NBA-ABA Conversion Charts
year min sco reb ast SS#
1968 .38 .64 .80 .90 782
1969 .73 .72 .85 .90 125
1970 .46 .80 .88 .90 611
1971 .74 .86 .90 .95 365
1972 .91 .90 .92 1.0 529
1973 .97 .91 .92 1.0 316
1974 .61 .92 .94 1.0 347
1975 .87 .92 .95 1.0 358
1976 .80 .92 .96 1.0 3425
The Minutes column is (NBA Min)/(ABA Min) -- averaged over the
sample for that year. In 1968, several players' rates are compared
to their last previous NBA season, which in some cases were 2-4
years prior.
Sco, Reb, and Ast are actually derived from averages of several
estimates: straight average, minutes-weighted, 3-year average, and
3-year/weighted by minutes. Then just smoothed over. 'Min' are not
smoothed, merely averaged.
Assists are so jumpy, I just crudely estimated them.
SS# is the sample size in player-games considered. Most years
(3-400 player-games) are equivalent to only 4-5 full player-seasons.
(The small 1969 sample is largely one guy, Rick Barry.)
The year of reference is the ABA season played. Whether Player X was in the NBA in 1971 and the ABA in '72; or in the ABA in '72 and NBA in '73; or in both leagues in '72; his numbers are averaged into the 1972 lot. Provided he had significant minutes in both appearances.
http://sports.groups.yahoo.com/group/APBR/message/16704I've fired up the old growler and scribbled the conversion rates,
for your perusal. There's a lot of uncertainty, and the numbers go
in fits and starts; but there's a trend or two.
year min sco reb ast SS#
1968 .38 .64 .80 .90 782
1969 .73 .72 .85 .90 125
1970 .46 .80 .88 .90 611
1971 .74 .86 .90 .95 365
1972 .91 .90 .92 1.0 529
1973 .97 .91 .92 1.0 316
1974 .61 .92 .94 1.0 347
1975 .87 .92 .95 1.0 358
1976 .80 .92 .96 1.0 3425
The Minutes column is (NBA Min)/(ABA Min) -- averaged over the
sample for that year. In 1968, several players' rates are compared
to their last previous NBA season, which in some cases were 2-4
years prior.
Sco, Reb, and Ast are actually derived from averages of several
estimates: straight average, minutes-weighted, 3-year average, and
3-year/weighted by minutes. Then just smoothed over.
Assists are so jumpy, I just crudely estimated them.
SS# is the sample size in player-games considered. (The small 1969
sample is largely one guy: Rick Barry.)
Since 1970 was the biggest sample year between the ABA's beginning
and end; and it's the last year I'm calling it "minor-league", lets
examine the sample of comers/goers who were in the ABA that year:
Haywood, Melchionni, Barnhill, Raymond, Dove, Orms, Niemann, Olsen,
Hamilton, Warlick, Kron, Workman.
If you've only heard of the first couple, don't feel lost; these are
in descending order of ABA minutes for 1970, and most were not
impact players.
Of this dirty dozen, only Tommy Kron got fewer ABA minutes than he
had in the NBA.
Of the 12, only Hamilton and Warlick (bit players, too) had a better
effective shooting % in the NBA.
Those same 2 were the only ones with better scoring rates in the NBA.
Melchionni and Barnhill had better rebound rates in the NBA, along
with Kron. These guys were guards with few rebounds in either
league.
Only Warlick had a better assist rate in the NBA.
In all, their minutes more than doubled in the ABA (or halved, if
they were going the other way). Eff% averaged 8% higher in the
ABA. Scoring rates were only 74% as good in the NBA; Rebounding
87%; Assists 68%.
These are per-minute rates. Compounded by the doubling of minutes,
the actual production was more than twice as great in the ABA.
From this type of evidence (player stats standardized year-by-year),
I can't agree that ABA competition was equal to NBA competition, in
1970.
But if you scroll back up to the chart, you'll see within 2 years
the competative difference had been cut in half.
And the 1976 figures should be strengthened for the ABA. Last year
we pretty much concurred that the dual effects of contraction (to
1977 NBA) and joining new teams might account for the majority of
the difference.