Defensive Score Sheet-Tampa | The Boneyard

Defensive Score Sheet-Tampa

Status
Not open for further replies.
Joined
Aug 24, 2011
Messages
3,023
Reaction Score
3,684
First off, I want to thank @tcf15 and @gampelcrazies for the uconnhuskygames.com site for the replay to this game and all the work they do. Charting games can sometimes require looking at individual plays up to 5 times, and the interface on that site is very useful and easy to use compared to watchespn and other sites.

Sorry for the long post in advance. I mentioned a few months ago how I began charting last season's defensive possessions. Long story short, I got busy and things fizzled out, and I didn't complete the season. I said I would have the results posted by the start of the season and I didn't, so for that I apologize. After a while it became difficult to brace myself to suffer through essentially re-watching two more losses to a bad Memphis team, among others. If it's any consolation, I've charted all of the 2014 NCAA tournament, and parts of the 2009 NCAA tournament. It is my hope to post the results of each of UConn's final four runs sometime in the future. Again, thanks to uconnhuskygames.com. I hope to chart every game I can this season and post the results here, although obviously sometimes life gets in the way and makes it difficult.

I originally started to think about charting sometime last spring as a way to look at and analyze the game on a closer level. My first thought was to chart games the way Luke Winn of Sports Illustrated did with UConn and Butler. In doing further research, I decided that I preferred Dean Oliver's method in the book Basketball on Paper. Both methods are good in their own ways, but I find Oliver's to be a little better and more universal in a way. Basketball on Paper was written over a decade ago so it's a little dated, but I really liked it. I think anyone that enjoys stats and basketball will like it. FWIW, Winn seems to be a fan of Oliver's method as well. The best link I found to help was the one in my original post, by David Hess, who operated a Kansas blog and was charting their defense about 5 seasons ago(I have no idea if he still does). And then obviously Oliver's book was very useful as well. For reference, here's a link to defensive charting that teamrankings.com did for a UConn-Cincinnati game about 4 years ago.

As far as the process goes for charting the games, the goal(at least for me) is to essentially assign either credit or blame to a player or players for the result of each defensive play. I'll copy and paste Hess' explanations for each of the terms below, but I basically watch each defensive play and then determine whether a given player or players a)forced a turnover, b) forced a missed shot, c) committed a foul that led to free throws, or d) gave up a made basket. Once I have charted a game in its entirety, the calculations begin based on the formulas Oliver created.

Defense in basketball is very much a team oriented ordeal, which Oliver is happy to point out. Narrowing down credit or blame to individuals for each play can be at times foolish, and often times doesn't tell the whole story. For example, defensive charting will not be able to point out bigs that effectively defended a screen and roll the way Ollie wanted to, or guards that rotated to the weak side to close out on a corner three in time to force a worse short. But I think this whole concept was borne out of the fact that many basic defensive statistics do not even come close to telling the whole story, and this at least comes a little bit closer. It is absolutely flawed, and by no means perfect, which I'll elaborate on further below after I post Hess' explanations, and then the results of the Tampa game:

