Player Efficiency Ratio (PER)

[VidNick]

A widely used basketball statistic (Player efficiency rating). Takes into account everything on a scoresheet. Does not account for the value of a shut-down defender.

I calculated the PER of several of our favorite players. Checked my work, but error is possible. Paige: 23.2, ONO: 19.6, Williams: 14, Edwards: 13.8, Caitlan: 26.3. EDD scored 28.6 in the only college year I found. Basically same as her pro stat.

A frame of reference is WNBA all-time leaders: Cooper: 28.72, EDD: 28.66, Jackson: 27.9, (8) Stewie: 25.3, (11) Maya 24.6, (15) DT: 23.1, (20) Charles: 21.1

 

[Big Mick]

For the St Johns game, here are the efficiency ratings:

Actual Minutes on the floor: ..... Extrapolated to 40 minutes:

.....

 

[huskyd]

I am not a fan of stats because they are malleable, ie, they can prove pretty much whatever you'd like.
Wikipedia: "PER takes into account accomplishments, such as field goals, free throws, 3-pointers, assists, rebounds, blocks and steals, and negative results, such as missed shots, turnovers and personal fouls." But what's the relationship to PER numbers and game outcomes? Inferred is that the higher the PER the more the player contributes to winning. In a low PER performance, say, with Paige against TN, what's the value of her making a game-winning (or game-securing) shot at the end? Bueckers also scored the last 13 points v. SC. Can we safely throw out everything she did up to that point? Actually we do because she won the game for UConn. That's all that mattered in the end. Stats don't determine legends (though they may confirm a player's greatness).
Finally, who would you like to have take the last-minute shot that determines a game's outcome? Think Geno looks at the PER to decide?

 

[UcMiami]

One of the stats that I see in International play that I would like to see are fouls drawn which can be instrumental in game outcomes.

The other International scoring adjustment that I really like is assigning assists on plays that result in a missed shot because the shooter was fouled in the act of shooting.

Agree that sports stats are somewhat limited, especially in a team game as complex and interconnected as basketball.

 

[GRIFFBIFF20]

For the St Johns game, here are the efficiency ratings:

Actual Minutes on the floor: ..... Extrapolated to 40 minutes:
View attachment 65354 ..... View attachment 65355

statistics speak louder than words, as you cannot change them. very interesting indeed. go aubrey griffin

 

[SVCBeercats]

Thanks! This demonstrates the fallacy of extrapolating to 40 minutes.

For the St Johns game, here are the efficiency ratings:

Actual Minutes on the floor: ..... Extrapolated to 40 minutes:
View attachment 65354 ..... View attachment 65355

 

[UcMiami]

The problem with this is Williams is a defensive stopper but does not get blocks or steals - is she less valuable if she causes a 20 ppg player to score only 6, vs someone who has two steals and a block?

Or the player that makes the tough pass to the player who then drops it off to the scorer, or the player who set the killer pick to free a shooter.

Don't get me wrong, I love stats. I just think the more complicated they become the further from 'eye test' they get.

Westbrook often does lots of unrecorded things to help the team in ways that are generally more important than Griffin, but stats aren't going to show that. And PERS can make it look even more of kilter.

 

[HuskyNan]

I am not a fan of stats because they are malleable, ie, they can prove pretty much whatever you'd like.
Wikipedia: "PER takes into account accomplishments, such as field goals, free throws, 3-pointers, assists, rebounds, blocks and steals, and negative results, such as missed shots, turnovers and personal fouls." But what's the relationship to PER numbers and game outcomes? Inferred is that the higher the PER the more the player contributes to winning. In a low PER performance, say, with Paige against TN, what's the value of her making a game-winning (or game-securing) shot at the end? Bueckers also scored the last 13 points v. SC. Can we safely throw out everything she did up to that point? Actually we do because she won the game for UConn. That's all that mattered in the end. Stats don't determine legends (though they may confirm a player's greatness).
Finally, who would you like to have take the last-minute shot that determines a game's outcome? Think Geno looks at the PER to decide?

“Statistics are like bikinis. What they reveal is suggestive, but what they conceal is vital.”​

― Aaron Levenstein

 

[oldhusky]

PER is a complicated calculation involving all of a player's statistics. It is adjusted for minutes played and also game pace. It is normalized to an average figure of 15. If your average PER for a team weighted by minutes played is above that you are a good team and visa versa. I generally think an average team might have starters that play 2/3rds of the minutes at an average of about 16.5, and the bench might average a PER of about 12 in a third of the minutes,to produce an overall of 15.

sports-reference.com is an excellent site that gives many advanced stats including including PER for free regarding the NBA, WNBA and Men's college basketball. Unfortunately they don't provide it for women's college hoops.
There is a paid site ($20/year) called herhoopstats.com that does give PER for college women and the WNBA.

Here are the Uconn PER numbers before today's game from that site.

31.8 Paige Bueckers

30.9 Olivia Nelson-Ododa

29.6 Aaliyah Edwards

28.3 Aubrey Griffin

23.6 Mir McLean

20.6 Christyn Williams

20.6 Evina Westbrook

13.1 Anna Makurat

12.1 Nika Muhl

and for the barely used players, Autumn 24.1, Piath 15.5, and Saylor -3.5

I should note that early in the season Nika was minus as well.

 

[VidNick]

I am not a fan of stats because they are malleable, ie, they can prove pretty much whatever you'd like.

Look at the NBA/WNBA PER stats: you will not find a bad player on the top part of the list and you will not see a great player missing. We already know who the great players are, but stats confirm and refine our understanding. Stats are having an enormous impact on sports. Why? Because they can provide insights not easily gathered in other ways. There are always intangibles of course and they can be very important too. But stats are what they are, not what you want them to be.

 

[VidNick]

Here are the Uconn PER numbers before today's game from that site.

Hey OldHusky: please let me know the address where you found these numbers. There are at least 3 stats floating around: , PER and Game Score Calculator. 30+ PER, for season stats would be very high. A single game, definitely possible.

 

[oldhusky]

Hey OldHusky: please let me know the address where you found these numbers. There are at least 3 stats floating around: , PER and Game Score Calculator. 30+ PER, for season stats would be very high. A single game, definitely possible.

The numbers I showed were from herhoopstats.com. As I indicated in my earlier post it is a paid site that charges $20 per year, but you can go there and get some information without paying. Another site that uses the same methodology as I understand it is sports-reference.com. They have advanced stats including PER for NBA, WNBA, and Men's College basketball, but unfortunately not for women's college hoops.