Correcting the ABA-NBA “equality” myth

Since reading Bill Simmons views on the history of the ABA and how the ABA favorably compared to the “too white” NBA when both were in existence (all of this, of course, written in his Book of Basketball) I wondered “How did the NBA actually compare to the ABA?” and “Were the NBA and ABA ever truly equal?”

A lot of people — not just Simmons — believe that in the last waning years of its existence the red-white-and-blue ball of the ABA caught up to, and perhaps even dribble drove right past, the traditional NBA orange ball we all grew up with and love.  Was that possible?  Was the NBA the inferior league?

Those that believe that point as evidence to the ABA’s growing dominance over the NBA when the rival Association’s faced off during the 1971-75 exhibition seasons.   The two leagues played a total of 155 times, yielding increasingly positive results for the upstart as the series went on.

Sure the ABA got their asses kicked pretty badly in the first couple of years, but the ABA and its supporters could legitimately brag that they indeed won the majority of the exhibitions in each of the last three years and actually won the overall series 79 games to 75.

Does that mean the ABA was actually better than the senior NBA?  Not so fast.

You can’t compare two leagues by simply pointing to a series of games and saying the bare results hold the comparative truth.  The truth does not arrive until you adjust the results so that one’s apples are being compared to the other’s apples, not its Buffalo Braves.

What nearly all who point to the ABA-NBA exhibition results as proof of ABA equality fail to mention is the location of the games and the matchups.  Since the NBA did not want to “legitimize” the ABA with a lot of games in NBA arenas, an overwhelming majority of the games were played in ABA gyms and were therefore officiated by ABA refs.  Thus to have any comparative value whatsoever the scores must first be adjusted to account for homecourt advantage.

Moreover, to get a true feel for the relative strength of each Association, you have to “neutralize” the two teams playing so that each game can stand as a reliable comparison.  The games weren’t match ups of relative equals like the Big Ten-ACC challenge in college basketball.  The NBA often used bottom feeder clubs in the exhibitions (the Lakers didn’t play in a single NBA-ABA game) while the ABA kept its bottom feeders at home and instead sent its marquee clubs.

So you must account for those disparities when considering the exhibition results and I below I did that.  And the new outcomes I came up with put the lie to any notion of ABA equality until the very last days of ABA basketball (1976 the ABA drew even after it contracted itself down to its “cream” six teams).

Here’s the method  I used.

Utilizing Country Club Scoring to Settle the Issue

To give a clear accounting of each Association’s strength with relation to its rival, I just applied golfers logic to all of the exhibition results.

When two golfers are of uneven strength the pair “handicap”  the score to create the illusion of even competition.

Similarly, I used Basketball-Reference’s “Simple Rating System” to adjust the exhibition outcomes so that, to the extent possible, every single game matched a fictitious “Average ABA team” versus a fictitious “Average NBA team” on even footing.  (So for example, in 1971-72 the Bucks were something like +10.0 points above NBA average on the Simple Rating System, so for every Bucks exhibition the Bucks had to give ten points and so on).

After that I gave +3.4 points, the standard Vegas homecourt adjustment, to the visiting team.  If the game was a “semi-home” game I gave +1.0 point to the visitor, and for seemingly “neutral” site games I gave nothing.  Given the passion that reports say the games brought to ABA arenas, that more than probably understates the value of homecourt advantage, but its close enough.  (Note that I threw out several games to keep the comparison legitimate.  For instance I threw out all of the Atlanta Hawks games in which they suited up Dr. J, and all of the Virginia Squires from that exhibition season that did not feature the Doctor).

