Untitled Document
Brown Columbia Cornell Dartmouth Harvard Penn Princeton Yale



 Page 3 of 6 « First<3456
Username Post: What do we know after 7 games        (Topic#20726)
AsiaSunset 
Postdoc
Posts: 4350

Reg: 11-21-04
What do we know after 7 games
11-26-17 11:15 AM - Post#237924    
    In response to TheLine

Just to be clear, I didn't say Mike doesn't watch games. I said it's clear that he didn't watch the Towson game and that his comments about the game reflect that.

As for Jeff - he sometimes reacts emotionally without reading posts or demonstrating an understanding of what has been written. I said above that the stats from the Monmouth game, taken in isolation, distort a key metric cited by Mike James The fact that Monmouth had 50 ft attempts and only 70 plus fg attempts negatively impacts that metric, but when you look at the box score, Penn had basically the same numbers. So it doesn't mean much to me other than the fact that this tells me more about the nature of the game rather than indicating a deteriorating metric that Mike used to support an argument that the Penn defense hasn't improved.

When dealing with small sample sizes, distortions are to be expected and there is often more to the story than cumulative metrics indicate. I made a similar comment about the likely movement of Antonio's ORAT after the game, but I ascribe little significance to that number either before or after the game. We'll know more in time. It seemed pretty obvious to many of us that Antonio would start this year and would be an important component were this team to succeed. Let's not forget that others like Jeff were lobbying for the likes of Sam Jones and Devon Goodman.

Additionally - I didn't offer an opinion of whether Penn played well or not last night. I didn't see the game. I do know that better Penn teams have struggled in that building, so it was nice to emerge with a win.

 
palestra38 
Professor
Posts: 32683

Reg: 11-21-04
Re: BU
11-26-17 11:17 AM - Post#237925    
    In response to penn62

Antonio looks quite a bit better when he isn't playing 3 games in 3 days, doesn't he?

It's that sort of thing that makes us groan when those magically silent until that 3 game tournament start touting the value of his ORAT and DRAT numbers after game 2 and 3 of that series. It's a long season, sample size does matter as does context. Anyone who ignores them isn't telling the real story.

 
Streamers 
Professor
Posts: 8141
Streamers
Loc: NW Philadelphia
Reg: 11-21-04
11-26-17 11:26 AM - Post#237926    
    In response to 10Q

  • 10Q Said:
Yes. That’s what the otb is for.



Like

 
Jeff2sf 
Postdoc
Posts: 4466

Reg: 11-22-04
Re: BU
11-26-17 11:32 AM - Post#237927    
    In response to palestra38

p38 and asia wrote a bunch of luddite, anti-fact garbage. let's leave it at that.

mike, i saw your post. quote formatting problems aside, i wish you left it up. I have no patience for this ish.

 
penn nation 
Professor
Posts: 21082

Reg: 12-02-04
11-26-17 11:35 AM - Post#237929    
    In response to 10Q

Yeah you know me!

  • 10Q Said:
Yes. That’s what the otb is for.




 
AsiaSunset 
Postdoc
Posts: 4350

Reg: 11-21-04
Re: BU
11-26-17 11:46 AM - Post#237931    
    In response to Jeff2sf

The final word from the President of the Sam Jones fan club.


 
Jeff2sf 
Postdoc
Posts: 4466

Reg: 11-22-04
11-26-17 11:49 AM - Post#237932    
    In response to AsiaSunset

That's some guy. I said I was open to seeing what he could do to help the lineup spacing. I also said I didn't believe the defensive metrics but was open to learning more. That spirit of not being blindly wed to something would suit you well.

You still don't realize you're arguing at an arithmetic level while mikes busy doing differential calculus.

 
mrjames 
Professor
Posts: 6062

Loc: Montclair, NJ
Reg: 11-21-04
Re: BU
11-26-17 11:56 AM - Post#237933    
    In response to palestra38

Listen, it's pretty simple. Right now, Penn's defense is holding opponents to the following efficiency ratings by shot type:

3PT Js: 115 ORAT
2PT Js: 83 ORAT
Layups: 128 ORAT
2 FTs: 147 ORAT

On 3PT Js, that is better than any Ivy achieved last year. On 2PT Js, that is better than any Ivy achieved last year. On layups, that's better than all but one team last year (PRIN) and would have been 2nd by 6 pts/100 poss.

