Streamers
Professor
Posts: 8304
Loc: NW Philadelphia
Reg: 11-21-04
|
11-19-17 03:20 PM - Post#236950
In response to 10Q
Of course it’s impactful when you lose your best defensive player, especially one a versatile as Matt was who could take on anyone from 2-4. He is not irreplaceable, but his absence forces adjustments and others to step up.
|
SomeGuy
Professor
Posts: 6415
Reg: 11-22-04
|
11-19-17 09:10 PM - Post#236962
In response to Jeff2sf
Well, I think you both may be right here. I do think that Matt Howard was a better defender last year than Woods is right now. I think he had greater versatility. However, losing that versatility has forced us to play a more conventional lineup defensively, and thus far that appears to be a good thing, or at least not a bad thing.
I do like Woods' strength. I'm never sure what makes certain smaller guards play bigger than they are and not get pushed around, but Woods seems to have it, whatever it is.
Assuming the defense (from Antonio and Max) and the lack of turnovers (from Antonio) aren't total mirages, and given the lack of decent ORATs generally from the bench, I wonder if simply lowering the usage rates for Woods and Rothschild (particularly) is an answer. Assuming we can keep 3 of AJ, Betley, Foreman, and Wood on the floor most of the time, we can carry a lower usage rate here and there if it comes with lots of other positives.
|
Jeff2sf
Postdoc
Posts: 4466
Reg: 11-22-04
|
11-19-17 09:21 PM - Post#236964
In response to SomeGuy
does no one understand Mike's point?
"You have to assess the defense as if it were allowing a reasonable percentage of jumpers to fall through the hoop. Because, at some point, it will. When it does, it's not going to look so great."
Cus this is essentially inarguable. it's been tested up and down basketball across thousands of games.
|
Streamers
Professor
Posts: 8304
Loc: NW Philadelphia
Reg: 11-21-04
|
11-19-17 11:23 PM - Post#236968
In response to Jeff2sf
Bad defense = open jumper
Good defense = hand in face late in the shot clock
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
11-20-17 09:02 AM - Post#236974
In response to Streamers
Penn is currently allowing its opponents to hit 17% of its threes. That's insane and will not sustain. So, rather than look at the defense with that anomaly baked in, you can leverage expected points per shot (XPPS) to understand what would happen in the long run given that shot selection.
The more important thing than the luck on oppt makes is that they're keeping opponents from shooting that shot (28% of all attempts this year; 40% last year).
I don't have my XPPS numbers in front of me, but I'd wager that Penn's defense so far grades out as a bit better than average - somewhere in line with where it was last year. It is getting there, though, in a much more aesthetically pleasing way - more pressure in man defense scaring shooters off the line and forcing more TOs but yielding more action at the rim versus last year when the zone yielded a ton of threes that oppts hit at high rates.
|
palestra38
Professor
Posts: 32871
Reg: 11-21-04
|
11-20-17 10:23 AM - Post#236977
In response to mrjames
OK, whenever we start talking "luck", my eyes glaze over. We are playing much better defense on the 3s because Woods and Foreman get out much quicker and more effectively than our guard combos last year, plus they know that they have their backs covered by the 2 bigs inside. Howard was a "tweener" in that he covered a big guard or small forward--very good defensive player especially on the drive. But these 2 guys get out on the 3 much better. While it is an extremely small sample, so far, Penn's Div 1 opponents have shot the 3 at a 23.5% clip (Fairfield, with fewer Foreman/Woods minutes), 15% (LaSalle) and 11% (Navy) while shooting at a 36%/26%/40% clip against their other (usually higher powered) opponents. You can't attribute this to "luck" although you can say it's too small a sample to make definitive conclusions. But the eye test confirms that these numbers are correlated with much tighter man defense on the 3 point shooters.
|
Jeff2sf
Postdoc
Posts: 4466
Reg: 11-22-04
|
11-20-17 10:28 AM - Post#236978
In response to palestra38
except you CAN attribute the make rate to luck. "I don't care what the numbers say, I'm an genius (paraphrased)" blah blah blah. Meanwhile I've watched all the games and you haven't. The opponents are doing the opposite of what you're saying with respect to "at the rim". They are taking a higher percentage of shots at the rim AND converting them at a higher rate.
I will say that I could buy the Woods role in successfully chasing people off the line. But I'm very skeptical with respect to Max's contribution on defense. Not only do I think he's not great, but he's making a guy I think could be great play out of position.
|
palestra38
Professor
Posts: 32871
Reg: 11-21-04
|
11-20-17 10:45 AM - Post#236980
In response to Jeff2sf
"Luck" in your vocabulary means "that which I cannot explain."
|
Jeff2sf
Postdoc
Posts: 4466
Reg: 11-22-04
|
11-20-17 10:53 AM - Post#236981
In response to palestra38
sometimes it means that. in this case it means "the shooting percentage is unsustainable and will regress to the mean".
