PG usage,shots & efficiency: season&playoffs, young & older
Posted: Thu May 05, 2011 5:42 pm
I write this elsewhere with main regard to Westbrook; but since it picked up on issues raised in the Rose thread, I briefly adapted much of the material to here.
Scott Brooks quote:
“He [Westbrook] is not a traditional point guard but who is now? I mean, Derrick Rose is the MVP of the league and he leads their team in scoring. He takes the most shots. He’s good. That’s how they play and that’s how they win, but that’s point guards now. Point guards are like that now.”
Westbrook and Rose are much more outliers than the general playoff pattern.
Over the last 5 years, only about 4-5 PGs per year over 25 minutes per game in the playoffs have averaged over 25% usage. So that is only about 25-30% of playoff starters who are over 25% usage and 70-75% who weren’t. Westbrook is at 35% usage.
1 team with a PG over 25% usage won a championship during that 5 year period (Parker at 29% usage) and 4 teams won with a PG under 25% usage, with an average starting PG usage of 19%. 19%, or barely half of Westbrook’s usage. That is how playoff PGs are on average and on recent champions.
Regular season and the playoffs show different trends.
PGs in 2003-4 regular season over 24% usage and 25 minutes per game: 5
# of such PGs in 2004 playoffs: 3
# of such PGs in 2010-11 regular season: 11
# of such PGs in 2011 playoffs: 5
So in the playoffs, the # over 24% usage and 25 minutes per game increased from 3 to 5 over 7 years. Is that a meaningful trend? It is still only a bit over 30% of playoff PGs.
# to get to the second round: 2
How many will make the conference finals?
Will the champion PG be over or under 24% this time?
Magic, the most successful modern PG title winner, never had playoff usage above 25% and that of course gives a pretty good amount of weight to his passing.
Parker’s playoff usage was 29% for his last title though he averaged just a bit over 26% for the 3 championships he was a part of.
Average for playoff usage of playoff PGs (including tweeners like Wade) over 25 minutes per game in 2004: 21.66%.
Average for playoff usage of playoff PGs over 25 minutes per game in 2011 so far: 20.30%.
A 6.3 percentage decrease.
There were only 4 over 15 FGAs / game this playoffs.
In 2004? The same, just 4. The champ PG that season, Billups, one of the best shooting PGs of all-time, on a team sometimes speculated to be a decent comparison or model for the Thunder, took just 12 FGAs per game.
That’s what the number say for the change in the recent period.
In 2004 the big minute PGs played an average of 15.2% more minutes per game in the playoffs. So adjusting to FGAs per minute, in the 2011 playoffs the average was almost exactly the same as 2004. So no decrease on a per minute basis. But no increase per minute on average either.
FTs were up a small fraction per game in 2011 compared to 2004 but assists were down by a larger fraction.
Shooting efficiency in 2011 for these guards was down across the board- 2 pt, 3 pt and from the line. So average usage went down, and average FGAs stayed the same. There were few cases of high usage or high shots per game then or now. And shooting quality went down. So how smart is a team strategy of an extremely high usage PG including high FGAs and currently below average FG%?
I went ahead and ran the regular season numbers for the big minute PG comparison. There was no increase in average shots per minutes from 2004 to 2011 for the regular season for these players. There was the tiniest fraction of a decline. There was the tiniest of increase in usage. A few hundredths of a percent. The impression of increase or a notable one for these stats, at least in recent years and for the player set selected, appears to be illusion. At both times only 5 PGs took 15 shots per game. The percentage that did so though actually went down significantly as it appears more PGs and tweener guards got 25+ minutes. So what people may have picked on was more PGs and tweener guards playing more total minutes, but few minutes per game per player, and not having higher personal usage or taking more shots per minute. In the regular season they shot better in 2011 than in 2004 from all distances so that may have fueled the increase in qualifying players and probably total minutes. But as noted earlier, this group shot worse in the playoffs in 2011 than in 2004 across the board. So this trend does not seem an especially favorable one to teams.
