Charting Duke v. Cal Poly Pomona (exh.)

[pfrduke]

The starting lineup as a lineup was -7 in the first half, and finished the game in the negative overall. Lots of mixing and matching - only two lineups played more than once (and they were collectively -2). The three-guard backcourt worked very well tonight, at least when those three guards were Irving, Smith, and Curry - the team was +15 and shutout Pomona with those three on the floor.

Individuals
Nolan Smith 58-36 (+22)
Kyle Singler 63-42 (+21)
Kyrie Irving 60-39 (+21)
Seth Curry 46-26 (+20)
Miles Plumlee 52-33 (+19)
Ryan Kelly 18-10 (+8)
Mason Plumlee 49-46 (+3)
Andre Dawkins 29-26 (+3)
Josh Hairston 22-27 (-5)
Tyler Thornton 8-15 (-7)

Per 40 Minutes
Miles Plumlee +36.2
Ryan Kelly +35.6
Nolan Smith +35.2
Seth Curry +34.8
Kyle Singler +30
Kyrie Irving +30
Andre Dawkins +7.5
Mason Plumlee +4.6
Josh Hairston -12.5
Tyler Thornton -35

Lineups (Score, times used, margin)
Irving-Smith-Curry-Singler-Kelly (6-0, x1, +6)
Irving-Smith-Curry-Singler-Miles (5-0, x1, +5)
Smith-Curry-Dawkins-Singler-Mason (6-2, x1, +4)
Irving-Smith-Curry-Mason-Miles (4-0, x1, +4)
Irving-Smith-Singler-Kelly-Miles (4-0, x1, +4)
Curry-Dawkins-Singer-Mason-Miles (5-2, x1, +3)
Thornton-Curry-Dawkins-Hairston-Miles (3-0, x1, +3)
Irving-Curry-Singler-Hairston-Mason (4-2, x1, +2)
Irving-Curry-Dawkins-Hairston-Miles (4-2, x1, +2)
Irving-Smith-Dawkins-Singler-Kelly (2-0, x1, +2)
Irving-Thornton-Curry-Hairston-Mason (3-2, x1, +1)
Irving-Curry-Dawkins-Mason-Miles (2-1, x1, +1)
Irving-Smith-Singler-Hairston-Kelly (4-4, x2, 0)
Smith-Curry-Singler-Hairston-Kelly (2-2, x1, 0)
Irving-Smith-Dawkins-Singler-Miles (2-2, x1, 0)
Irving-Smith-Dawkins-Singler-Mason (0-0, x1, 0)
Irving-Thornton-Dawkins-Hairston-Kelly (0-0, x1, 0)
Smith-Dawkins-Singler-Mason-Miles (3-4, x1, -1)
Irving-Smith-Singler-Mason-Miles (20-22, x4, -2)
Irving-Curry-Singler-Hairston-Kelly (0-2, x1 -2)
Thornton-Curry-Dawkins-Hairston-Kelly (0-2, x1, -2)
Irving-Thornton-Dawkins-Hairston-Mason (0-2, x1, -2)
Thornton-Curry-Dawkins-Hairston-Mason (2-9, x1, -7)

[Kedsy]

I'm not doubting you, but your numbers are a lot different from scACChoops (http://www.scacchoops.com/ViewHDGame.asp?hSchedule=7164&bView=0). Not sure why.

[billyj]

Interesting. Would Singler switch to the 4? This is going to be difficult giving his future will be a 3.

[pfrduke]

I'm not doubting you, but your numbers are a lot different from scACChoops (http://www.scacchoops.com/ViewHDGame.asp?hSchedule=7164&bView=0). Not sure why.

I'm not sure why either, but at a quick glance, that site doesn't have it right. For example, when you add up all the players' +/-, it has to end up in a multiple of 5 (to account for the fact that there are 5 players on the court at all times). That box represents a total +/- across out players of +167, which a) is not a multiple of 5, and b) is way too high anyway - we won by 21, so the total +/- for our guys has to be +105.