Here are my predictions and empirical explanations. We can re-visit these at the end of the season:

Penn will not finish with a lower DRAT on 3PT Js than the lowest team from last year (though that's not a long fall with HAR at 116 last year). Penn's rebounding rate on oppt missed threes is 25%, so as Penn allows opponents to make more threes (Penn's 3PT% allowed will finish at the very lowest in the 32% range), it's not like that's going to wipe out many misses that opponents were getting back. I'd suspect Penn will finish closer to the middle of the pack from last year (125-131 DRAT on oppt 3s).

2PT Js are a total crapshoot (because you WANT opponents to shoot them) and tend to show a tight banding in DRAT that looks a lot like DRAT on FTs. That will invariably rise to the tight band in that 96-102 range.

Layup DRAT will rise as well. The best DRAT teams when it comes to layups are above average at blocking shots. Penn hasn't been since AJ was moved out of his zone positioning. As Penn stops fouling as much at the rim, I'd suspect this number will rise back into the range of the better (non-PRIN) teams from last year (mid-130s). Penn might be able to sustain this if it fouls anyone with a good look, but my presumption is that will be discouraged (for the very simple reason of keeping Max and AJ on the floor). Also, Penn's rebounding 74.5% of its opponents at rim misses, which is astounding and of which I'm a bit skeptical (teams tend to be far worse at rebounding at rim misses than 3s and 2PTJs, but Penn's oreb allowed is basically 25% across the board).

2FTs DRAT might rise marginally. Last year's DRAT went from 145 to 156 across the league with the average in the low 150s, so maybe a trivial rise in that is likely.

Aside from increasing turnovers (DRAT on those possessions is 0), these are the components of defensive efficiency. And these are the reasons why I believe that Penn's defensive efficiency will rise from its current 100.9 as the season wears on.

 
AsiaSunset 
Postdoc
Posts: 4350

Reg: 11-21-04
11-26-17 12:04 PM - Post#237937    
    In response to Jeff2sf

There’s is a lot written about defensive metrics and the relevance of team on individual DRAT numbers. It’s also not irrelevant that starters are often matched against starters so comparing their DRAT numbers with role players can be totally misleading.

Everything must be viewed within context. Jeff - you speculated about a lot of really unlikely outcomes on this board based solely on relying on numbers that sometimes provide clarity but often distort reality.

 
Jeff2sf 
Postdoc
Posts: 4466

Reg: 11-22-04
11-26-17 12:06 PM - Post#237938    
    In response to AsiaSunset

I haven't cited an individual DRAT number in at least 5 years. perhaps never.

you honestly have no idea what you're arguing about at this point.

 
Jeff2sf 
Postdoc
Posts: 4466

Reg: 11-22-04
11-26-17 12:09 PM - Post#237939    
    In response to Jeff2sf

I just checked. I've written the letters DRAT three times on this board including the time in the post above. The other two times were at the beginning of the season where I said I had no idea how to predict individual DRATs in context of predicting Antonio Woods season. The third was in 2011 where I referenced team DRAT.

So that's it, bro. You're creating straw men on the fly that are demonstrably false.

 
AsiaSunset 
Postdoc
Posts: 4350

Reg: 11-21-04
11-26-17 12:40 PM - Post#237942    
    In response to Jeff2sf

And of course I never said you did. I simply responded to your post. The discussion of DRAT numbers in this thread arose in the context of a single metric Mike threw out and my only comment directed to you was that you seemed to either not understand or purposely misrepresented my comment. That comment was about the Monmouth game and its impact on that metric. I was simply pointing out how numbers without context can lead to the wrong conclusions. But - yes - I do think individual DRAT numbers are less valuable than individual ORAT numbers and we probably all agree on that.

