Box Score
BasketballAn objective approach to basketball analysis through the simplicity of the "Box Score" TM
Back to Definitions
Back to Home Page
Back to Previous Page
GPM - Game Pace Multiplier
The Game Pace Multiplier (GPM) is used to normalize both player and team statistics so that all teams and players can be compared if they were to play at the same pace of play. For example, in the 2004-05 season the Phoenix Suns played most of their games shooting the ball very early in each possession. This is strikingly evident when looking at the number of shots, rebounds, assists and every other statistic in both the Suns stats and their opponent's stats for the season.
TEAM GPM 04-05 ATL 1.002 BOS 0.993 CHI 0.992 CHR 0.984 CLE 1.005 DAL 0.985 DEN 0.991 DET 1.024 GSW 0.957 HOU 1.029 IND 1.051 LAC 1.011 LAL 0.982 MEM 1.045 MIA 1.011 MIL 0.998 MIN 0.998 NJN 1.036 NOH 1.023 NYK 1.007 ORL 0.973 PHI 0.980 PHO 0.926 POR 0.998 SAC 0.958 SAS 1.027 SEA 1.022 TOR 0.991 UTA 1.053 WAS 0.973
2) The team with the ball gives up the ball to the opponent: a Turnover (TO)
3) A player is fouled by the defensive ("PF")
Note: An offensive foul is counted as a turnover.
What we did is compile the total "events" -- the number of FGA, TO, and PF for each team as well as their opponents. We then determined the league average and divided each teams event value by the league average. By glancing at the column to the left, you can see the "percent" increase and decrease of game pace that directly affects a player's ability to gain statistics. The larger the number, the larger the multiplier required to normalize a slow paced team to an average paced team. Memphis had a GPM of 1.045, which means that every player lost "4.5 percent" of the ability to earn statistics compared to the league average. Also, the Phoenix Suns players had "great years" because they were a great team AND they played at a pace which enabled thier players to gain extra statistics: their GPM was 0.926, which is number needed to "normalize" their statistics. One can say from a comparison standpoint that 7.4% (1.000 minus 0.926 = 0.074 = 7.4%) of their stats were inflated due to their pace.
So what does this all mean?
Basically, all you need to do in order to compare two players who played on teams with two very different styles of game, is "normalize" their stats using the multipliers to the left.
Example: Compare the scoring production of Joe Johnson (17.1 points per game) of Phoenix (PHX .926 GPM) to that of Andrei Kirilenko (15.6 points per game) of Utah (UTA 1.053 GPM).
Johnson = 15.83 PPG on the league "normalized" pace (multiply scoring by GPM)
Kirilenko = 16.43 PPG on the league "normalized" pace (multiply scoring by GPM)
Our method does not even consider the fact that Johnson played
39.5 minutes per game and Kirilenko 32.9 MPG.
Kirilenko = 17.48 PPG (normalized using GPM and 35 minute game)
Johnson = 14.03 PPG (normalized using GPM and 35 minute game)
Can we take it another step further? Let's assume Kirilenko and Johnson swapped teams (there are many
things to consider in this situation, especially the question as to how the GPM's would be affected. But
game pace is the philosophy of the coach, and swapping these players would only by a small % of floor minutes).
Assuming Kirilenko and Johnson swapped teams, their predicted scoring averages would have been
Kirilenko 22.66 Points per game (in 39.5 minutes and Phoenix's pace)
Johnson 11.10 Points per game (in 32.9 minutes and Utah's pace)
This is such a contrast to the raw stats of Johnson scoring 17.1 PPG vs Kirilenko's 15.6 before any analysis!
Kirilenko, from just looking at points per game (in the papers), would look to be "twice" the scorer.
Back to Definitions
Back to Home Page
Back to Previous Page
All contents copyright © 2006, 2007, Greg W. Lowe. All rights reserved.