I went from the official play-by-play provided at the Live Stats site linked from the GoDuke page (because, for some reason, the official play-by-play in the box score doesn't have substitution data).

[pfrduke]

Interesting. Would Singler switch to the 4? This is going to be difficult giving his future will be a 3.

Well, as many others have observed, he's likely to spend at least some time playing the 4 this season, if for no other reason than that a lineup with Smith, Irving, and Curry (or Dawkins) on the floor is pretty darn dangerous, particularly against zones. But he'll also spend plenty (probably most) of his time at the 3 because of the Plumlees and Kelly.

[Duvall]

Interesting. Would Singler switch to the 4? This is going to be difficult giving his future will be a 3.

It certainly ruined the careers of Shane Battier and Luol Deng.

I think you may be taking too much from a few possessions of a November exhibition game.

[riverside6]

I'm not doubting you, but your numbers are a lot different from scACChoops (http://www.scacchoops.com/ViewHDGame.asp?hSchedule=7164&bView=0). Not sure why.

We had some problems with +/- for this game. I'll rebuild it tomorrow sometime, I'm sure pfrduke is dead on with his numbers.

[pfrduke]

Inspired by Basketball Prospectus, which has been running these (http://www.basketballprospectus.com/article.php?articleid=1243) as their daily recaps for the NBA games, below is an "enhanced" box score of the Pomona game. The box reflects: Plays, the number of possessions each player was on the court; Usage, the percentage of the team's possessions each player used while on the court; Points Created Per Possession, how efficiently each player used his offensive possessions; +/-; and game rating. The game ratings of all players total the team's offensive rating for the game (the decimal points may be off due to rounding.


Plays USG PCP +/- GR
Singler 54 18.0 1.59 21 20.0
Irving 54 22.1 1.14 21 17.7
Miles 41 28.1 1.05 19 15.5
Mason 50 14.9 1.49 3 14.3
Smith 48 19.3 1.11 22 13.4
Hairston 31 27.3 0.81 -5 8.8
Curry 45 17.8 0.77 20 7.9
Kelly 17 31.9 0.40 8 2.8
Dawkins 31 11.2 0.64 3 2.8
Thornton 16 6.8 0.91 -7 1.2

Pace ORTG eFG% oRB% TO% FTA%
77.4 104.7 0.500 0.351 0.181 0.159

The numbers at the bottom are the team numbers and reflect: Pace, the number of possessions in the game; ORTG, team points per 100 possessions; efg%, shooting percentage, weighting for 3s; oRB%, how many of our misses we rebounded; to%, how often our possessions ended in a turnover; and fta%, how often our possessions resulted in a trip to the line.

[ice-9]

Didn't watch the game, but was surprised to see from stats that Josh played more minutes than Kelly (16 to 9). Is this a game specific thing, or possibly an indication for the rest of the season? Kelly seemed to have contributed more in limited minutes.

[Bob Green]

Didn't watch the game, but was surprised to see from stats that Josh played more minutes than Kelly (16 to 9). Is this a game specific thing, or possibly an indication for the rest of the season? Kelly seemed to have contributed more in limited minutes.

Kelly fouled out. Five fouls in nine minutes. Per reports in the game thread, he was the victim of multiple bad calls.

[ice-9]

Kelly fouled out. Five fouls in nine minutes. Per reports in the game thread, he was the victim of multiple bad calls.

Ah, thanks for the clarification! Heading over to the game thread now. :)

[loran16]

@PFRDuke, thanks for doing the work. A Pace of 77.4 Possessions Per Game is way higher than last year (I think we were around 60). Interesting.

[Kedsy]

Inspired by Basketball Prospectus, which has been running these (http://www.basketballprospectus.com/article.php?articleid=1243) as their daily recaps for the NBA games, below is an "enhanced" box score of the Pomona game. The box reflects: Plays, the number of possessions each player was on the court; Usage, the percentage of the team's possessions each player used while on the court; Points Created Per Possession, how efficiently each player used his offensive possessions; +/-; and game rating. The game ratings of all players total the team's offensive rating for the game (the decimal points may be off due to rounding.


