Michael Jordan -- Complete Playoff Shooting Data

[70sFan]

Due to popular demand, here is the playoff midrange data for some legends. MJ looks pretty good!

Player: FG/FGA (%); FGA/game

Michael Jordan: 1040/2279 (45.6%); 12.7
Demar Derozan: 208/559 (37.2%); 8.9
Dirk Nowitzki: 563/1230 (45.8%); 8.5
Kobe Bryant: 759/1854 (40.9%); 8.4
Kevin Durant: 601/1238 (48.5%); 7.3
Chris Paul: 374/743 (50.3%); 5.0
Steve Nash: 232/517 (44.9%); 4.3
Stephen Curry: 225/487 (46.2%); 3.3

GOAT-level midrange shooter, at least among players we have data from.

[jjgp111292]

Due to popular demand, here is the playoff midrange data for some legends. MJ looks pretty good!

Player: FG/FGA (%); FGA/game

Michael Jordan: 1040/2279 (45.6%); 12.7
Demar Derozan: 208/559 (37.2%); 8.9
Dirk Nowitzki: 563/1230 (45.8%); 8.5
Kobe Bryant: 759/1854 (40.9%); 8.4
Kevin Durant: 601/1238 (48.5%); 7.3
Chris Paul: 374/743 (50.3%); 5.0
Steve Nash: 232/517 (44.9%); 4.3
Stephen Curry: 225/487 (46.2%); 3.3

So I'm assuming that in your methodology, mid-range jumpers = jumpshots outside of the paint, not simply all 2-point jumpers, right? Because in that context - yeah, those numbers are very strong. Because by basketball reference splits, MJ in 97 was 150/326 (46%) on 2-point jumpers and 142/352 (40.3%) in 98. So yeah, like lessthanjake said, I'd expect a jump if you split the data in line with basketball reference's style, and so people going "that's not that impressive!" if they're comparing the numbers to other people's BBR splits is a hasty judgment lol

[lessthanjake]

I think the talk of “expectations” needs to be calibrated in a realistic way. We have always known exactly what Jordan’s playoff FG% is. So if you think something in Djoker’s data is lower than you expected, then definitionally something else must also be better than you expected (whether it’s a higher FG% than you expected from a different range, or a more efficient distribution in terms of shot area than you expected). Either that or your “expectations” were simply not actually calibrated to the objective information we already had.

[jjgp111292]

I think the talk of “expectations” needs to be calibrated in a realistic way. We have always known exactly what Jordan’s playoff FG% is. So if you think something in Djoker’s data is lower than you expected, then definitionally something else must also be better than you expected (whether it’s a higher FG% than you expected from a different range, or a more efficient distribution in terms of shot area than you expected). Either that or your “expectations” were simply not actually calibrated to the objective information we already had.

Yeah, it's really easy to not be impressed when you set some arbitrary expectation based on "mythology." Almost like...bad faith

[Redmoon]

Even relative to era I'm not really sure Jordan faced significantly tougher defenses than other all-time greats. Also wouldn't include the Jazz in that, even in 97 they were only 10th, not noteworthy in the grand scheme of things. Also would echo falco, fantastic numbers especially considering the volume, but nothing out of this world.

They were a physical team with a large-ish frontline and some good perimeter defenders. Their O was better than their D, obviously, and they didn't stand up to teams with Ewing or Mourning sure, but they were no pushovers.

The perimeter shooting IS out of this world on the volume he was producing (45.6%, remember), and 65% at the rim in-era was pretty good. His short game is the part that's underwhelming, and you'll note it represented 12.6% of his total volume.

Food for thought.

I mean sure they were good but they aren't in the same realm as the Pistons or Knicks who were legit ATG defenses, especially in 98.

And based off the data I would feel reasonably comfortable calling him the GOAT midrange shooter, mostly due to volume, but Dippers data originally had him at like 52%, which I always thought was probably too high, but maybe my expectations were unrealistic. And would love peer data on the rim finishing but regardless after 91 his rim finishing merely looks good as opposed to excellent, and honestly worse than I would've expected. Based off this, I don't really think he has an argument as the best slasher ever.

Dippers data includes the regular season. The numbers do checkout its dippers synergy stuff and on-off that is prob wrong.
MJ around 50% in midrange at his peak in regular season is very believable. Might be a percentage point or 2 higher from the sample of games he has (skew towards good games).

I had the same impression about his rim numbers as well. The peak years are fantastic of course. Those compare favorably to anyone even Wade. But his finishing always hovered around 60% after those years. Seems to coincide with his steady decline in athleticism and changing his game to more post ups and jumpers.

Whats interesting to me is his numbers at the end of his second 3 peat. 1997 and especially 1998 was relatively bad for his standards. The league was slowing down significantly and becoming a grindfest.

[Djoker]

They were a physical team with a large-ish frontline and some good perimeter defenders. Their O was better than their D, obviously, and they didn't stand up to teams with Ewing or Mourning sure, but they were no pushovers.

The perimeter shooting IS out of this world on the volume he was producing (45.6%, remember), and 65% at the rim in-era was pretty good. His short game is the part that's underwhelming, and you'll note it represented 12.6% of his total volume.

Food for thought.

I mean sure they were good but they aren't in the same realm as the Pistons or Knicks who were legit ATG defenses, especially in 98.

And based off the data I would feel reasonably comfortable calling him the GOAT midrange shooter, mostly due to volume, but Dippers data originally had him at like 52%, which I always thought was probably too high, but maybe my expectations were unrealistic. And would love peer data on the rim finishing but regardless after 91 his rim finishing merely looks good as opposed to excellent, and honestly worse than I would've expected. Based off this, I don't really think he has an argument as the best slasher ever.

Dippers data includes the regular season. The numbers do checkout its dippers synergy stuff and on-off that is prob wrong.
MJ around 50% in midrange at his peak in regular season is very believable. Might be a percentage point or 2 higher from the sample of games he has (skew towards good games).

I had the same impression about his rim numbers as well. The peak years are fantastic of course. Those compare favorably to anyone even Wade. But his finishing always hovered around 60% after those years. Seems to coincide with his steady decline in athleticism and changing his game to more post ups and jumpers.

