of regular season games, then Jordan's real 3P% during that period was 34.5%. That's 9.9% less than his l percentage from Dipper's sample: 38.2%.

[Gregoire]

This is great stuff, although I'm questioning the selection of games for this shot chart due to his ridiculous 3pt percentage

Yeah, that's unfortunately the biggest problem here. And that's why it would be better to do only playoffs games - as all of them are available, so although sample would be smaller, but at the same time more reliable, without "highlights bias".

Maybe Dipper could separate give us only PO numbers...

[Dipper 13]

Yeah, that's unfortunately the biggest problem here. And that's why it would be better to do only playoffs games - as all of them are available, so although sample would be smaller, but at the same time more reliable, without "highlights bias".

Yes it is fair to note most of the available games on the internet are the ones where he has a big performance. He did however have some bad shooting games in the 1992 Playoffs, especially vs. Cleveland and New York. For the most part I watched these games in chronological order and I recall earlier on his midrange percentage was in the 55% ballpark and his restricted area percentage was 78-79%, but his overall FG% was approaching 60%. In this (incomplete) footage he averages 33.7 points while shooting 53.5% from the floor (1672/3128), which is basically what he averaged in the 1987-88 and 1988-89 regular seasons. If his outside shot back then was not what it would become later, it would almost certainly mean he was an even better finisher at the basket.

But I don't see how this highlight bias affects his Synergy breakdown (1.98 PPP in Transition), or his Clutch statistics in the 82games.com format. His performance in crunch time (shooting over 60%) may be the most impressive in this thread. I know many statistical posters here do not believe in the idea of crunch time, but I certainly do if only for the fact that we know the mental pressure affects the player's performance (beit positive or negative). In a close game in the 4th quarter the emphasis on not wasting possessions will become that much greater. Not to say the first 3 quarters don't count, certainly they do seeing as basketball is a game of runs.

I was also surprised by the Off Court Offensive Rating. Even more telling is that Pippen was in the game for a lot of those off court minutes for Jordan, since Phil Jackson would almost always keep one of the two players in the game to play with the bench unit (unless it was garbage time). The Bulls bench was inconsistent, though I think the statistics here underrate their capabilities. Even players like Will Perdue (great passer) were excellent players within their roles off the bench. In Pippen's case, he was an excellent creator and deadly in transition, though his outside jump shot was shaky. What is most impressive how he really improved his outside shooting over his career, considering how bad he was as a rookie.

But that bench came through for them in big games as well, notably Game 6 vs. Portland in the Finals. For a good portion of that game it looked like the Blazers were going to force a Game 7, but early in the 4th quarter the bench unit led by Pippen, Hansen, etc went on a run to cut the big Blazers lead down. When Jordan subbed back in he helped to finish them off the last 6 minutes.

The only downside is the poor video quality of some of these games compared with today. Imagine seeing an explosive move off the dribble like this in high definition.

[youtube]http://www.youtube.com/watch?v=oF2iBb6fp5g[/youtube]

[colts18]

Did you keep track of MJ's defense?

[Dipper 13]

Did you keep track of MJ's defense?

If you mean individual defense then no, except in Crunch Time.

15/55 FG (27.3%)

This is primarily measuring his 1 on 1 defense in isolation, post up, and close outs on the spot up shooter. Remember 1 of the 2 Bulls playoff losses in 1991 came on a Hersey Hawkins game winner with Jordan closing out on him.


[youtube]http://www.youtube.com/watch?v=PuiQBs_z_bQ[/youtube]

[lorak]

In this (incomplete) footage he averages 33.7 points while shooting 53.5% from the floor (1672/3128), which is basically what he averaged in the 1987-88 and 1988-89 regular seasons.

We don't know how many minutes he played in your sample. So only stats we can compare to his real stats are 3P and 2P.

Over that period ('90-'92), with all playoff games included and regular season games proportionally equal to the amount in your sample he in reality was shooting 53.5% 2P and 34.5% 3P. Your sample says 54.7% 2P and 38.3% 3P. So it's 9.9% and 2.2% difference from his real stats over that period, so it seems as his stats from your sample are increased by highlight bias by about 6.1%.

But I don't see how this highlight bias affects his Synergy breakdown (1.98 PPP in Transition), or his Clutch statistics in the 82games.com format

Very easily - most of these games are better Jordan's games, so his numbers, also synergy and clutch numbers, are affected and increased by that. (and of course 52 FGA are missed from your sample, most of them probably missed shots, and it could decrease his FG% even by 1 percentage point).

As I said, I wish you would do it at least with separation between playoffs and regular season. Because in form like you presented these information might be significantly away from truth. (I mean, for example instead of being +70% at rim, he rather was just below 70%, or instead of +50% from midrange he was 48%.)

[Dipper 13]

^I was also referring to his tendencies, especially driving left vs. right in isolation. I would be shocked if he was below 90% FG in transition, but you never know. From this point on it's up to the NBA to release the pre-1997 data, though now it is clear why they haven't done so. They don't want to upset all the statistical fans who manipulate certain numbers into preaching how the 2000's is the greatest era in NBA history. They don't make money promoting the past, but rather the present.

[lorak]

From this point on it's up to the NBA to release the pre-1997 data, though now it is clear why they haven't done so. They don't want to upset all the statistical fans who manipulate certain numbers into preaching how the 2000's is the greatest era in NBA history. They don't make money promoting the past, but rather the present.

That's silly. NBA is very high on the past, they praise old player and history of the league.

[Dipper 13]

Over that period ('90-'92), with all playoff games included and regular season games proportionally equal to the amount in your sample he in reality was shooting 53.5% 2P and 34.5% 3P. Your sample says 54.7% 2P and 38.3% 3P. So it's 9.9% and 2.2% difference from his real stats over that period, so it seems as his stats from your sample are increased by highlight bias by about 6.1%.

What do you mean by proportionally equal to my sample? How did you make the adjustment from 299 games to 126 games, and determine the effect that would have on his shooting percentage? Including regular season and playoffs for those 3 seasons he shot 54.0% from 2 Pt Range.

[lorak]

Over that period ('90-'92), with all playoff games included and regular season games proportionally equal to the amount in your sample he in reality was shooting 53.5% 2P and 34.5% 3P. Your sample says 54.7% 2P and 38.3% 3P. So it's 9.9% and 2.2% difference from his real stats over that period, so it seems as his stats from your sample are increased by highlight bias by about 6.1%.

What do you mean by proportionally equal to my sample? How did you make the adjustment from 299 games to 126 games, and determine the effect that would have on his shooting percentage? Including regular season and playoffs for those 3 seasons he shot 54.0% from 2 Pt Range.

For example in your sample he has XX games from '90 regular season. So we look at his avg stats from '90 season multiple by XX... do the same for two other seasons, add all stats from three playoffs run and that's all.

[Dipper 13]

^The only way to know for sure would be for the NBA to release the data. Chances are they probably used overhead cameras to determine the shot locations. I would bet on his restricted area percentage being lower as opposed to his midrange if anything, seeing as quite a few of his missed FGA at the rim were on some incredible manufactured shots that most others wouldn't even consider attempting. Or he could have missed even more shots in the paint from the 5-15 foot range, seeing as that is where it was easiest for the defense to give help. I would almost call Jordan (and Bird) the Bill Russell of offense in his playing style and how he took advantage of the defensive gaps that most others couldn't even see.




Russell Rules: 11 Lessons on Leadership From the Twentieth Century's Greatest Winner - Bill Russell

When I played against Oscar Robertson, for example, I thought for a while that he had 360-degree vision. He seemed to know where everyone on the court was at all times; he could thread a pass through the eye of a needle, through a football scrimmage of bodies. But he didn't really have 360-degree vision at all; he had this peculiar ability great athletes have to focus. Where an ordinary player would see whatever was in front of him and whatever his peripheral vision took in, a player like Oscar would be able to eliminate everything between him and the player he wanted to pass to or the area where he had to get to shoot or set up a play. His vision then narrowed like a laser beam. Within that beam he could slow down or speed up the action as need demanded. But then everything outside that beam wasn't visible to him! I realized his great seeing power made him blind! And that allowed me to work invisibly against him. I could move into these blind zones and attack the ball or the player without being seen.

[ardee]

+44.9 On/Off....

Inconceivable. That's double of the best we've seen.

[Dipper 13]

playoffs

Playoff Only Crunch Time Statistics

24 Games Total

http://i.imgur.com/09XcdJP.png







Plus/Minus Total: +87

Plus/Minus Per 48: +42.2

Win Percentage: 70.8%





Total Statistics

http://i.imgur.com/mkjVAW4.png



Per 48 Statistics

http://i.imgur.com/pXUCEnY.png

[Gregoire]

playoffs

Playoff Only Crunch Time Statistics

24 Games Total

http://i.imgur.com/09XcdJP.png







Plus/Minus Total: +87

Plus/Minus Per 48: +42.2

Win Percentage: 70.8%





Total Statistics

http://i.imgur.com/mkjVAW4.png



Per 48 Statistics

http://i.imgur.com/pXUCEnY.png

And how about other stats for the playoffs only: on/off, shot chart, plus/minus ect?

[Gregoire]

playoffs

Playoff Only Crunch Time Statistics

24 Games Total

http://i.imgur.com/09XcdJP.png







Plus/Minus Total: +87

Plus/Minus Per 48: +42.2

Win Percentage: 70.8%





Total Statistics

http://i.imgur.com/mkjVAW4.png



Per 48 Statistics

http://i.imgur.com/pXUCEnY.png

His playoffs peak clutch scoring seems not human-like: 73,7 points/48min with 71 TS%.... For three years... its just couldnt be understanded.