Plays USG PCP +/- GR
Singler 54 18.0 1.59 21 20.0
Irving 54 22.1 1.14 21 17.7
Miles 41 28.1 1.05 19 15.5
Mason 50 14.9 1.49 3 14.3
Smith 48 19.3 1.11 22 13.4
Hairston 31 27.3 0.81 -5 8.8
Curry 45 17.8 0.77 20 7.9
Kelly 17 31.9 0.40 8 2.8
Dawkins 31 11.2 0.64 3 2.8
Thornton 16 6.8 0.91 -7 1.2

Pace ORTG eFG% oRB% TO% FTA%
77.4 104.7 0.5 0.351 0.181 0.159

The numbers at the bottom are the team numbers and reflect: Pace, the number of possessions in the game; ORTG, team points per 100 possessions; efg%, shooting percentage, weighting for 3s; oRB%, how many of our misses we rebounded; to%, how often our possessions ended in a turnover; and fta%, how often our possessions resulted in a trip to the line.

Thanks for these. Interesting that the players with the highest usage numbers were Ryan, Miles, and Josh, in that order.

[gam7]

Inspired by Basketball Prospectus, which has been running these (http://www.basketballprospectus.com/article.php?articleid=1243) as their daily recaps for the NBA games, below is an "enhanced" box score of the Pomona game. The box reflects: Plays, the number of possessions each player was on the court; Usage, the percentage of the team's possessions each player used while on the court; Points Created Per Possession, how efficiently each player used his offensive possessions; +/-; and game rating. The game ratings of all players total the team's offensive rating for the game (the decimal points may be off due to rounding.


Plays USG PCP +/- GR
Singler 54 18.0 1.59 21 20.0
Irving 54 22.1 1.14 21 17.7
Miles 41 28.1 1.05 19 15.5
Mason 50 14.9 1.49 3 14.3
Smith 48 19.3 1.11 22 13.4
Hairston 31 27.3 0.81 -5 8.8
Curry 45 17.8 0.77 20 7.9
Kelly 17 31.9 0.40 8 2.8
Dawkins 31 11.2 0.64 3 2.8
Thornton 16 6.8 0.91 -7 1.2

Pace ORTG eFG% oRB% TO% FTA%
77.4 104.7 0.5 0.351 0.181 0.159

The numbers at the bottom are the team numbers and reflect: Pace, the number of possessions in the game; ORTG, team points per 100 possessions; efg%, shooting percentage, weighting for 3s; oRB%, how many of our misses we rebounded; to%, how often our possessions ended in a turnover; and fta%, how often our possessions resulted in a trip to the line.

Hey pfr, very interesting stuff. I'm interested in learning more about how to interpret these new stats. Is there a good write-up somewhere? I didn't see it in a quick look at the basketball prospectus link you provided. I don't have a frame of reference for whether 77.4 is a fast or slow pace, etc. In fact, I'm not even sure how the pace (number of possessions) can have a decimal (77.4).

Do you happen to have these advanced stats for last year's Duke team? Seeing this comparison would give us some sort of a frame of reference at least.

[loran16]

Hey pfr, very interesting stuff. I'm interested in learning more about how to interpret these new stats. Is there a good write-up somewhere? I didn't see it in a quick look at the basketball prospectus link you provided. I don't have a frame of reference for whether 77.4 is a fast or slow pace, etc. In fact, I'm not even sure how the pace (number of possessions) can have a decimal (77.4).

Do you happen to have these advanced stats for last year's Duke team? Seeing this comparison would give us some sort of a frame of reference at least.


IIRC Possesions can have decimals based upon the idea that to calculate possessions you add up both teams # of possessions then divide. If one team has more possessions than the other (held balls, steals in the backcourt, etc.)