One more important twist to my data. In the 2011 regular season, the big minute PGs 23 years old and under at the start for the season had a 10% higher usage than those older and shot 26% more often. They shot near identical FG%s. In the playoffs the young ones had 21% higher usage and shot 43% more than the older ones.
There is a trend of big minute PGs having higher usage and shooting more often… but it is among those under 23 years old and not the entire big minute PG population. The average trend might extend beyond 23 years old but 24 no longer seems “young” to me for this. These young guys on average have shot better in the playoffs than their older counterparts especially from 3 point land… but not Russell. His FG%s are real close to the older guys.
Of the 25 PGs who started 60+ games (as a simple average, not minutes weighted) in the regular season, the 11 who had usage above the average had an average regular season Offensive Rating about 1 point below those with usage under the mean, though about 1.5 higher in the playoffs.
Of those 11 with higher than average usage, 5 made the playoffs but only 2 remain in it (Russell and Derrick). Of the 19 with lower than usage. 10 made the playoffs (Hinrich is missing from my dataset for lack of 60 starts for the Hawks) and 5 remain. Similar fractions advancing. So it is not “pivotal” on average. but it still “could be” in specific cases. Westbrook has the lowest playoff Offensive Rating of any PG still playing.
PGs under the mean on usage were still twice as common as over the mean. Some of those over the mean usage were way over. Of those over 25% usage, the regular performance was about the same but the playoff Offensive Rating is about 2 points less than for those below the mean usage.
Edthesportsfan: “OKC is 15-7 (68%) in the regular and post-season combined when Westbrook outshoots Durant…versus 44-22 (66%) when he doesn’t. That’s 22 of a possible 81 regular season games played. In other words, Westbrook outshot Durant in 27% of all of OKC games this year. OKC is 13-5 (72%) in the regular season, and 7-5 (58%) versus playoff teams. In 4 playoff games versus Denver/Memphis, OKC was 2-2. 2-1 versus Denver, and 0-1 versus Memphis.”
2-2 in 4 playoff games is a small sample and not dramatically bad for the playoffs, but there may be little room for a problem, this possible “problem” or any problem in a close series.
Scott Brooks quote:
“He [Westbrook] is not a traditional point guard but who is now? I mean, Derrick Rose is the MVP of the league and he leads their team in scoring. He takes the most shots. He’s good. That’s how they play and that’s how they win, but that’s point guards now. Point guards are like that now.”
Westbrook and Rose are much more outliers than the general playoff pattern.
Over the last 5 years, only about 4-5 PGs per year over 25 minutes per game in the playoffs have averaged over 25% usage. So that is only about 25-30% of playoff starters who are over 25% usage and 70-75% who weren’t. Westbrook is at 35% usage.
1 team with a PG over 25% usage won a championship during that 5 year period (Parker at 29% usage) and 4 teams won with a PG under 25% usage, with an average starting PG usage of 19%. 19%, or barely half of Westbrook’s usage. That is how playoff PGs are on average and on recent champions.
Regular season and the playoffs show different trends.
PGs in 2003-4 regular season over 24% usage and 25 minutes per game: 5
# of such PGs in 2004 playoffs: 3
# of such PGs in 2010-11 regular season: 11
# of such PGs in 2011 playoffs: 5
So in the playoffs, the # over 24% usage and 25 minutes per game increased from 3 to 5 over 7 years. Is that a meaningful trend? It is still only a bit over 30% of playoff PGs.
# to get to the second round: 2
How many will make the conference finals?
Will the champion PG be over or under 24% this time?
Magic, the most successful modern PG title winner, never had playoff usage above 25% and that of course gives a pretty good amount of weight to his passing.
Parker’s playoff usage was 29% for his last title though he averaged just a bit over 26% for the 3 championships he was a part of.
Average for playoff usage of playoff PGs (including tweeners like Wade) over 25 minutes per game in 2004: 21.66%.
Average for playoff usage of playoff PGs over 25 minutes per game in 2011 so far: 20.30%.
