- Average Player Efficiency Rating (PER) of the four Hornets frontcourt players (Davis, Anderson, Lopez, Smith): 21.1
- Average PER of the four Hornets backcourt players (Vasquez, Rivers, Roberts, Mason): 11.15
- To reference Mason’s Standards, 21.1 is smack dab between Great and Elite.
- 15 is average.
- 11.15 is between Poor and Bad.
To quote John Hollinger, the creator of PER, the replacement level of a backcourt player is between 10.5 and 11.0. “Replacement level” is a common term in sports analytics for the level of talent readily available to be picked up off the scrap heap. In NBA terms, it’s either a minimum contract guy at the end of the bench or some other team’s bench, a veteran off the free-agent scrap heap or a call-up from the D-League.
12 responses to “Stat of the Week”
There are too many intervening variables that could interfere with that statistic… For example, assume age is a variable and you compared PERS of Kobe and Nash with the Laker with us… So our problem is in the backcourt… How can it be remedied without our superstar… Ever team needs its Westbrook and Durant assuming it competitive…
Davis is a rookie potentially dropping 30 on some nights.. We need leadership which we have in Vasquez who is also maturing… How will he work with EG… In the draft we need the top backcourt shooter… Imagine Thronton on this team… Currently we are gambling on a healthy Eric Gordon…
Stats are great for discussion but what are we comparing…
When thinking positive, Rivers may turn into a starter at PG in a year or 2. Vasquez is ok for now, & makes a good future backup PG. Gordon could come back soon & be a star SG, a major upgrade in our backcourt.
Meanwhile, I’m confused about trade deadlines ( Dec 15 ? ) and options should EG not return. I know we can fix our backcourt in the off season, but surely there is something we could do now besides wait on EG. WWDD? What can he?
Even though I feel it’s easy to get too caught up in advanced stats in terms of player ability, these seem like a great representation of where our team is right now.
I dont understand why we would be looking at an “average” PER rating on a group of players, when it is actually an “individual” statistic meter.
When looking at PER, it should be looked on a per player basis, specifically when referencing “replacement level” backcourt players. Those numbers can be rather skewed when giving an average, which can lead to unfairly judging an individual player.
I would like to see individual PER’s. This would provide the true telling statistics. Well, if we want to be judgemental, that is.
Davis: 25.8
Smith: 21.6
Anderson: 20.4
Roberts: 16.8
Lopez: 16.5
Vasquez: 12.5
Mason: 9.9
Rivers: 5.4
I actually would not be opposed to seeing what a Vasquez, Aminu, Davis, Anderson, Smith lineup could do for the first 8 minutes of a game. To be totally honest, I think it would be lethal on defense, and above average enough of offense to make it worthwhile.
Would you like it enough to click here? http://www.basketball-reference.com/teams/NOH/2013.html
That’s all there is to it.
The average of such a stat may not have any mathematical or predictive weight, and it fails to completely characterize the play of the groups. It does, however, give a succinct and relevant description of the respective levels of play to this point of the two groups.
Two things I love about this analysis;
– It points out we don’t need to add a superstar PG. A average starting PG would be a big improvement, as would a quality back-up SG.
– It confirms what we all know about the strength of this team being its frontcourt players and the backcourt, especially without Gordon, being a weakness.
Here is how the Hornet’s remaining players (SFs) stack up:
Aminu: 16.39
Miller 8.69
Thomas 4.79
I really don’t think this a valid measure with Eric Gordon… The backcourt will transition and provide coach a reliable scorer… All statistical methods require reliability and validity… Is the method true when replicated…
What does that even mean? What would be a sufficient replication for you?
Statisticians seek reliability and validity of their methods… Certainly you need a league wide measure but comparing PER for an established backcourt s rookies seems a bit unfair…
What method? This is clearly a descriptive situation, not a modeling, predictive, or inferential situation.
Again, what would be a sufficient replication for you?