But Mike has now written a pretty extensive post on Penn team DRAT numbers that is informative. I think his suggestion that we revisit at the end of the season is a good one, especially when comparing Ivy teams. We all play different OOC opponents and having that base of 14 games against primarily common opponents makes the analysis far more useful.

I do think many Penn posters enjoy Mike's work and in fact were begging him for analysis when we were only two games into the season. The fact that I may discount the importance of some of these early season numbers doesn't mean that others don't hunger for them.

 
10Q 
Professor
Posts: 23197

Loc: Suburban Philly
Reg: 11-21-04
11-26-17 12:57 PM - Post#237943    
    In response to AsiaSunset

I hunger for better foul shooting. We put the foul in foul shooting.

 
Stuart Suss 
PhD Student
Posts: 1439

Loc: Chester County, Pennsylva...
Reg: 11-21-04
11-26-17 03:07 PM - Post#237950    
    In response to 10Q

I would have preferred to have the Villanova numbers to include in this discussion, but many of you are addressing the topic now. My analysis will be mainly narrative. If anyone is interested in the specific numbers, or my methodology, feel free to contact me.

The comparison is between last year’s non-conference schedule (INCLUDING the Princeton tournament game) and this year’s non-conference schedule (EXCLUDING the Brandywine game).

I look at the Offensive Efficiency Rating (OER, which is different from the ORat). Offensively we are essentially identical to the OER from last year’s non-conference schedule. Our three point shooting percentage is down, but we are not taking proportionately as many three point attempts (3FGA/FGA is down from last year’s 44.3% to this year’s 39.2%). Our biggest improvement is getting to the foul line (FTA/FGA improving from 23.1% to 34.4%). It is a shame that our actual foul shooting is not capitalizing on this improvement. So, while our Effective FG% and True FG% are slightly poorer, we have mostly compensated for that with a better turnover percentage. Offensive rebounding impact has netted out to basically the same as last year.

Defensively (DER) we have improved from 0.99 to 0.93. Our Effective FG% defense is identical to last season (47.1%) and True FG% defense is almost the same. The biggest negative is putting teams on the foul line (FTA/FGA deteriorating from 37.2% to 46.6%). Our opponents are not capitalizing on their more frequent trips to the foul line. There is a slight deterioration in forcing turnovers. Our biggest defensive improvement is much better defensive rebounding, probably arising from the greater use of two big men.

What do I conclude from this? Our non-conference schedule to date is weaker than last year’s non-conference schedule. We defeated a Monmouth team playing without their leading scorer (Micah Seaborn). We will soon play a top 100 team (St. Joe, currently battling injuries), a top 50 team (Temple), and a top 5 team (Villanova). We have three upcoming opponents who are currently ranked in Pomeroy’s bottom 50. But, we play two of them on the road where anything can happen.

It is prudent to expect our defensive three point percentage (29.7%) to regress to the national mean (currently 34.7%). The same principle would suggest that our own three point shooting (31.5%) will improve. It is even more prudent to expect our defensive free throw percentage (66.0%) to regress to the national mean (currently 70.4%). It is aspirational to expect our own free throw percentage to improve.

We may obtain contributions from our freshmen as one recovers from injury (Jelani), as one develops (Jarrod) and as one is assimilated into the rotation (Eddie). On the other hand, other teams are also waiting for injured players to return and for freshmen to contribute. Any improvement we have is an improvement which may be met in equal or greater proportion by our upcoming opponents.

If, as of today, we are only slightly improved in comparison to last year, and with the tougher part of the non-conference schedule yet to be played, some curbed enthusiasm may be in order.


 
10Q 
Professor
Posts: 23197

Loc: Suburban Philly
Reg: 11-21-04
11-26-17 03:11 PM - Post#237951    
    In response to Stuart Suss

Consider my parade officially rained on.

 
SomeGuy 
Professor
Posts: 6391

Reg: 11-22-04
Re: BU
11-26-17 10:51 PM - Post#237998    
    In response to palestra38

To me, you're the one artificially limiting this to a small sample size. We've now got, what, 48 games to determine an ORAT (or anything else) for Woods? And he keeps not improving his ORAT.

As for the 3 games in 3 days comment, last night it was the opposite. If the game had ended in regulation, he'd have had lousy numbers again. It was staying out there for an extra 20 minutes that allowed him to turn it around. So I'm not sure I that the narrative that he played too many days in a row works. Yesterday, it seemed like he outlasted everybody else instead of wearing down.

 
SomeGuy 
Professor
Posts: 6391

Reg: 11-22-04
11-26-17 11:02 PM - Post#237999    
    In response to Jeff2sf

Yes, I said that I would start Sam Jones over Woods at the start of the season. Of course, there isn't anything in the numbers so far that says I'm wrong -- Jones has a higher ORAT than Woods, and Woods is the only guy who is sub 100 among the starters. Of course, you can't really just plug a bunch of 115 ORAT guys together and expect a 115 ORAT -- they have to be able to do other things in order to actually achieve that. No other words, Sam's ORAT might go down playing more, or the guys around him might go down. More likely, we'd score more and give up more.

 
penn nation 
Professor
Posts: 21082

Reg: 12-02-04
11-26-17 11:49 PM - Post#238004    
    In response to Stuart Suss

  • Stuart Suss Said:
It is even more prudent to expect our defensive free throw percentage (66.0%) to regress to the national mean (currently 70.4%). It is aspirational to expect our own free throw percentage to improve.




I nominate this for post of the year.

 
Mike Porter 
Postdoc
Posts: 3614
Mike Porter
Loc: Los Angeles, CA
Reg: 11-21-04
Re: BU
11-27-17 12:50 AM - Post#238009    
    In response to palestra38

  • palestra38 Said:
Antonio looks quite a bit better when he isn't playing 3 games in 3 days, doesn't he?

It's that sort of thing that makes us groan when those magically silent until that 3 game tournament start touting the value of his ORAT and DRAT numbers after game 2 and 3 of that series. It's a long season, sample size does matter as does context. Anyone who ignores them isn't telling the real story.



Count me as someone who does think Woods numbers will improve, but I wouldn't hold up last nights game as proof.

That second half was the worst half I watched in some time and at the end Woods was pivotal in letting Monmouth close the gap through bad turnovers and missed shots/free throws. He made it up in overtime but his play last 10 mins of regulation was brutal.


 
Chip Bayers 
Professor
Posts: 6997
Chip Bayers
Loc: New York
Reg: 11-21-04
11-27-17 07:10 AM - Post#238013    
    In response to Mike Porter

The second half box is brutal for everyone save Eddie Scott, but it’s salient that Penn blew a 15-point lead, while shooting 5-27 from the floor, with Woods and Foreman dominating the action, taking the first and second most shots while:

* shooting a combined 2-12 (Woods 2-7, Foreman 0-5)
* committing 4 TOs against only 1 assist (Foreman, Penn’s only assist of the half)
* going 0-3 on FTs (all by Woods, including a missed front end of a 1-and-1 and an 0-2 with Penn up 5 with 1:13 remaining)
* committing a combined 6 fouls

Jelani Williams can’t be ready fast enough.


 
 Page 3 of 6 « First<3456
Icon Legend Permissions Topic Options
Report Post

Quote Post

Quick Reply

Print Topic

Email Topic

5858 Views





Copyright © 2004-2012 Basketball U. Terms of Use for our Site and Privacy Policy are applicable to you. All rights reserved.
Basketball U. and its subsidiaries are not affiliated in any way with any NCAA athletic conference or member institution.
FusionBB™ Version 2.1 | ©2003-2007 InteractivePHP, Inc.
Execution time: 0.239 seconds.   Total Queries: 16   Zlib Compression is on.
All times are (GMT -0500) Eastern. Current time is 01:45 PM
Top