What are your thoughts on the IBM Award given out from 1984-2002

[Wally450]

The IBM Award was an award given out to National Basketball Association players from 1984 to 2002. The award was sponsored and calculated by technology company IBM and was determined by a computer formula, which measured a player's statistical contribution to his team. The player with the best contribution to his team in the league received the award.

The award did seem to favor front court players more.

Seems like something they could bring bag, especially with how analytically driven the league is now.

https://en.wikipedia.org/wiki/IBM_Award


Here are the past winners of the award...

1984 - Magic Johnson, Los Angeles

1985 - Michael Jordan, Chicago

1986 - Charles Barkley, Philadelphia

1987 - Charles Barkley, Philadelphia

1988 - Charles Barkley, Philadelphia

1989 - Michael Jordan, Chicago

1990 - David Robinson, San Antonio

1991 - David Robinson, San Antonio

1992 - Dennis Rodman, Detroit

1993 - Hakeem Olajuwon, Houston

1994 - David Robinson, San Antonio

1995 - David Robinson, San Antonio

1996 - David Robinson, San Antonio

1997 - Grant Hill, Detroit

1998 - Karl Malone, Utah

1999 - Dikembe Mutombo, Atlanta

2000 - Shaquille O'Neal, L.A. Lakers

2001 - Shaquille O'Neal, L.A. Lakers

2002 - Tim Duncan, San Antonio


I'm sure someone could use the formula to calculate who the winners would be from 2003-present.

[Soundwave]

Feels like it's very heavily weighed towards rebounding ... I mean I like Dennis Rodman, but no chance in hell he was even one of the top 5 players in the NBA in 1992.

[warriorfan]

Damn I’ve never heard of that. Pretty interesting. Looks like a precursor to PER. Surprised to see they were already starting to think of this stuff 40 years ago.

[Wally450]

Damn I’ve never heard of that. Pretty interesting. Looks like a precursor to PER. Surprised to see they were already starting to think of this stuff 40 years ago.

Same. Seems like a pretty cool award. They'd definitely have to change the formula to make it more fair and balanced for backcourt players to win. Too heavily favored towards front court guys.

[tpols]

The main gripe I have with this is playoff performance AKA showing up when it matters. The data here is long term derived for the regular season mostly. It doesn't detail who played the best when championships were on the line. The best examples are '95 D-Rob vs Hakeem, '98 Malone vs Jordan, and 1984 "Tragic" Johnson.

[iamgine]

Formula is too simple. I'd hope if replicated today we'd have a much more advanced formula.

[90sgoat]

Formula is too simple. I'd hope if replicated today we'd have a much more advanced formula.

I could easily make a better formula just by using box score stats and correlating them to wins in regular season and Michael Jordan.

[AussieSteve]

It's basically team wins multiplied a players share of the teams counting stats.

Now that we have play by play data, this formula could be adjusted by dividing all counting sets into three buckets:
1. Clutch (within 5 points, less than 5 mins remaining)
2. Blow out (margin >15)
3. Other

Clutch weighted higher than other. Other weighted higher than blowout.

Add weighted scores and normalise, to arrive at the final player score.