The “New” Results

Here are my new adjusted results season-by-season:

1971-1975ABA-NBA Exhibition

Handicapped Results









ABAPoint Spread



NBARelative Winning% ABARelative Winning%
1971-72 (21 games) 14 7 +5.1 -5.1 .648%(13.6 wins) .351% (7.4 wins)
1972-73 (30 games) 22 8 +8.4 -8.4 .745%(22.3 wins) .255% (7.7 wins)
1973-74 (24 games) 9 15 +4.6 -4.6 .633%(15.2 wins) .367% (8.8 wins)
1974-75 (22 games) 7 15 +4.7 -4.7 .637%(14.0 wins) .363% (8.0 wins)
1975-76 (48 games) 18 30 +0.5 -0.5 .513% (24.6 wins) .487% (23.4 wins)
OVERALL (145 games) 70 75 +4.1 -4.1 .620% (89.9 wins) .380% (55.1 wins)

ABA was about 80% of the NBA

As  you can see, the adjusted results paint a different picture than the one propagated by ABA enthusiasts, one of more prolonged and consistent NBA dominance.   At no point prior to 1976 was the ABA anywhere near the NBA’s equal. (note:  “Relative Winning %s” for each Association were determined according to each season’s adjusted average point differential — which I for some reason referred to as “point spread”… sorry — and the number of games a fictitious team would likely win with such an average point differential — using this formula).

The results, I think, need to be read in two year increments.

It seems in the first two exhibition seasons the NBA thoroughly dominated, both actually and in adjusted terms.  Then in the next two seasons the ABA leveraged the matchups and locations but nevertheless the adjusted point spreads — which ended up being remarkably similar despite the sundry adjustments — painted a continuing picture of dominance, albeit adjusted dominance.

Then in 1975-76, the year before the merger, the ABA placed itself basically on equal footing with the NBA.  Based upon the results I would go so far as to say all six of the ABA teams that were then on-going concerns were in fact NBA worthy teams — not just the “ABA Four” that were ultimately allowed to merge and that continue to this day.  The Kentucky Colonels and the Spirits of St. Louis could have been very respectable NBA franchises, and their fan bases were deprived of that chance (although the owners of the Spirit cut one of the most famous and best “buy-out” deals in the history of contract law).

What happened in Year Two?

The one result that makes no logical sense is the second exhibition season.  Even though it was the second longest exhibition schedule, the results were out of whack with the rest of the series.  The NBA simply whooped up on the ABA.  In fact the adjusted result would have been even worse had I not thrown out five NBA wins that were tainted by Dr. J’s weird decision to sign with — and play two exhibitions for — an Atlanta Hawks team that had no write to sign him — notwithstanding that he was still under contract to the ABA Squires (your Milwaukee Bucks owned his NBA rights.  How on Earth his agent advised him that signing with the Hawks would somehow have a constructive result no one has ever explained to my satisfaction.  If I were the Bucks or the Squires I would have hit the Hawks with a tortious interference action, not just an injunction.  They had no colorable right to Julius Erving… they just decided to sign him and play him!)

I really don’t know what happened or why the NBA delivered such a beatdown.  I think it may be an aberration.

Or perhaps it has something to do with the strength or waning strength of the homecourt advantage enjoyed by the ABA that season.  At any rate, if you adjust the 1973-74 results and put them in line with the (+5.1) point advantage held by the average NBA team over the average ABA team in the preceding season, then the final result would favor the NBA by about +3.7 relative point advantage.  That would mean over the entire existence of the ABA we could conclude that the average NBA team was about +3.7 points better than the average ABA team.

No big deal?  Actually that’s pretty substantial.  That translates into a winning percentage advantage that would be about +20.2% on average.

This I think is the correct ABA equivalency number — somewhere in the neighborhood of 20% less than the NBA.  I think that way because it would comport exactly with another study I did of the Win Score production of 25 random ABA-NBA performers which I outline below.

ABA-NBA Win Score comparison

I used Professor Berri’s basketball analytic known as Win Score to compare how 25 of the biggest ABA stars did when they played in the NBA.  The group of 25 I came up with either were named to the “All-Time ABA team”, are well known, or were mentioned by Bill Simmons as prominent ABA players in his Book of Basketball.