A 6.3 percentage decrease.
There were only 4 over 15 FGAs / game this playoffs.
In 2004? The same, just 4. The champ PG that season, Billups, one of the best shooting PGs of all-time, on a team sometimes speculated to be a decent comparison or model for the Thunder, took just 12 FGAs per game.
That’s what the number say for the change in the recent period.
In 2004 the big minute PGs played an average of 15.2% more minutes per game in the playoffs. So adjusting to FGAs per minute, in the 2011 playoffs the average was almost exactly the same as 2004. So no decrease on a per minute basis. But no increase per minute on average either.
FTs were up a small fraction per game in 2011 compared to 2004 but assists were down by a larger fraction.
Shooting efficiency in 2011 for these guards was down across the board- 2 pt, 3 pt and from the line. So average usage went down, and average FGAs stayed the same. There were few cases of high usage or high shots per game then or now. And shooting quality went down. So how smart is a team strategy of an extremely high usage PG including high FGAs and currently below average FG%?
I went ahead and ran the regular season numbers for the big minute PG comparison. There was no increase in average shots per minutes from 2004 to 2011 for the regular season for these players. There was the tiniest fraction of a decline. There was the tiniest of increase in usage. A few hundredths of a percent. The impression of increase or a notable one for these stats, at least in recent years and for the player set selected, appears to be illusion. At both times only 5 PGs took 15 shots per game. The percentage that did so though actually went down significantly as it appears more PGs and tweener guards got 25+ minutes. So what people may have picked on was more PGs and tweener guards playing more total minutes, but few minutes per game per player, and not having higher personal usage or taking more shots per minute. In the regular season they shot better in 2011 than in 2004 from all distances so that may have fueled the increase in qualifying players and probably total minutes. But as noted earlier, this group shot worse in the playoffs in 2011 than in 2004 across the board. So this trend does not seem an especially favorable one to teams.
One more important twist to my data. In the 2011 regular season, the big minute PGs 23 years old and under at the start for the season had a 10% higher usage than those older and shot 26% more often. They shot near identical FG%s. In the playoffs the young ones had 21% higher usage and shot 43% more than the older ones.
There is a trend of big minute PGs having higher usage and shooting more often… but it is among those under 23 years old and not the entire big minute PG population. The average trend might extend beyond 23 years old but 24 no longer seems “young” to me for this. These young guys on average have shot better in the playoffs than their older counterparts especially from 3 point land… but not Russell. His FG%s are real close to the older guys.
Of the 25 PGs who started 60+ games (as a simple average, not minutes weighted) in the regular season, the 11 who had usage above the average had an average regular season Offensive Rating about 1 point below those with usage under the mean, though about 1.5 higher in the playoffs.
Of those 11 with higher than average usage, 5 made the playoffs but only 2 remain in it (Russell and Derrick). Of the 19 with lower than usage. 10 made the playoffs (Hinrich is missing from my dataset for lack of 60 starts for the Hawks) and 5 remain. Similar fractions advancing. So it is not “pivotal” on average. but it still “could be” in specific cases. Westbrook has the lowest playoff Offensive Rating of any PG still playing.
PGs under the mean on usage were still twice as common as over the mean. Some of those over the mean usage were way over. Of those over 25% usage, the regular performance was about the same but the playoff Offensive Rating is about 2 points less than for those below the mean usage.
Edthesportsfan: “OKC is 15-7 (68%) in the regular and post-season combined when Westbrook outshoots Durant…versus 44-22 (66%) when he doesn’t. That’s 22 of a possible 81 regular season games played. In other words, Westbrook outshot Durant in 27% of all of OKC games this year. OKC is 13-5 (72%) in the regular season, and 7-5 (58%) versus playoff teams. In 4 playoff games versus Denver/Memphis, OKC was 2-2. 2-1 versus Denver, and 0-1 versus Memphis.”
2-2 in 4 playoff games is a small sample and not dramatically bad for the playoffs, but there may be little room for a problem, this possible “problem” or any problem in a close series.