Whats interesting to me is his numbers at the end of his second 3 peat. 1997 and especially 1998 was relatively bad for his standards. The league was slowing down significantly and becoming a grindfest.

The 1997 and 1998 numbers (especially 1997) are known to be pretty wonky. About 1/4 of all field goals is reportedly mis-labelled according to this article.

https://www.sports-reference.com/blog/2020/04/nba-play-by-play-lineup-and-shooting-stats-added-back-to-1996-97/

I plan to track those two postseasons in the future as well. I was actually debating whether to do those first and then release all the data but I wanted to release something around Christmas!

[jjgp111292]

I mean sure they were good but they aren't in the same realm as the Pistons or Knicks who were legit ATG defenses, especially in 98.

And based off the data I would feel reasonably comfortable calling him the GOAT midrange shooter, mostly due to volume, but Dippers data originally had him at like 52%, which I always thought was probably too high, but maybe my expectations were unrealistic. And would love peer data on the rim finishing but regardless after 91 his rim finishing merely looks good as opposed to excellent, and honestly worse than I would've expected. Based off this, I don't really think he has an argument as the best slasher ever.

Dippers data includes the regular season. The numbers do checkout its dippers synergy stuff and on-off that is prob wrong.
MJ around 50% in midrange at his peak in regular season is very believable. Might be a percentage point or 2 higher from the sample of games he has (skew towards good games).

I had the same impression about his rim numbers as well. The peak years are fantastic of course. Those compare favorably to anyone even Wade. But his finishing always hovered around 60% after those years. Seems to coincide with his steady decline in athleticism and changing his game to more post ups and jumpers.

Whats interesting to me is his numbers at the end of his second 3 peat. 1997 and especially 1998 was relatively bad for his standards. The league was slowing down significantly and becoming a grindfest.

The 1997 and 1998 numbers (especially 1997) are known to be pretty wonky. About 1/4 of all field goals is reportedly mis-labelled according to this article.

https://www.sports-reference.com/blog/2020/04/nba-play-by-play-lineup-and-shooting-stats-added-back-to-1996-97/

I plan to track those two postseasons in the future as well. I was actually debating whether to do those first and then release all the data but I wanted to release something around Christmas!

I've never trusted MJ's percentage at the rim for the 97 regular season, especially considering how much better the numbers are in 98 when he's less athletic lmao. Although the 93-96 playoff numbers seem to align with it so who knows.

So for 1996-97, in particular, and later 90s seasons to a lesser extent, field goal percentages at the rim and on layups are artificially deflated and field goal percentages on jump shots are artificially inflated

Yup, sounds about right

[Caneman786]

Due to popular demand, here is the playoff midrange data for some legends. MJ looks pretty good!

Player: FG/FGA (%); FGA/game

Michael Jordan: 1040/2279 (45.6%); 12.7
Demar Derozan: 208/559 (37.2%); 8.9
Dirk Nowitzki: 563/1230 (45.8%); 8.5
Kobe Bryant: 759/1854 (40.9%); 8.4
Kevin Durant: 601/1238 (48.5%); 7.3
Chris Paul: 374/743 (50.3%); 5.0
Steve Nash: 232/517 (44.9%); 4.3
Stephen Curry: 225/487 (46.2%); 3.3

Chris Paul is such a demon

[DraymondGold]

Fantastic work Djoker! Clearly a lot of effort went in, and it’s really great stuff.

A. Multi-year Averages
I think it’s informative to break things into different time spans. Multi-year averages have less noise than single year data, but allow for more detail then just full career averages.

Looking at three-year averages:

Spoiler:

Timespan: Rim FG%, Paint (no rim) FG%, Midrange FG%, 3P FG%
85-87: 59.3, 35.8, 42.2, 28.6
88-90: 70.4, 34.7, 47.1 (11.5 FGA), 30.7
89-91: 70.8, 36.6, 47.1 (11.1 FGA), 32.4
90-92: 67.7, 37.5, 48.1 (12.9 FGA), 35.8
91-93: 63.3, 35.6, 47.9, 38.7
96-98: 63.1, 47.2, 42.4, 29.7

It’s impressive how well this tracks with Jordan’s perceived career ark. Going from his early career 85-87 to peak 88-90/89-91: we see a massive boost in rim finishing, clear improvement in midrange, and slight improvement in 3 point shooting. I would expect some of Jordan’s worse rim shooting early on to be explained by the small sample of games and the opponents. The 85 Bucks and 86-87 Celtics were both fantastic defenses, and were the only opponents MJ faced in the first 3 years. Still, I would believe some of the improvement at the rim was real growth as Jordan peaked. The 1986 series stands out in its paint (no-rim) scoring. Jordan was blazing hot from that area, which tracks with what I remember watching those games. Still, this series does bring up Jordan’s 3-year average from that region -- comparing 85/87 to the following peak years, Jordan also improved in the paint (no-rim) scoring.

Looking at his peak: 88-91 seems to be the clear best 4-year run at the rim, with 1990 and 1991 being the best 2 years in that order, and a drop off going into 1992. Looking at the midrange, we see Jordan level up in 1988 and become more consistent in the midrange in 1990. The 3 point shooting is super small sample in the first 4 years (and still small sample even in the other years), but if we squint, we can see an increase in 3 point volume in 1989, and an increase in accuracy over 89 to 91. Moving through to Jordan’s late peak (e.g. 90-92), we see a gradual decline at the rim, gradual improvement in midrange, and gradual improvement from 3.

1990 and 1991 stand out as Jordan’s best shooting years overall. This coincides with Jordan’s plus minus data — neat! 1990 looks marginally better at the rim and in the midrange, 1991 look marginally better in the paint (non-rim) and from 3. Both are close enough in raw percents for differences between region to be explained away by noise, context, or opponents faced. Still, both years stand out as a timespan where Jordan put it all together — his rim shooting, his midrange, and his 3 point shooting (era-relative, especially 1991). To me, one of the things that gives Jordan the best GOAT peak argument is that he ‘put it all together’ better than the other GOAT peak candidates, who peaked in different skills at different times. This shooting data supports this idea that Jordan’s rim pressure, midrange ability, and distance shooting peaked simultaneously around ~1990–1991.

Looking past his peak: we see a decline in rim shooting, but sustained GOAT midrange shooting at higher volume, and good era-relative 3 point shooting. In 1995, we see Jordan’s lowest rim numbers of any of the 90s Bulls years, supporting the idea he wasn’t back in full shape yet. By 1997–1998, Jordan’s distance shooting from midrange and 3 begins to taper off. In 1997, we do see some evidence of inconsistent tracking in the official NBA data, as Djoker and others discussed earlier. Looking at both Jordan’s data and the rest of the nba, I believe some of this is the misattribution of shots at the rim that were wrongly counted as shots in the paint (no rim) region. We see an unexpected dip in Jordan’s numbers at the rim at the same time as an unexpected jump in Jordan’s paint (no rim) numbers in 97 compared to other years. Adjusting for this, I’d expect 2nd 3-peat Jordan’s true rim numbers to look a little better and his paint numbers to look a little worse.

Here are Jordan’s numbers broken up into other time frames:

Spoiler:

Timespan: Rim FG%, Paint (no rim) FG%, Midrange FG%, 3P FG%
4-year peak
88-91: 70.9, 35.7, 47.1, 32.5
89-92: 67.5, 36.9, 47.6, 34.2

5-year peak
87-91: 69.7, 34.8, 46.8, 32.8
88-92: 68.1, 36.2, 47.6, 34.2
89-93: 65.9, 35.6, 47.6, 35.7
91-96: 62.3, 35.4, 47.5, 38.9

7-season prime
87-93: 66 , 34.6, 47.4, 35.7
88-95: 65.9, 35.4, 47.8, 35.8
89-96: 64.8, 35.5, 47.4, 36.7

8-season prime
86-93: 65.8, 35.4, 47.3, 36
88-96: 65.5, 35.2, 47.3, 36.6

10-year prime
87-96: 65 , 34.7, 47.2, 36.7

10-season prime
86-96: 64.9, 35.3, 47.1, 36.9
87-97: 64.5, 37.9, 46.7, 33.8
88-98: 65.1, 39.1, 45.8, 33.3

12-year prime
86-97 64.3, 38.3, 46.6, 33.9
87-98 64.7, 38.6, 45.7, 33.4

Note: by X-year prime, I’m looking at an X year timespan, and so am not expanding the sample for Jordan’s missed playoffs in 1994. For Y-season prime, I’m looking at Y playoffs, and so will add an adjacent year to replace Jordan’s missed playoffs in 1994 (so e.g. a 10-season prime covers and 11-year time frame).

B. Comparison with other players.

It’s of course interesting to compare Jordan’s numbers to other all-time scorers in different regions of the floor. I’m not sure the best way to do this. Nba.com player stats have the best data (breaking it down by region like we have here for Jordan), but their UI is pathetic. BBR has much better UI, but with different definitions of zones per above. I found these reddit threads from 5 years ago showing career averages (not sure if RS, or RS+PS) from players scoring at the rim (https://www.reddit.com/r/nba/comments/gcg79t/who_are_statistically_the_best_finishers_at_the/) and in the midrange (https://www.reddit.com/r/nba/comments/g53347/who_are_the_greatest_mid_range_shooters/) which give an initial sense of the best scores in each region. It’s a lot of the usual suspects.

In the midrange, Dirk looks the best, being 2nd in FG% (just behind Nash) and first with some separation in FGA.
At the rim, Wade looks best among guards, being 2nd in FG% (just behind Bradley beal) and first with some separation in FGA. As expected, Shaq and LeBron look best at the center and wing positions respectively.

Let’s compare to Dirk and Wade. Era differences will make this comparison more difficult — see section C for discussion there.

Dirk’s midrange :

Spoiler:

source: https://www.nba.com/stats/player/1717/shooting?Season=2003-04&SeasonType=Playoffs
Year Midrange FGM FGA FG%
2004 23 54 42.6
2005 48 117 41
2006 93 203 45.8
2007 19 45 42.2
2008 17 41 41.5
2009 45 87 51.7
2010 31 60 51.7
2011 96 196 49
2012 20 35 57.1

3-year, FG%
04-06 43.9
06-08 44.6, (8.5 FGA)
07-09 46.8, (8.2 FGA)
08-10 49.5, (9.0 FGA)
09-11 50.1, (9.3 FGA)
10-12 50.5

4-year peak, FG%
08-11 49.2
09-12 50.8

5-year peak, FG%
06-10 47
07-11 48.5
08-12 49.9

7-season prime, FG%
04-10 45.5
05-11 46.6
06-12 48.1

8-season prime, FG%
04-11 46.3
05-12 47.1

So Jordan’s 3-year peak in 89-91 and 91-92 has higher midrange FG% than any Dirk year up until 2010. In 3-year peaks, 89-91 Jordan shot +2.5 FG% (+0.3 FG%) better than 06-08 (07-09) Dirk with +2.6 (+2.9) more FGA per game. 90-92 Jordan’s advantage is even larger. On the other hand, Dirk does seem to have a higher raw midrange FG% than Jordan when looking at his late peak years on lower volume. Comparing late peaks, 90-92 Jordan shot -2.0 FG% worse than 09-11 Dirk but on +3.6 more FGA per game. In 5-year peaks, 88-92 Jordan shot -0.9 FG% worse than 07-11 Dirk but on +2.8 FGA. Super impressive stuff from both.

In longer prime samples (e.g. 8-season primes), we see Jordan’s nearly unparalleled consistency at play, helping Jordan have higher midrange FG%. And of course Jordan has the clear volume advantage over his prime. Both 86-93 Jordan and 88-96 Jordan shot +1.0 FG% (+0.2 FG%) at +3.5 FGA (+3.6 FGA) over 04-11 (05-12) Dirk.

Era-relative context probably plays a role both in peak Dirk’s raw efficiency advantage and peak Jordan’s raw volume advantage.