ABA-NBAPerformance Comparisons

using Win Score / 48

Julius Erving, F 18.39 14.08
Rick Barry, F/G 10.65 9.15
Billy Cunningham, F 15.06 12.73
Spencer Haywood, C/F 21.30 13.18
Connie Hawkins, F/C 17.89 12.86
David Thompson, G/F 11.23 9.78
Zelmo Beatty, C 17.40 14.02
George McGinnis, PF 16.18 12.92
Artis Gilmore, C 21.21 16.78
Charlie Scott, G 4.49 2.12
Dan Issel, C/F 12.94 14.06
Bobby Jones, F 17.28 13.41
Billy Knight, SG 12.21 8.72
Maurice Lucas, PF 11.64 11.66
George Gervin, SG 10.64 8.49
Jim Chones, PF 14.72 10.15
Swen Nater, C 19.45 15.87
Super John Williamson, SG 2.13 1.69
Ron Boone, G 5.98 4.17

Marvin Barnes, F

15.56 7.92

Caldwell Jones, F/C

16.29 11.37

ML Carr, F/G

10.21 8.04
Larry Kenon, F 12.45 11.69
Don Buse, G 10.18 8.11

Tom Owens, C

13.85 10.94

20.2% Rule

As you can see, nearly every one of the 25, save for Dan Issel, saw a decline in his productivity when he brought his game to the NBA, with the average decline being 20.2%, and the median being 20.8%.

So everything seems to come up “20% reduction” when evaluating the strength of the ABA visavis the NBA.  Even Dr. J took a 20% haircut when he made his famed jump to the senior circuit.

But none of this should be read to diminish the achievements of the ABA.  Frankly, I’m stunned at what they were able to accomplish, given the fact that they were somehow able to run a respectable “Shadow NBA”, paying top dollar for talent, when they had little attendance money and no television money to draw upon.  Frankly, they had some balls to give it a go.

And give it a go they did.  And bare this in mind Bucks fans.  Without the ABA, there’s probably no Milwaukee Bucks.  The Bucks improbable bid for franchise came as part of the “panic response” sudden expansionist movement the NBA undertook after the ABA tugged at its contented tail.

So thank you ABA.  But just don’t try to say you were equal to the NBA.  You weren’t.  Not till closing time you weren’t.

Footnote: I just found an eerily similar but much better written piece comparing the NFL to the AFL using virtually the same techniques and some of the same kind of data. (my writings getting so sloppy lately.  Like I told you, I have this weird habit of mirroring the linguistic patterns of the author I’m reading at the moment and at the moment I’m reading “Mr. Everything’s a Digression”.  I gotta throw that fucking book away now. But it has given me a lot of ideas for posts that don’t begin with “The Bucks STILL can’t shoot” so I owe it that much.)


Tags: , , , , , , , , ,

4 Responses to “Correcting the ABA-NBA “equality” myth”

  1. Greg Magarian Says:


    This is a really interesting post; I was just glancing around the site and found it. I want to raise two quibbles, the first of which I think is meaningful; the second I’m not so sure about.

    First — Your analysis of the exhibitions is based on adjusting for distorting factors. Fair enough. But by the same basic logic, shouldn’t you adjust the player comparisons for natural declines based on age? I don’t know the history of pro basketball especially well, but I imagine your player comparisons are measuring a lot of 20-something ABA seasons against a fair number of 30-something NBA seasons. From reading Simmons I assume that some of those players also hit serious drug trouble in their NBA years, though that would be harder to account for.

    Second — In making your adjustments for the exhibition games, you’re assuming that the teams’ play translates in a linear fashion to different circumstances. But isn’t it true that teams who are winning blowouts preserve some of their effort, especially in exhibitions, by playing lesser players or just slowing down the pace? Likewise, don’t refs behave differently under different circumstances: What I’m getting at is, if your adjustments are converting, say, a 110-90 ABA win into a 102-100 NBA win, wouldn’t we have to assume that the ABA would compensate for the squeeze by playing more of its best players down the stretch, pushing the tempo, etc.? Wouldn’t the refs do more to help the home teams? I’m just pulling that example out of my butt, but it seems to me that the adjustments you’re making (different players, different refs) would affect the underlying data in ways you don’t (and probably can’t) account for.

    Again, I really like the piece. Thanks for doing all the work.

  2. Greg Magarian Says:

    You did the analysis; I thought it was something you would be interested in thinking / talking more about. My mistake. Sorry to have wasted your time; I won’t do so again.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: