Thursday, February 15, 2007
Threes, consistency, and hoary old stereotypes
It is said that teams that shoot more threes are more inconsistent. (And by "it is said" I mean I say it too--see yesterday's post on Florida, for starters.) Is that true in the Big Ten this year?
Yes and no. Shooting percentages for threes are inherently more volatile than shooting percentages for twos. This much we knew--the basket's just further away.
But the surprising part is that some of your most inconsistent shooting teams don't shoot a lot of threes--and some of your most consistent teams do shoot a lot of threes. What the?...
Take Iowa.
Being a Hawkeye fan is hazardous to your expectations
Steve Alford might notice a few gray hairs in his distinctive coif this season: only Minnesota exhibits more erratic shooting from game to game:
Serenity now--shooting, from most to least consistent
(standard deviation in effective FG pct.: conference games only, through Feb. 14)
1. Wisconsin (4.7)
2. Northwestern (6.2)
3. Indiana (6.4)
4. Ohio State (6.4)
5. Purdue (8.0)
6. Illinois (8.4)
7. Michigan (8.8)
8. Penn State (9.4)
9. Michigan State (9.7)
10. Iowa (10.2)
11. Minnesota (12.9)
(The Gophers on offense are what grim hoops analysts in white lab coats refer to delicately as "chaos in human form." No Big Ten team is more erratic in any form of shooting than Minnesota is when they shoot threes.)
Funny thing is: the Hawkeyes are this inconsistent despite the fact that they hardly ever shoot threes. Only Michigan and Michigan State have shot fewer threes than the Hawks in conference play.
And then there's the flip side: look at Northwestern and Ohio State on this list. The Wildcats shoot more threes by far than any other Big Ten team. And yet they're the very model of consistency. (Which actually is bad news in Evanston: the 'Cats have been bricking their threes at a prodigious, albeit consistent, rate.) Likewise, the Buckeyes shoot more than their fair share of threes and yet are able to post a consistent level of shooting success from game to game.
(And Wisconsin? Steady as she goes. They're the third-best shooting team in a conference that isn't shooting all that well this season. You know what you're going to get from Bo Ryan's team.)
Bottom line: shooting threes is more volatile than shooting twos. But consistency, or the lack thereof, is less about where the shot is attempted on the floor and more about who's attempting the shot.
In today's less Wonk-ish venues....
Ohio State beat Penn State 64-62 in State College last night. In an eerie recurrence of the game against Michigan State in Columbus, the Buckeyes led by as many as 24 in the second half but ended up watching Mike Walker shoot a last-second three that would have cost them the win. In the game's last 13 minutes, during which time OSU scored just ten points, the Buckeyes shot eight threes and missed all of them. The Nittany Lions, symmetrically enough, also shot eight threes over this span--but they made five of theirs. Greg Oden posted a 15-10 dub-dub for Ohio State, while Geary Claxton recorded a 19-10 for the Nittany Lions. (Box score.)
Wisconsin beat Minnesota 75-62 last night in Minneapolis. Classic Badger win: a +13 in FTAs and a -9 in turnovers. Truly Bo-dacious. The range of options open to Ryan with this team--and, more importantly, his willingness to use them--is truly impressive. Last night he went eight minutes with a backcourt comprised of Kammron Taylor, Jason Bohannon, and Trevon Hughes. Alando Tucker led all scorers with 29 points on 20 shots. Lawrence McKenzie made 4-of-6 threes and scored 21 for the Gophers. (Box score (pdf).)
Indiana plays Purdue tonight in West Lafayette. "For us—where we are right now with out record—we have to win our home games if we expect to have a chance to get into the NCAA tournament," says Matt Painter. Coverage of the Painter-Earl Calloway kerfuffle from last time here.
Northwestern plays Iowa in Iowa City tonight. The Wildcats have won four of their last five games against the Hawkeyes. (Then again three of those were in Evanston.)
Rumors are a-swirlin' around Illinois guard Jamar Smith's one-car accident Monday night, a collision with a tree that has landed teammate Brian Carlwell in the hospital.
Wonk back!
Don't just mutter ineffectually; email me!
Axes to grind
Yesterday I posted a tempo-free aerial for the SEC and noted that teams score a lot of points in that thar conference. The alert readers respond!
John,
The "bad offense, good defense" quadrant in the SEC aerial is empty only because the axes are placed at 1.00 PPP. It looks like the SEC averages about 1.05 PPP in conference play. There's nothing sacred about1.00, so perhaps we'd get a better division of teams if the axes were placed at the conference average. Then to compare teams from different conferences, we could use both the raw and conference-specific ratings to discuss a team.
I really don't think that the SEC and ACC are populated with bad defensive teams. I just think the style of play there encourages higher PPP. I think Tennessee's position on the graph is more like Purdue's than Iowa's (if the Big Ten's average PPP is truly 1.00, that is).
Robin F.
Ah, a chance for me to wax Rorty-esque! Cool!
Of course there's nothing sacred about 1.00 PPP. It's an arbitrary round number. Nor is there anything sacred about 0 on a thermometer. Or the top 500 companies as identified by Fortune. Or batting .400. Round numbers are handy cognitive short-cuts--and I'm all about short-cuts.
And as far as grouping everything around a conference average, sure, I do a lot of that. But I also like seeing how conferences differ. I devote a good deal of effort to parsing out these differences in pace, style, and coaches' haircuts. In fact, watch for an end-of-year post holding forth on each conference's average for points-per-possession--and the implications thereof.
Still, I can see where the axes could lead one to believe that I'm defining "good" defense as 0.99 or below in every case for every team now and forever more. And I'm not. So maybe I should just kill the axes. We'll see. ¶ 6:29 AM
It is said that teams that shoot more threes are more inconsistent. (And by "it is said" I mean I say it too--see yesterday's post on Florida, for starters.) Is that true in the Big Ten this year?
Yes and no. Shooting percentages for threes are inherently more volatile than shooting percentages for twos. This much we knew--the basket's just further away.
But the surprising part is that some of your most inconsistent shooting teams don't shoot a lot of threes--and some of your most consistent teams do shoot a lot of threes. What the?...
Take Iowa.
Being a Hawkeye fan is hazardous to your expectations
Steve Alford might notice a few gray hairs in his distinctive coif this season: only Minnesota exhibits more erratic shooting from game to game:
Serenity now--shooting, from most to least consistent
(standard deviation in effective FG pct.: conference games only, through Feb. 14)
1. Wisconsin (4.7)
2. Northwestern (6.2)
3. Indiana (6.4)
4. Ohio State (6.4)
5. Purdue (8.0)
6. Illinois (8.4)
7. Michigan (8.8)
8. Penn State (9.4)
9. Michigan State (9.7)
10. Iowa (10.2)
11. Minnesota (12.9)
(The Gophers on offense are what grim hoops analysts in white lab coats refer to delicately as "chaos in human form." No Big Ten team is more erratic in any form of shooting than Minnesota is when they shoot threes.)
Funny thing is: the Hawkeyes are this inconsistent despite the fact that they hardly ever shoot threes. Only Michigan and Michigan State have shot fewer threes than the Hawks in conference play.
And then there's the flip side: look at Northwestern and Ohio State on this list. The Wildcats shoot more threes by far than any other Big Ten team. And yet they're the very model of consistency. (Which actually is bad news in Evanston: the 'Cats have been bricking their threes at a prodigious, albeit consistent, rate.) Likewise, the Buckeyes shoot more than their fair share of threes and yet are able to post a consistent level of shooting success from game to game.
(And Wisconsin? Steady as she goes. They're the third-best shooting team in a conference that isn't shooting all that well this season. You know what you're going to get from Bo Ryan's team.)
Bottom line: shooting threes is more volatile than shooting twos. But consistency, or the lack thereof, is less about where the shot is attempted on the floor and more about who's attempting the shot.
In today's less Wonk-ish venues....
Ohio State beat Penn State 64-62 in State College last night. In an eerie recurrence of the game against Michigan State in Columbus, the Buckeyes led by as many as 24 in the second half but ended up watching Mike Walker shoot a last-second three that would have cost them the win. In the game's last 13 minutes, during which time OSU scored just ten points, the Buckeyes shot eight threes and missed all of them. The Nittany Lions, symmetrically enough, also shot eight threes over this span--but they made five of theirs. Greg Oden posted a 15-10 dub-dub for Ohio State, while Geary Claxton recorded a 19-10 for the Nittany Lions. (Box score.)
Wisconsin beat Minnesota 75-62 last night in Minneapolis. Classic Badger win: a +13 in FTAs and a -9 in turnovers. Truly Bo-dacious. The range of options open to Ryan with this team--and, more importantly, his willingness to use them--is truly impressive. Last night he went eight minutes with a backcourt comprised of Kammron Taylor, Jason Bohannon, and Trevon Hughes. Alando Tucker led all scorers with 29 points on 20 shots. Lawrence McKenzie made 4-of-6 threes and scored 21 for the Gophers. (Box score (pdf).)
Indiana plays Purdue tonight in West Lafayette. "For us—where we are right now with out record—we have to win our home games if we expect to have a chance to get into the NCAA tournament," says Matt Painter. Coverage of the Painter-Earl Calloway kerfuffle from last time here.
Northwestern plays Iowa in Iowa City tonight. The Wildcats have won four of their last five games against the Hawkeyes. (Then again three of those were in Evanston.)
Rumors are a-swirlin' around Illinois guard Jamar Smith's one-car accident Monday night, a collision with a tree that has landed teammate Brian Carlwell in the hospital.
Wonk back!
Don't just mutter ineffectually; email me!
Axes to grind
Yesterday I posted a tempo-free aerial for the SEC and noted that teams score a lot of points in that thar conference. The alert readers respond!
John,
The "bad offense, good defense" quadrant in the SEC aerial is empty only because the axes are placed at 1.00 PPP. It looks like the SEC averages about 1.05 PPP in conference play. There's nothing sacred about1.00, so perhaps we'd get a better division of teams if the axes were placed at the conference average. Then to compare teams from different conferences, we could use both the raw and conference-specific ratings to discuss a team.
I really don't think that the SEC and ACC are populated with bad defensive teams. I just think the style of play there encourages higher PPP. I think Tennessee's position on the graph is more like Purdue's than Iowa's (if the Big Ten's average PPP is truly 1.00, that is).
Robin F.
Ah, a chance for me to wax Rorty-esque! Cool!
Of course there's nothing sacred about 1.00 PPP. It's an arbitrary round number. Nor is there anything sacred about 0 on a thermometer. Or the top 500 companies as identified by Fortune. Or batting .400. Round numbers are handy cognitive short-cuts--and I'm all about short-cuts.
And as far as grouping everything around a conference average, sure, I do a lot of that. But I also like seeing how conferences differ. I devote a good deal of effort to parsing out these differences in pace, style, and coaches' haircuts. In fact, watch for an end-of-year post holding forth on each conference's average for points-per-possession--and the implications thereof.
Still, I can see where the axes could lead one to believe that I'm defining "good" defense as 0.99 or below in every case for every team now and forever more. And I'm not. So maybe I should just kill the axes. We'll see. ¶ 6:29 AM