Wade’s Rim scoring:

Spoiler:

source: https://www.nba.com/stats/player/2548/shooting?Season=2004-05&SeasonType=Playoffs
Year Midrange FGM FGA FG%
2005 61 102 59.8 23 44 52.3
2006 93 147 63.3 23 60 38.3
2007 15 26 57.7 4 9 44.4
2009 20 37 54.1 9 17 52.9
2010 31 39 79.5 6 12 50
2011 95 144 66 26 70 37.1
2012 97 140 69.3 40 89 44.9
2013 70 109 64.2 27 72 35.4
2014 45 65 60 44 72 61.1

3-year, FG%
05-07 61.5 44.2
06-09 61 41.9
07-10
09-11 66.4 41.4
10-12 69 42.1
12-14 67.5 47.6

4-year peak, FG%
06-10 63.9 42.9
07-11 65.4 41.7
09-12 67.5 43.1

5-year peak, FG%
05-10 62.7 45.8
06-11 64.6 40.5
07-12 66.8 43.1

7-season prime, FG%
05-12 64.9 43.5
06-13 65.6 41

8-season prime, FG%
05-13 64.8 42.4
06-14 65.9 44.6

So peak Jordan seems to be a full level above Wade (the best guard in previously available data at rim scoring) in raw rim FG%. Comparing 3-year peaks, 89-91 Jordan’s better by +4.3 FG% over 09-11 Wade and +1.8 FG% over 10-12 Wade, a full 20 years earlier. Impressive stuff! He also has the volume advantage: 89-91 Jordan shot +0.6 FGA over 09-11 Wade and +0.7 FGA over 10-12 Wade.

Compared to early LeBron:

Spoiler:

LeBron Rim
Year Rim FGM FGA FG%
2008 13 61 87 70.1
2009 14 81 122 66.4
2010 11 52 76 68.4
2011 81 120 67.5
2012 23 136 181 75.1
2013 23 114 165 69.1

3-year, FGA, FG%
08-10 7.5 68.1
09-11 6.9 67.3
11-13 7 71

4-year peak , FG%
09-12 70.1
10-13 70.7

5-year peak, FGA, FG%
08-12 7.1 70.1
09-13 7.2 69.9

To my great surprise… peak Jordan actually has higher efficiency at the rim than 1st Cavs LeBron! 89-91 Jordan’s +2.7 FG% (-0.2 FG%) over 08-10 (11-13) LeBron, at -0.2 FGA (+0.3 FGA) per game. Again, this is a full 20 years earlier, so this likely comes with significantly less spacing. Wow! In 5-year peaks, 88-92 Jordan’s -0.9 FG% on +0.5 FGA compared to 09-13 LeBron. Early peak 87-91 Jordan’s -0.2 FG% (-0.4 FG%) on +0.6 FGA (+0.7 FGA) compared to 09-13 (08-12) LeBron. LeBron’s efficiency shoots up in his later years, e.g. in his 2nd Cleveland stint. I doubt LeBron actually improved at the rim as his athleticism declined. Instead, I think the continued growth of 3 point spacing started to make driving easier. Which brings us to…

C. Era-relative Comparison:

I haven’t seen much discussion in this thread about era-relative differences — which to me seems paramount before claiming Jordan’s raw FG% looks more or less impressive than some more recent player.

As should be obvious to anyone here, 3 point shooting has continuously increased since Jordan’s time, improving spacing and improving raw FG% from a variety of zones on the floor. In a cursory look at percentages across era, efficiency in each zone starts clearly increasing by the mid 2010s compared to earlier eras. So comparing the raw FG% of 90s Jordan (or another early 2000s star) to a player after the early 2010s is a bit of an apples to oranges comparison. The improvement in spacing varied by team — some teams certainly had more spacing in the early 2010s and late 2000s too. And it’s worth emphasizing we have no real measure of efficiency by zone in the early 90s or late 80s, although we may be able to do a rough projection. There’s a full ~20 years difference between peak Jordan and peak Wade/Dirk, and even more when comparing Jordan to later players.

At the rim: I would break things down into 3 forms of scoring. (1) Rim in the fast break, (2) open rim scoring in the half court from e.g. a break down in defense, and (3) rim scoring over a defender or through traffic. I would expect increased 3 point shooting to increase teams ability to break down the defense in the half court (2), and potentially generate more fast breaks since opposing 3 point attempts generate longer rebounds (1). Relaxed dribbling rules and widened driving lanes should even increase raw rim FG% while defended in the half court (3).

For a true comparison of rim scoring, we should adjust for how much scoring comes from each category, and for the spacing of the team/era. We might be able to project by looking at rim FG% over time, or perhaps by seeing what variables correlate most strongly with rim FG% (e.g. team 3PA for spacing or pace for fast breaks) and using that to project back. This sort of projecting is made harder by the time and style changes of the early/mid 2000s. The shift towards a more defense-focused rule set and league partially counteracts the increase in 3s, making the trend less clear when comparing the late 90s to the mid 2000s.

In the paint (no rim): here’s where Jordan’s raw FG% is ‘most disappointing’ when comparing to later era players, but this is also the region of the floor that benefits most from spacing. Here, I would expect we don’t have to worry as much about fast break scoring or rate of open shots biasing our numbers as much. So the lower raw FG% is almost certainly due to the more cramped spacing.

In the midrange: era-relative changes certainly help explain Jordan’s volume advantage, although he’s enough of a singular outlier that he’a still all-time in volume accounting for era. In efficiency, we see slight improvements in midrange efficiency by the mid 2010s as the floor opens up. Some of this may also be driven up by what I might call ‘Morey Defense’ (the defensive equivalent of Morey assists), which the defense tries to drive people off the 3 point line then uses zone principles to discourage shots too close to the rim. This is akin to how players like Durant and a Chris Paul are more often able to get to their midrange spots. If a player generates a defensive breakdown or a 5 on 4 e.g. in the pick and roll, the defense might try to preemptively collapse, preferring to give up a more open shot at the midrange or the elbow than the rim. Similarly, since players prioritize getting 3 pointers or shots at the rim more recently, they may only take midrange shots if they’re relatively more open. This also drives down volume and boosts efficiency. Curry, Durant, and Chris Paul all seem more efficient in raw FG% in the midrange, but all shot at lower volume with far more spacing. It’s not quite clear to me how much to do a fair comparison adjusting for the volume and spacing difference.