[WillyNYC]

Can someone who understand all this stuff better than me put it in laymen's terms or compare it to the same type of stats for modern players like Kobe/Bron/Wade etc.? Preferably an apples to apples comparison (the reg season #'s versus regular season and the playoff numbers above vs. others' playoff #'s).

[O_6]

Fantastic job Dipper, if I could give you 10 And1’s I would. I was a big fan of the Barkley shot charts but you went above and beyond with this Jordan post. Thank you.

The thing that really stands out to me in that post is the clutch performance. Jordan is already considered as the clutchest player in NBA history yet his clutch numbers look too good to be true. I’m just amazed at those team and individual stats in the clutch. Just to bring them up again…

78.3 PTS – 12.1 REB – 8.1 AST – 5.2 STL – 3.2 BLK – 1.8 TOV – 62.4% FG – 40.0% 3P – 88.4% FT

1v1 Defense: 15/55 (27.3 FG%)
Plus/Minus: +48.8

A per 48 line of 78/12/8 on essentially 60/40/90 shooting while being extremely careful with the ball (1.8 TOV) and also being an absolute menace on both help D (8.4 stocks) and man to man D (27.3% opp FG%). That is just insanity and his team performance of +48.8 pts per 100 backs it all up and proves how much of an impact Jordan had at the end of games. This kind of data backs up the Jordan clutch mythology and then some.

I decided to compare Jordan’s performance in the clutch during this time period with the best 2 year stretch of clutch play we’ve seen from LeBron and Kobe. Jordan played 242 “clutch minutes” during this incomplete sample over 3 years, which is roughly the same as a full 2 year span. I used “82games” clutch numbers with some help from BBRef to figure out the best 2 year stretch of clutch performance from both Kobe and LeBron. I settled on 2008-2009 for Kobe and 2009-10 for LeBron, I believe this was the best 2 year streak of regular season clutch performance from either of these two players based on the stats.

Here are what LeBron and Kobe averaged per 48 over their best 2 season stretch of clutch performance, with Jordan’s 90-92 stretch next to them for comparison….

Code: Select all

PER 48 CLUTCH STATS
Player   PTS    REB   AST   STL   BLK   TOV   eFG%   FT%    +/-   Mins
Kobe     54.4   8.3   5.8   0.9   0.3   3.6   .493   .883   22.5   279
LeBron   61.7   15.2  10.1  3.3   2.6   4.4   .572   .823   40.1   262
Jordan   78.3   12.1  8.1   5.2   3.2   1.8   .652   .884   48.8   242

As great as Kobe is, he's a clear 3rd in this comparison like he is with most comparisons involving these 3. There's nothing bad about 54/8/6 per 48 on a very solid .493 eFG% (Kobe has a career .487 eFG%). The fact that his teams had a +22.5 plus/minus shows that Kobe had a really strong impact during these moments. But LeBron and Jordan trump Kobe when it comes to both clutch scoring and clutch non-scoring. Jordan and LeBron could dominate the defensive end of the game during these moments, an underrated aspect of both of their "clutchness".

And when it comes to LeBron vs. Jordan, MJ does come out clearly ahead as good as LBJ was during these moments. MJ averaged 17 points more while shooting significantly better from the field. LeBron's clutch performance during these 2 epic Cavs years are historic and among the best we've seen in the 10+ years of data we have on clutch play, but Jordan's 90-92 run trumps it.

There could be some highlight bias but I feel like this would be less of a problem for "clutch" performance since it's more likely to have complete videos for these end of game situations. And even if there is a slight highlight bias that could make these numbers look a little better for Jordan, there's also the fact that he played in the playoffs for a big chunk of those 242 clutch minutes. It's much tougher to perform highly in the clutch during the playoffs. Kobe and LeBron's numbers are regular season only. So I think even if there is highlight bias, the playoff factor makes up for it.

Long story short... Jordan is considered the best clutch player ever and yet he was probably even more clutch than people think.

[parapooper]

So for a 40 minute game (puts it better in context for me personally) that would be
65/10/6.7/4.3/2.7

Not sure what is more insane - 65 pts @ 65eFG%, 10 rbds for an SG, ast/TO ratio of 4.5 or the 4.3 stls + 2.7 blocks. It's mindboggling in every single aspect of the game and defense outside of boxscore as well apparently.
I always thought LeBrons 09/10 had a good shot at GOAT clutch but MJ's numbers are absolutely unbelievable. There is no way anyone will get close to that ever. Does someone have an excel sheet to calculate the PER for that - it's probably in the 60s.

MJ must have been totally phoning it in in non-clutch time if he can do that when required.