  • Taken from the box score
    • Min – Minutes played
    • DREB – Defensive REBounds
  • Tracked by me
    • FM - Forced field goal Miss – when a defender forces an offensive player to miss a shot from the field. Oliver separates FM from Blocks, but I’ve lumped them together here.
    • FTO – Forced TurnOver – when a defender forces an offensive player to turn the ball over. Again, Oliver separates out Steals, but I’ve combined them, partly because because I don’t know which plays the official scorekeeper would actually count as steals. One thing to note here is that a player who draws an offensive foul is always credited with a FTO, even if it’s just a moving screen.
    • FFTA – Forced missed Free Throw Attempt – missed foul shots resulting from a defender’s foul
    • DFGM – allowed Defensive Field Goal Made – when a defender allows an offensive player to score a field goal over him or by dribbling by him
    • DFTM – allowed Free Throw Made – made free throws resulting from a defender’s foul
  • Calculated Tallies
    • Stops – the credit a defensive player gets for actions that contributed to ending an opponent possession. This isn’t as simple as adding FM + FTO + 0.4*FFTA, because the credit for a missed shot has to be shared with the defensive player who rebounds it. The formula is more complex than you might think, and includes a sliding weight for FM vs. DREB, based on how difficult those actions seem to be in each particular game, so I’ll just refer you to Appendix 3 of Basketball On Paper.
    • ScPos – Scoring Possessions allowed by a player. This is essentially just DFGM plus a FT-related factor. I’ll again refer you to Basketball On Paper for details.
    • DPoss – [Stops + ScPos] – total Defensive Possessions that were credited to (or blamed on) a player.
  • Calculated Metrics
    • Stop% – [Stops/DPoss] – the fraction of an individual player’s credited defensive possessions that ended with 0 points. Essentially the inverse of offensive Floor%.
    • %DPoss – [(Min/40)*DPoss/TeamDefensivePossessions] (for a non-OT game) - the percentage of team defensive possessions faced by an individual defender. Analogous to %Poss on offense.
    • DRtg – [(1–%DPoss)*TeamDRtg + %DPoss*(100*TeamDefPtsPerScPoss*(1-Stop%))] – individual Defensive Rating. Gives a player credit for stops and scoring possessions he was directly involved in, then assumes a nebulous team-average performance in the other possessions. This is the analog of offensive rating.
So without further ado, here are the results, and keep in mind that I am human and my math skills are not the best so I may have very well made some errors in my calculations. I'm also not the most artistic person,so I apologize in advance for how the table looks. I did not post any of the percentages or D rating for the two walk-ons, primarily because they only played a minute each, and also because I did not think it was fair to Noyes, because his numbers were obviously going to be extremely distorted thanks to playing for such a short time. The "team" player there is essentially just there to fill in the gaps for unforced turnovers. Any field goals attributed to the "team" were basically instances where there were fast breaks, and Tampa had numbers and thus it would be unfair to assign blame to just one player. In certain cases, two players contested a shot or forced a turnover, so they received .5 of a FM or FTO. Even though the defensive ratings are bolded, I think stop percentage is slightly more valuable or useful, which I'll explain below.

Defensive Scoresheet-Tampa.png


The most valuable thing you can do on defense in Oliver's estimation is force turnovers. That's what he assigns the most credit for when it comes to stops. That makes sense, as turnovers prevent a shot from being attempted by the opposing team(and thus a potential offensive rebound), and also possibly allow the defense to transition into a fast break which typically yields more points than in a half court setting. That's why Ryan Boatright always had a pretty good stop percentage even though he had a pretty poor percentage of defending threes. He was good for at least three forced turnovers by himself each game, which is a pretty insane number. Depending on the game and the prevalence of offensive rebounds, Oliver typically makes defensive rebounds worth about
2/3rds of what forced field goal misses are worth. I think that may be a little high when it comes to factoring in the value or worth of a defender, but I'm curious to see how you guys feel about that. It's a similar debate to what @James and @mauconnfan (among others) have been arguing about with Brimah. The rebounding coefficient and lack of turnovers that he forces really hurts his number of stops, even though he is great at creating misses on defense. That's why I added the eFG% to the table even though Oliver does not include that, as I think it's useful. Which brings me to another point, which is that Oliver does not factor in threes at all in his calculations. The basketball community at large has placed such a high importance on threes, especially from an offensive standpoint, that it seems silly that none of Oliver's calculations include three pointers. Again, that's why I added eFG% to my table, to perhaps paint a more complete picture. My other problem is that Oliver is very reliant on the overall team defensive rating to compute an individual's defensive rating. At a certain point it actually is worse for your D rating if you're involved in a high percentage of your team's defensive possessions. The extreme case of that is illustrated with Brimah and Gibbs here, as Gibbs has a worse stop percentage but a better D rating, because he was involved in far fewer possessions. Naturally big men are going to be involved in more possessions than guards just due to the way the game is played. FWIW, I think Gibbs played pretty well on defense, not involving yourself in a lot of possessions can often be a very good thing and a sign of good defense and it's something that 'Bazz did very well during the '14 tourney. It also essentially gives you credit for when you foul a guy and he misses a free throw. This is something that Nolan did a lot last year, and it helped his overall numbers. I understand that Oliver wants to show the overall picture for how an individual contributed on defense, but he seems to give too much credit to missed free throws, at least in my opinion.

That's all I can think of for now, please reply below with any comments, suggestions, or questions you have. Like I said, I very well may have made some mistakes along the way.

TL;DR: here's a table of defensive charting I did for the Tampa game based on this link.
 
Last edited:
Joined
Sep 2, 2011
Messages
1,337
Reaction Score
14,440
Very impressive thought process that I will reread and focus on later. Thanks for your effort and contribution which I am sure is a labor of love, but get some sleep, you're posting this at 4am ish.
 
Joined
Aug 28, 2011
Messages
26,914
Reaction Score
65,041
John, since you watched the entire game over and over, and since numbers don't tell the whole story, I would be interested in your opinion of who played good defense and who didn't.
 
Joined
Jan 10, 2014
Messages
571
Reaction Score
1,720
A couple questions.

If I'm reading it correctly a lower D Rating is better than a higher d rating? I'm basing this off Cassell's number being so high when it says he forced .5 misses and had 3 makes.

Also, is there a general benchmark that is used to classify what is an average, good, very good defender. It appears around 100 is average and below 90 is very good?
 
Joined
Nov 3, 2014
Messages
366
Reaction Score
665
Great work. You have an effective patience ratio of about 20:1 greater than I!
 
Joined
Nov 22, 2014
Messages
196
Reaction Score
468
And a couple of comments:
According to the numbers, Cassell was bad and Enoch was great. Watching the game, I thought they both were a little below average. Their numbers are skewed by having few minutes on the floor. I guess you need at least 15 minutes on the floor for your number to have meaning.
I wasn't impressed by Dham's defense, but his rating says he was good. I probably was looking at the floor and not counting Drebs enough.
I think it will be interesting to see what trends develop over the season.
 
Joined
Aug 27, 2011
Messages
16,029
Reaction Score
23,088
Do the team statisticians and coaches do all of this? how far do they go with stats vs this guys playing good?
 
Joined
Aug 24, 2011
Messages
3,023
Reaction Score
3,684
A couple questions.

If I'm reading it correctly a lower D Rating is better than a higher d rating? I'm basing this off Cassell's number being so high when it says he forced .5 misses and had 3 makes.

Also, is there a general benchmark that is used to classify what is an average, good, very good defender. It appears around 100 is average and below 90 is very good?

Yes, you want your D rating to be as low as possible. 100 is average in the NBA. I'm sure in college it's a bit lower because teams generally don't score as much on a per-possession basis, but I'm not sure what the benchmark would be. Anything in the 80's is really, really good. And anything in the low 90's is good as well.
 
Joined
Aug 24, 2011
Messages
3,023
Reaction Score
3,684
And a couple of comments:
According to the numbers, Cassell was bad and Enoch was great. Watching the game, I thought they both were a little below average. Their numbers are skewed by having few minutes on the floor. I guess you need at least 15 minutes on the floor for your number to have meaning.
I wasn't impressed by Dham's defense, but his rating says he was good. I probably was looking at the floor and not counting Drebs enough.
I think it will be interesting to see what trends develop over the season.

You're completely right about the low number of minutes and the low number of possessions. It's a very small sample size. That's why I hope to do this for every game so that some sort of consensus forms for each player. I think that Cassell is overall a below average defender and obviously it's too early to tell on Enoch, but I initially thought he wasn't that good on defense on Saturday, but in watching it again I thought he looked pretty good. He only played 9 minutes and a lot of that was in garbage time at the end against Tampa's backup bigs. But he positions himself pretty well and he recovers well, too. He's a little erratic, but I liked what I saw on a second viewing.

Hamilton is always an interesting case on defense because he's such a good defensive rebounder, which is always going to help his numbers for something like this. He's also pretty good on the ball I would say, he doesn't lose his guy often and he contests shots pretty well. He's pretty shaky off the ball, especially in inbounds situations. That's how he gave up at least one of his made field goals against Tampa.
 
Joined
Aug 24, 2011
Messages
3,023
Reaction Score
3,684
John, since you watched the entire game over and over, and since numbers don't tell the whole story, I would be interested in your opinion of who played good defense and who didn't.

I thought Miller was very good on defense throughout. There were several times where a Tampa player caught it on the block or faced him up with the intention to score and they had to resort to a harmless pass because Miller stopped them. He looks like he'll be the rock of the defense if last Sunday's game is any indication. I also thought Gibbs played pretty good defense to be honest. He committed a couple silly fouls for a fifth year senior, but he clearly understands how to defend ball screens and the importance of limiting penetration for opposing guards. And I thought overall Brimah played well on defense, too, but rebounding is still a big problem for him, and his numbers aren't going to be be any better this year if he continues to log a lot of minutes with Hamilton and Miller on the floor with him at the same time. But he is so good at contesting shots that he is very valuable on defense. I also thought Calhoun looked pretty good.

I think Cassell struggled a bit and Purvis needs to be better if we're going to take the next step, although if Purvis is playing well on offense you live with his defense. Cassell seems to either lose concentration sometimes or he just isn't great at contesting jump shooters.
 
Status
Not open for further replies.

Online statistics

Members online
559
Guests online
3,404
Total visitors
3,963

Forum statistics

Threads
155,762
Messages
4,030,789
Members
9,863
Latest member
leepaul


Top Bottom