Other factors: All this doesn’t account for defenses faced in the playoffs (or small samples or player injuries), which makes a much larger defense in the playoffs. We could get a first-pass estimate looking at average opponent defense faced, which might be worth doing. But this wouldn’t account for varying defensive matchups, or for how team defenses might affect each zone differently. Tracking data suggests good individual defenders usually have a larger effect on rim efficiency than 3 point efficiency, but it does suggest 3 point efficiency varies with how open the shot is, and better defenders/defenses might be better at preventing open shots. I haven’t looked at how much individual defenders or team defenses affect each zone — rim, paint, midrange, 3 — in any detailed way. Perhaps someone else might devote the time.

[falcolombardi]

Fantastic work Djoker! Clearly a lot of effort went in, and it’s really great stuff.

1990 and 1991 stand out as Jordan’s best shooting years overall. This coincides with Jordan’s plus minus data — neat! 1990 looks marginally better at the rim and in the midrange, 1991 look marginally better in the paint (non-rim) and from 3. Both are close enough in raw percents for differences between region to be explained away by noise, context, or opponents faced. Still, both years stand out as a timespan where Jordan put it all together — his rim shooting, his midrange, and his 3 point shooting (era-relative, especially 1991). To me, one of the things that gives Jordan the best GOAT peak argument is that he ‘put it all together’ better than the other GOAT peak candidates, who peaked in different skills at different times. This shooting data supports this idea that Jordan’s rim pressure, midrange ability, and distance shooting peaked simultaneously around ~1990–1991.

To my great surprise… peak Jordan actually has higher efficiency at the rim than 1st Cavs LeBron![/u] 89-91 Jordan’s +2.7 FG% (-0.2 FG%) over 08-10 (11-13) LeBron, at -0.2 FGA (+0.3 FGA) per game. Again, this is a full 20 years earlier, so this likely comes with significantly less spacing. Wow! In 5-year peaks, 88-92 Jordan’s -0.9 FG% on +0.5 FGA compared to 09-13 LeBron. Early peak 87-91 Jordan’s -0.2 FG% (-0.4 FG%) on +0.6 FGA (+0.7 FGA) compared to 09-13 (08-12) LeBron. LeBron’s efficiency shoots up in his later years, e.g. in his 2nd Cleveland stint. I doubt LeBron actually improved at the rim as his athleticism declined. Instead, I think the continued growth of 3 point spacing started to make driving easier. Which brings us to…

Well for starters the 80's and early 90's were a higher efficiency era than the 00's and early 10's, not the other way around

Second if lebron is less efficient at the rim yet a more efficient overall player with a) worse free throw shooting, b) less ft calls, c) worse midrange shooting yet less efficient at the rim too? Where does the efficiency advantage come for lebron then?

Only explanation would be much higher rim volume which, again, would make any small efficiency edge jordan would have be at the expense of much lesser volume

This is with lebron spacing being potentially worse in a post illegal defense but pre 3 point explosion era

Also efficiency is era adjusted lol, so even if rim efficiency was tougher in jordan era he would be being compared to players who also had it tougher

ben taylor used say that jerry west was better than oscsr at every area of the floor ( the rim, jumpshooting, long range) despite being less efficient overall (slightly) than oscar so where was the efficiency from oscar coming if he was worse at every scoring area of the floor?

I feel the same way here when people argue jordan is simultaneously, somehow, a better finisher, ft shooter, jumpshooter all at the same time and somehow still less efficient

[DraymondGold]

Fantastic work Djoker! Clearly a lot of effort went in, and it’s really great stuff.

1990 and 1991 stand out as Jordan’s best shooting years overall. This coincides with Jordan’s plus minus data — neat! 1990 looks marginally better at the rim and in the midrange, 1991 look marginally better in the paint (non-rim) and from 3. Both are close enough in raw percents for differences between region to be explained away by noise, context, or opponents faced. Still, both years stand out as a timespan where Jordan put it all together — his rim shooting, his midrange, and his 3 point shooting (era-relative, especially 1991). To me, one of the things that gives Jordan the best GOAT peak argument is that he ‘put it all together’ better than the other GOAT peak candidates, who peaked in different skills at different times. This shooting data supports this idea that Jordan’s rim pressure, midrange ability, and distance shooting peaked simultaneously around ~1990–1991.

To my great surprise… peak Jordan actually has higher efficiency at the rim than 1st Cavs LeBron![/u] 89-91 Jordan’s +2.7 FG% (-0.2 FG%) over 08-10 (11-13) LeBron, at -0.2 FGA (+0.3 FGA) per game. Again, this is a full 20 years earlier, so this likely comes with significantly less spacing. Wow! In 5-year peaks, 88-92 Jordan’s -0.9 FG% on +0.5 FGA compared to 09-13 LeBron. Early peak 87-91 Jordan’s -0.2 FG% (-0.4 FG%) on +0.6 FGA (+0.7 FGA) compared to 09-13 (08-12) LeBron. LeBron’s efficiency shoots up in his later years, e.g. in his 2nd Cleveland stint. I doubt LeBron actually improved at the rim as his athleticism declined. Instead, I think the continued growth of 3 point spacing started to make driving easier. Which brings us to…

Well for starters the 80's and early 90's were a higher efficiency era than the 00's and early 10's, not the other way around

During Jordan’s peak and the surrounding years (say 1986-1993), the league average eFG% ranged from 48.7 to 49.3. During Jordan’s later years (1995-1998), average ranged from 47.8 to 50.0.

From Jordan’s 2nd retirement through to the end of hand-checking (1999-2004), league average ranged from 46.6 to 47.8 — definitely slightly lower, although you’ll see that I mentioned this in my post in discussions about changing rules and styles.

But then with the abolition of hand checking in 2005 (to counteract the start of zone defenses and the general dead ball style), we see it rise to 48.2, then 49.0, then 49.6. From 2007 to 2016, league average eFG% never gets below 49.6 and rises to 50.2 — clearly more efficient than Jordan’s league during his peak and surrounding prime. Since 2017, league average began at 51.4 and rose to 54.7 — as described by the mid 2010s jump discussion in my post.

eFG% corrects for the change in 3 point volume but doesn’t correct for changes in shooting zone from 2 point range. The thing to do would be to look at league efficiency by each zone (rim, paint, midrange, 3, overall) over time. I wouldn’t be surprised if part of the inefficiency from 1999-2004 was an increase in midrange and a decrease in fast break offense at the rim (the style changes I alluded to), in addition to the defense-centric rules from 2002-2004 (zone defense and hand checking simultaneously). But at least in eFG%, from 2007 onward (which includes most of the years we care about for Wade, Dirk, LeBron, etc), league average efficiency was higher than what it was during peak and surrounding prime Jordan, particularly after 2016. Pretty much like I said in my post — do you read the trends differently?

Second if lebron is less efficient at the rim yet a more efficient overall player with a) worse free throw shooting, b) less ft calls, c) worse midrange shooting yet less efficient at the rim too? Where does the efficiency advantage come for lebron then?

Only explanation would be much higher rim volume which, again, would make any small efficiency edge jordan would have be at the expense of much lesser volume

You’re welcome to check the efficiency by each zone too. But in short — we’re talking about raw playoff Fg% in each zone, so I’m not actually sure he’s more efficient. 2008-2010 (11-13) PS LeBron’s at 47.4 FG% (48.7 FG%), while 89-91 PS Jordan’s at 51.6 Fg%.

This is with lebron spacing being potentially worse in a post illegal defense but pre 3 point explosion era

Re: LeBron’s spacing, I definitely disagree here. I would be quite surprised if 09 LeBron had less spacing than 91 Jordan, with a team shooting 4x as many 3s (20.4 vs 5.2) a full ~20 years later. The film analysis I’m aware of on the topic — eg Thinking Basketball’s analysis of spacing across history and rule changes across history — certainly supports there being an increase in spacing from 1990 to 2010.

Also efficiency is era adjusted lol, so even if rim efficiency was tougher in jordan era he would be being compared to players who also had it tougher

Indeed. That’s why I brought up era relative in the first place lol

ben taylor used say that jerry west was better than oscsr at every area of the floor ( the rim, jumpshooting, long range) despite being less efficient overall (slightly) than oscar so where was the efficiency from oscar coming if he was worse at every scoring area of the floor?

I feel the same way here when people argue jordan is simultaneously, somehow, a better finisher, ft shooter, jumpshooter all at the same time and somehow still less efficient

It’s quite possible for someone to be more efficient from every zone on the floor, and less efficient overall, if they have a larger proportion of their volume from their less efficient zone. Player A could be more efficient in every zone than Plauer B, but have a much higher percentage of their volume from one of their worse zones (e.g. midrange), while Player B has a much higher percentage of their volume from one of their better zones (e.g. rim). There’s a difference between shooting ability and shot selection. The same phenomenon happens with Garnett and Duncan in some years. But as it relates here, Jordan actually does have better FG% overall than LeBron in the 1st Cavs years I mentioned — so the point doesn’t quite apply.

[jjgp111292]

Yeah the arguments in favor of LeBron in this context would make more sense if you were talking about the late 90s-mid 00s. But excluding 2012 and 13, the league in LeBron's prime prime(Late Cleveland and Miami) was just as efficient as the 80s and early 90s and while spacing isn't what is today, it was certainly better than the early 80s and 90s. And yeah, even with the 09 playoffs included, LeBrons overall efficiency during that period was worse than Peak MJ (+4 rts for Bron 08-13 Bron vs +5 rts for 88-93 Jordan. And im guesstimating this because im on my phone and I know league average TS during all these years besides 2012 hovered around 53-54 ts, never going higher than like 54.1...actually since the TS in these eras was virtually the same Ill just do the raw numbers lmao - 57.7 for Lebron, 58.3 for MJ)

[jjgp111292]

Also just checked BBall ref for 97 again...according those splits, league average FG% at the rim was 55% compared to 63% in 98 and 62% in 99, both way less efficient years overall....so yeah, that 97 number CAN'T be right Because if even washed Wizards MJ was finishing 62% at the rim there's no way in hell 97 Jordan was at 52%

[Jaqua92]

Due to popular demand, here is the playoff midrange data for some legends. MJ looks pretty good!

Player: FG/FGA (%); FGA/game

Michael Jordan: 1040/2279 (45.6%); 12.7
Demar Derozan: 208/559 (37.2%); 8.9
Dirk Nowitzki: 563/1230 (45.8%); 8.5
Kobe Bryant: 759/1854 (40.9%); 8.4
Kevin Durant: 601/1238 (48.5%); 7.3
Chris Paul: 374/743 (50.3%); 5.0
Steve Nash: 232/517 (44.9%); 4.3
Stephen Curry: 225/487 (46.2%); 3.3

I know some people here say that the numbers demythify Jordan as a mid range jump shooting GOAT...

But who else is it? Dirk and KD? They're all in the same ball park, with KD being the most effecient. But Jordan's volume is nearly doubled.

He's the best rid range jump shooter ever. There's no debate.

[TheGOATRises007]

Due to popular demand, here is the playoff midrange data for some legends. MJ looks pretty good!

Player: FG/FGA (%); FGA/game

Michael Jordan: 1040/2279 (45.6%); 12.7
Demar Derozan: 208/559 (37.2%); 8.9
Dirk Nowitzki: 563/1230 (45.8%); 8.5
Kobe Bryant: 759/1854 (40.9%); 8.4
Kevin Durant: 601/1238 (48.5%); 7.3
Chris Paul: 374/743 (50.3%); 5.0
Steve Nash: 232/517 (44.9%); 4.3
Stephen Curry: 225/487 (46.2%); 3.3

I know some people here say that the numbers demythify Jordan as a mid range jump shooting GOAT...

But who else is it? Dirk and KD? They're all in the same ball park, with KD being the most effecient. But Jordan's volume is nearly doubled.

He's the best rid range jump shooter ever. There's no debate.

Disingenuous posting honestly. Typical on here lately.

You're either on 1 side or the other, and there's no in between.