Last year Duke had a pace of 65.5 possessions per game. For reference, if Duke was to maintain a pace of 77.4 all year, it would have come in as the 3rd fastest team in all of D1:


http://kenpom.com/stats.php?y=2010&s=4

[sagegrouse]

We had some problems with +/- for this game. I'll rebuild it tomorrow sometime, I'm sure pfrduke is dead on with his numbers.

Thanks for weighing in and keeping the discussion on an even keel.

sagegrouse

[alteran]

Kelly fouled out. Five fouls in nine minutes. Per reports in the game thread, he was the victim of multiple bad calls.

The last one was particularly ridiculous, made by the ref who made at least two of the other calls on him, and when he made it he glared at Kelly.

Usually you have to go to a high school game to see that level of bull.

K rode that official off and on for the remainder of the game.

[riverside6]

We rebuilt our +/- scores and we are now pretty much the same as pfrduke (except for Dawkins and Mi Plumlee)

http://www.scacchoops.com/ViewHDGame.asp?hSchedule=7164&bView=0

[MulletMan]

Interesting. Would Singler switch to the 4? This is going to be difficult giving his future will be a 3.

As K has said many times, Duke doesn't really play "positions". Players play.

That's part of the reason that we'll miss Lance so much this year on the defensive end. He could guard anyone and allowed us the flexibility to have an extra big on the floor. It didn't matter what switching needed to occur because he could guard the PG or the C and everything in between.

[pfrduke]

Hey pfr, very interesting stuff. I'm interested in learning more about how to interpret these new stats. Is there a good write-up somewhere? I didn't see it in a quick look at the basketball prospectus link you provided. I don't have a frame of reference for whether 77.4 is a fast or slow pace, etc. In fact, I'm not even sure how the pace (number of possessions) can have a decimal (77.4).

Do you happen to have these advanced stats for last year's Duke team? Seeing this comparison would give us some sort of a frame of reference at least.

Ken Pomeroy's website (http://kenpom.com/rate.php) is a useful repository for this sort of thing, and offers helpful explanations (http://kenpom.com/blog/index.php/weblog/stats_explained/). As loran said, 77.4 is quite fast - as an illustration, assume you had a game where both teams used 30 seconds off the shot clock every time down the court. That would be a 40 possession game (each team would have 1 possession each minute). At 77.4, we're almost doubling that speed, meaning that the average possession takes just over 15 seconds.

As for the decimal, that comes because the box score doesn't give us a perfectly accurate number of possessions - mostly due to free throws. Possessions is shots attempted plus turnovers minus offensive rebounds plus (free throw attempts *.475). People a lot smarter than I am have figured out that .475 is the right weight to give to an attempt to account for the fact that sometimes free throws are and-ones (meaning no extra possession past the shot attempt), and sometimes the front-end of one-and-ones are missed (meaning only one FTA in a possession, instead of two). I run the numbers for both teams, then average them to get the pace.

Pomeroy also has much of these for last year's team (http://kenpom.com/team.php?team=Duke). No +/- (it's very hard to get reliable play-by-play data for college games - we're lucky that Duke usually has it), but all the rest is there.

[NSDukeFan]

As K has said many times, Duke doesn't really play "positions". Players play.

That's part of the reason that we'll miss Lance so much this year on the defensive end. He could guard anyone and allowed us the flexibility to have an extra big on the floor. It didn't matter what switching needed to occur because he could guard the PG or the C and everything in between.

I was thinking this as well. When the team extends its pressure, it's nice to have the bigger players on the floor able to pressure and move their feet well enough to recover. Lance would have been great defending in this kind of role. I look forward to continued improvements from Miles, Mason and Ryan defending on the perimeter and inside and being in solid position to help out when players get beat.

[Kedsy]

We rebuilt our +/- scores and we are now pretty much the same as pfrduke (except for Dawkins and Mi Plumlee)

http://www.scacchoops.com/ViewHDGame.asp?hSchedule=7164&bView=0

You now have Andre Dawkins at +18 (in 16 minutes), while Basketball Prospectus and pfrduke have him at +3. That's a huge difference. Did one of you or the other miss a shift on the court for him?