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
11-20-17 10:59 AM - Post#236982
In response to Jeff2sf
I can guarantee that Penn's 3PT% allowed will rise from here. And that it will be at least 10 percentage points higher from here on out - and very likely more.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
11-20-17 11:02 AM - Post#236983
In response to palestra38
We are playing much better defense on the 3s because Woods and Foreman get out much quicker and more effectively than our guard combos last year, plus they know that they have their backs covered by the 2 bigs inside.
Penn's defensive rating thus far is 12 points per 100 poss higher when Woods and Foreman are on the floor together versus all other lineups. Not that 92 poss are a good representation of what we'll see moving forward, but just looking at what has happened, Penn's defensive outcomes have been worse with them on the floor together than not.
|
palestra38
Professor
Posts: 32871
Reg: 11-21-04
|
11-20-17 11:28 AM - Post#236986
In response to mrjames
I readily agree that the sample is small. What I expect to keep seeing is that Penn's defense against the 3 is superior to the other teams playing our opponents. Time will either bear this out or not.
|
yoyo
Senior
Posts: 365
Reg: 03-25-09
|
11-20-17 11:38 AM - Post#236987
In response to palestra38
I think as Antonio continues to get in basketball shape, his defense will only improve.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
Steve Donahue's assessment
11-20-17 12:14 PM - Post#236991
In response to palestra38
And that's fine, but magnitudes are important here.
Right now, Penn is allowing 17% shooting from 3. If we look at the middle 300 team defenses from last year, they allowed opponents to shoot 31.6% to 38.4% from three. So, while we can quibble about where Penn will end up in that range, the gap to the bottom of that range at present is over 2X the entire size of that range. Whether we'd add 7 threes to the made total of opponents thus far (31.2%) or 8 threes (33.3%) or 9 threes (35.4%) or 10 threes (37.5%), it doesn't change the fact that Penn is somewhere around an average defense (either a bit above or a bit below) versus looking like a Top 25 defense with its current 89 raw DRAT (only trivially worse adjusted with opponent ORATs at 99.8).
|
TheLine
Professor
Posts: 5597
Age: 60
Reg: 07-07-09
|
11-20-17 12:31 PM - Post#236994
In response to palestra38
"Luck" in your vocabulary means "that which I cannot explain."
No.
It means that some part of Penn's defensive success is due to the other team missing a greater percentage of shots they ordinarily make and/or the other team missing a greater percentage of shots other teams ordinarily make. Holding the opponents to a 17% make percentage on 3's is unsustainable. Even the best defense is incapable of doing that for an entire season.
I do agree Penn's defense has been doing a good job. None of the starting 5 are liabilities. Mike is just stating that there will be an inevitable regression to the mean.
|
palestra38
Professor
Posts: 32871
Reg: 11-21-04
|
11-20-17 12:33 PM - Post#236995
In response to TheLine
I agree that the current numbers are probably unsustainable. But the test is if we consistently hold opponents to a lower 3 point percentage than other teams. I think that is likely.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
11-20-17 12:43 PM - Post#236997
In response to palestra38
The test should be whether you generally hold opponents to a lower percentage of 3pters taken as a percentage of overall shots. That’s a far more impressive component of what Penn is doing right now than its random success in keeping shots from falling.
|
palestra38
Professor
Posts: 32871
Reg: 11-21-04
|
11-20-17 12:52 PM - Post#237001
In response to mrjames
Why wouldn't both be relevant? By discounting missed 3s as "luck", you're assuming that all 3s are taken wide open. There is a huge difference between having that hand in your face as someone comes up on you much more rapidly than you expect when you set to take the shot. Having guys that can get out quickly is the hallmark of good defense. You don't always want to play so tight that they don't take the shot, because that means there are openings to take it inside.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
11-20-17 12:56 PM - Post#237003
In response to palestra38
I don't have the time to have this argument again. Please read these for more:
https://kenpom.com/blog/offense-vs-defen se-3point-...
https://kenpom.com/blog/offense-vs-defen se-threepo...
|
Jeff2sf
Postdoc
Posts: 4466
Reg: 11-22-04
|
11-20-17 01:01 PM - Post#237006
In response to mrjames
seriously. this is so intellectually lazy. it's like you never read anything besides what you type.
And the fact you don't realize you play the opposite role of this on the OT board is just so... something.
|