[lessthanjake]

So I ran some numbers to get a sense of the efficiency of these mid-range shots by era for Jordan, Dirk, and Durant.

Specifically, here’s what I did:

For each player, I took a weighted average of the league’s regular season eFG% each year they played in the playoffs, weighted by the number of mid-range shot attempts they took in those years. This is meant to get a weighted average of general league efficiency when these players took their mid-range shots. I then compared that weighted-average eFG% and compared it to these players’ mid-range FG% (using Djoker’s data for Jordan and using data from the NBA website for 2-point shots outside the paint in the playoffs for Dirk and Durant). By subtracting that weighted-average eFG% from the mid-range FG% for these players, we get a relative eFG% on mid-range shots.

The result for these players was essentially the same. Jordan’s was a relative FG% on mid-range shots of -3.33, Dirk’s was -3.27, and Durant’s was -2.74. Jordan is very slightly lower than these guys, but they’re basically the same, and Jordan’s substantially higher volume makes him look the most impressive here IMO.

Of course, I will caveat this analysis by saying that it is a pretty blunt instrument. Since we do not have league data by different ranges for the vast majority of Jordan’s career, we can’t get relative eFG% on mid-range shots specifically. Doing it relative to league effective FG% is a substitute, but obviously is also dependent on league efficiency on other types of shots. In a sense, though, it does give us a broader sense of how efficient these players’ mid-range shots were within their league context, which may be just as useful as a more targeted analysis of league mid-range FG% might be. It is also using the relative eFG% compared to league average in those seasons, as opposed to basing it on their playoff opponents. The latter might potentially be a better way to do it, but I don’t think is something I can really compile at the moment since I don’t have opponent-specific shots data for Jordan (and, even if I did, doing the analysis this way would take a lot of time).

I will also note the obvious fact that these relative FG%’s on mid-range shots being negative obviously doesn’t mean they were actually bad offense when seen in context. Obviously the league’s overall eFG% includes a whole lot of easier shots, such as transition opportunities, other shots at the rim, open threes, etc. Mid-range shots tend to come in half-court offense—and often near the end of the shot clock—so the FG% on them should be expected to be lower than the overall eFG%. I think it’s safe to say that for all of these guys, their mid-range shots would have a positive relative eFG% if compared only to league average in analogous situations. Relatedly, since this is playoff numbers, we also need to keep in mind that they’re all generally playing above-average defenses, while the league eFG% is obviously not.

[DraymondGold]

Following up on jake’s post — I also thought it might be nice to get some sense of an opponent and era adjustment.

Here I’ll be using opponent’s allowed FG% from two-point range to approximate relative efficiency. So if I say “Jordan: 60 Fg% (+10)”, I mean Jordan scored 60 Fg% from some zone of the floor, his opponents allowed 50% FG% from two-point range, and Jordan outscored that by +10%.

Like Jake said, two point Fg% includes all shots at the rim, paint, and midrange — so the approximate relative efficiency is likely to overrate true relative efficiency at the rim and underrate it in the midrange. The hope is that the extent to which it overrates/underrates rim/midrange efficiency is constant across era, to allow more fair comparison. It’s an approximation, and people are welcome to try something else if they’d prefer.

This measure should incorporate some amount of correction for defenses faced in playoffs here, and for effects of spacing. But like anything, it’s imperfect.

Rim scoring:

Spoiler:

3-year Scoring at the rim:
88-90 Jordan: 70.4 FG% (+22.9) on 8.2 FGA/game
89-91 Jordan: 70.4 FG% (+23.4) on 7.3 FGA/game
90-92 Jordan: 67.7 FG% (+20.4) on 6.9 FGA/game

08-10 LeBron: 68.1 FG% (+21.0) on 7.5 FGA/game
09-11 LeBron: 67.3 FG% (+20.2) on 6.9 FGA/game
10-12 LeBron: 71.4 FG% (+24.1) on 6.9 FGA/game
11-13 LeBron: 71.0 FG% (+24.4) on 7.0 FGA/game

4-year Scoring at the rim:
88-91 Jordan: 70.9 FG% (+23.3) on 7.7 FGA/game
09-12 LeBron: 70.1 FG% (+23.2) on 7.2 FGA/game
10-13 LeBron: 70.7 FG% (+23.8) on 6.9 FGA/game

5-year Scoring at the rim:
87-91 Jordan: 69.7 FG% (+22.4) on 7.8 FGA/game
88-92 Jordan: 68.1 FG% (+20.5) on 7.7 FGA/game
08-12 LeBron: 70.1 FG% (+23.2) on 7.1 FGA/game
09-13 LeBron: 69.9 FG% (+23.3) on 7.2 FGA/game

They look pretty comparable to me, with different players eeking out a small advantage in efficiency or volume in different sample sizes. Peak Jordan looks slightly better than 1st Cavs LeBron (MJ has higher efficiency, comparable volume), but slightly worse than Miami LeBron (lower efficiency, only slightly higher volume).

Context matters here. LeBron faced tougher defenses on average relative to era, who held opponents 0-2% FG% lower than Jordan’s opponents did. This helps Miami LeBron’s efficiency advantage. But Jordan had much less spacing to work with (~4x fewer 3s in the 3-year samples). Thinking Basketball did an analysis of spacing over time on film, and found that there was indeed more spacing by 2006 onward than in 1990. Doing a linear fit to team rim FG% vs 3PA, we might expect it to be ~2-2.5% harder to shoot at the rim in a Jordan’s era than LeBron’s, but rule changes and style changes presumably make this non-linear. The relative efficiency should help correct for both, but as above, it’s an imperfect measure.

In LeBron’s favor, I suspect using 2 point FG% to approximate relative efficiency might boost Jordan’s numbers slightly: they most likely took more long twos in Jordan’s era rather than threes, which presumably drag the 2 point FG% down slightly relative to the later era. In Jordan’s favor, I suspect some of the perceived improvement at the rim from LeBron is from Miami’s added spacing, rather than LeBron actually being better at the rim in 2012/2013 vs 2009/2010. The Heat were ahead of their time by 2012-2013, with some proto-small ball lineups that had LeBron at the four or Bosh at the center, which opened up the floor significantly compared to Jordan’s era. Not all spacing can be captured by just 3PA, so correcting for this added spacing might reduce Miami LeBron’s efficiency advantage.

In any case, they seem surprisingly comparable at the rim statistically. And there’s plenty of statistical and contextual uncertainty to argue for one or the other.

Midrange:

Spoiler:

3-year scoring in midrange:
89-91 Jordan: 47.1 FG% (-1.2) on 11.1 FGA/game
90-92 Jordan: 48.1 FG% (+0.0) on 12.9 FGA/game
91-93 Jordan: 47.2 FG% (-0.1) on 13.2 FGA/game
06-08 Dirk: 44.6 FG% (-2.6) on 8.5 FGA/game
07-09 Dirk: 46.8 FG% (-1.6) on 8.2 FGA/game
08-10 Dirk: 49.5 FG% (-1.1) on 9.0 FGA/game
09-11 Dirk: 50.1 FG% (1.2) on 9.3 FGA/game
10-12 Dirk: 50.5 FG% (1.3) on 9.4 FGA/game

4-year scoring in midrange:
88-91 Jordan: 47.1 FG% (-1.1) on 11.2 FGA/game
90-93 Jordan: 48.0 FG% (-0.1) on 13.6 FGA/game
06-09 Dirk: 46.3 FG% (-0.9) on 8.5 FGA/game
07-10 Dirk: 48.1 FG% (-2.5) on 8.6 FGA/game
08-11 Dirk: 49.2 FG% (+0.0) on 9.1 FGA/game
09-12 Dirk: 50.8 FG% (+2.2) on 9.2 FGA/game

5-year scoring in midrange:
88-92 Jordan: 47.6 FG% (-0.7) on 11.7 FGA/game
91-96 Jordan: 47.5 FG% (-0.3) on 13.5 FGA/game
06-10 Dirk: 47.0 FG% (-1.5) on 8.7 FGA/game
08-11 Dirk: 48.5 FG% (-0.9) on 8.9 FGA/game
09-12 Dirk: 49.9 FG% (+1.1) on 9.1 FGA/game

So Jordan looks outright better (comparable or higher efficiency, higher volume) than pre-2011 Dirk from midrange. Including 2011 on, Dirk gains higher efficiency (+1-2.5% better) but at lower volume (~1.7-3.9 fewer FGA/game).

Again, context matters here. Jordan’s volume advantage is likely aided by his era, which tended towards the midrange over the three. Doing a linear fit to midrange FGA vs 3PA, we’d predict Jordan to have comparable or slightly more volume (~0-1.8 FGA more) than Dirk if they were in the same era. Like with LeBron, using opponent FG% on twos to predict relative efficiency favors Jordan slightly, as the larger midrange volume probably reduces the opponent’s allowed FG% compared to the later era. And like with LeBron, the added spacing in the later era like helps give Dirk an efficiency boost. Doing a linear fit to midrange FG% vs 3PA, we’d predict Jordan to have ~ +0.7 FGA% higher playing in Dirk’s era. Both linear fits here are imperfect, considering we’re fitting across different rulesets and styles, but they give an approximate sense for the kind of correction we might apply.

In sum, peak Jordan seems to be at a statistically similar level to young peak-athleticism LeBron at going to the rim, and slightly better than prime Dirk (less efficient, higher volume than peak Dirk) in the midrange. He’s disappointing in the non-rim paint area, although it’s unclear how much this is driven by a lack of spacing of his era. And peak Jordan’s roughly league average from 3 (very slightly better volume and perhaps efficiency), so he’s not giving anything up there. Like lessthanjake said, we’ve always known Jordan’s total volume and efficiency, so this shouldn’t change people’s overall evaluation much, but it does give a sense for how Jordan was doing his damage, scoring-wise. Fun stuff!

[Jaqua92]

Due to popular demand, here is the playoff midrange data for some legends. MJ looks pretty good!

Player: FG/FGA (%); FGA/game

Michael Jordan: 1040/2279 (45.6%); 12.7
Demar Derozan: 208/559 (37.2%); 8.9
Dirk Nowitzki: 563/1230 (45.8%); 8.5
Kobe Bryant: 759/1854 (40.9%); 8.4
Kevin Durant: 601/1238 (48.5%); 7.3
Chris Paul: 374/743 (50.3%); 5.0
Steve Nash: 232/517 (44.9%); 4.3
Stephen Curry: 225/487 (46.2%); 3.3

I know some people here say that the numbers demythify Jordan as a mid range jump shooting GOAT...

But who else is it? Dirk and KD? They're all in the same ball park, with KD being the most effecient. But Jordan's volume is nearly doubled.

He's the best rid range jump shooter ever. There's no debate.

Disingenuous posting honestly. Typical on here lately.

You're either on 1 side or the other, and there's no in between.

Respectfully, it's an opinion. Let's drop the elitist crap, yeah? It's absurd to make up rules about how people should feel or what conclusions they should draw.

That's disingenuous.

[jjgp111292]

I know some people here say that the numbers demythify Jordan as a mid range jump shooting GOAT...

But who else is it? Dirk and KD? They're all in the same ball park, with KD being the most effecient. But Jordan's volume is nearly doubled.

He's the best rid range jump shooter ever. There's no debate.

Disingenuous posting honestly. Typical on here lately.

You're either on 1 side or the other, and there's no in between.

Respectfully, it's an opinion. Let's drop the elitist crap, yeah? It's absurd to make up rules about how people should feel or what conclusions they should draw.

That's disingenuous.

i don't want to speak for him but I'm pretty sure he was agreeing with you and adding on to your point

[TheGOATRises007]

I know some people here say that the numbers demythify Jordan as a mid range jump shooting GOAT...

But who else is it? Dirk and KD? They're all in the same ball park, with KD being the most effecient. But Jordan's volume is nearly doubled.

He's the best rid range jump shooter ever. There's no debate.

Disingenuous posting honestly. Typical on here lately.

You're either on 1 side or the other, and there's no in between.

Respectfully, it's an opinion. Let's drop the elitist crap, yeah? It's absurd to make up rules about how people should feel or what conclusions they should draw.

That's disingenuous.

I was agreeing with you