Results 1 to 13 of 13

Thread: A new way judge a team's defense

  1. #1
    Administrator Unclebuck's Avatar
    Join Date
    Jan 2004
    Posts
    32,912

    Default A new way judge a team's defense

    Not much to comment on but I found this interesting


    http://proxy.espn.go.com/nba/columns...ohn&id=1997975






    Orlando's D deserves some credit
    By John Hollinger, ESPN Insider
    John Hollinger Archive

    Quick, who has a better defense, Orlando or Indiana?

    Probably 99 percent of you will say Indiana. Ask a group of NBA beat writers and columnists the same question, and they'd all stump for the Pacers too.

    Even the two teams would agree. The Magic consider themselves to be a terrible defensive team, as general manager Jon Weisbrod recently vented.

    "Our personnel doesn't warrant us being the worst defensive team in the league," he said after a recent loss to Golden State. "I think we need to look elsewhere for where the issues are."

    Meanwhile, Indiana believes its defense is what's keeping it in the playoff chase. On Monday, a newspaper column opined "Indiana is once again one of the league's premier defensive teams, but have underachieved because of an offense stuck in neutral."

    At first glance, the stats seem to support this idea. Orlando ranks a measly 27th in the league in points allowed per game, giving up 100.5, while Indiana is a lofty fifth at 93.4. So that settles it, right? All the conventional wisdom says Indiana is the superior defense.

    But there's one catch: Orlando is just as good.

    The entire argument for Indiana relies on the premise that points allowed per game is a good indicator of a team's defense. In truth, it's about as reliable as your uncle's '83 El Camino.

    Ranking teams by points allowed per game is flawed because of one massive omission that many observers overlook: pace. Some teams play much faster than others, leading to more possessions for each side. For instance, if one team is running and pressing and using a 100 possessions a game, while the other is milking the shot clock and only using 50, it's going to be awfully hard for the second team to outscore the first one.

    Yet the NBA rankings pretend this distinction doesn't exist. It would be like major league baseball ranking pitchers by "runs allowed" with no regard for how many innings they'd pitched.

    Thus, factoring in pace is a critical part of evaluating team defense. I measure pace with a statistic I call Pace Factor, which is the number of possessions a team uses in an average 48-minute game. The tricky part to calculating Pace Factor is dealing with free-throw attempts. If every free-throw attempt came on a two-shot foul, each would be worth half a possession. However, we have to account for those occasions when a player draws a foul and makes the basket, as well as oddities like technical fouls, clear-path fouls, flagrant fouls and lane violations. Based on these factors, the average free throw uses approximately 0.44 possessions.

    With that knowledge in hand, determining Pace Factor is easy. For those of you playing at home, here's the math: Take a team's free-throw attempts and multiply by 0.44, add its field-goal attempts and turnovers, and subtract its offensive rebounds. That's the number of possessions the team used on offense. Do the same for its defensive stats and you have the opponents' possessions. Average the two, divide by minutes played, and multiply by 48. Voila you have the team's Pace Factor.

    Pace Factor: 2004-05 League Leaders
    Team Possessions per Game
    Orlando 97.7
    Phoenix 97.6
    Philadelphia 96.3
    Boston 95.8
    Dallas 95.5
    League Average 93.2


    When it comes to pace, the league has two jackrabbits: Orlando and Phoenix. The Suns' run-and-gun style is well known, but Orlando actually is the fastest-paced team in the league. Orlando's opponents get 97.7 possessions per game, compared to the league average of 93.2, meaning Orlando's opponents get nearly five extra trips per game. Once we consider the pace, it's hardly a revelation that Orlando allows more points than do most teams.

    For a counter-example, look at Indiana. The Pacers are the tortoise to Orlando's hare. Indiana uses just 90.0 possessions per game only Detroit and New Orleans play slower. As a result, the Pacers' opponents have eight possessions a game fewer than Orlando's, making it fairly easy for Indiana to allow fewer points.

    So who's really better, Indiana or Orlando? Due to the differences in pace, it's impossible to make a relevant comparison using points allowed per game. Fortunately, there's an easy way to compare apples to apples. It's a measure that I call Defensive Efficiency, which is the number of points a team gives up for every 100 opponent possessions. Better yet, it's easily calculated once we know a team's Pace Factor. To do so, start with the number of points a team allows per 48 minutes and divide by its Pace Factor. Then multiply the result by 100.

    Using Defensive Efficiency, we can see that Orlando gives up 102.3 points per 100 possessions, while Indiana gives up 102.2. In other words, the Magic actually play defense just as well as the Pacers, despite allowing nearly seven more points per game (and yes, these numbers include Monday night's beatdown at the O-Rena).

    Lies and Statistics
    Team PPG Allowed Rank Def. Efficiency Rank
    Orlando 100.5 27 102.3 13
    Indiana 93.4 5 102.2 12
    League Average 96.7 103.2


    The Magic and the Pacers aren't the only teams whose pros and cons are masked by per-game averages. Take Utah, for instance. The Jazz's 99.1 points allowed per game looks somewhat respectable, ranking 18th in the league. But that ranking is pure fiction. Utah is, in fact, the worst defensive team in captivity, giving up 108.0 points for every 100 opponent possessions. The only thing making them look good is one of the league's slowest Pace Factors at 91.6 possessions a game.

    On the other hand, of the six teams allowing more than 100 points per game, only Toronto has a truly bad defense. Phoenix gives up the most points in the league at 102.3 per game, but it ranks 17th in Defensive Efficiency. Golden State allows 100.4 points per game, just like Orlando, but ranks 18th. Sacramento and Washington are a bit more deficient, ranking 20th and 22nd, respectively, but aren't nearly as bad as their per-game averages suggest.

    But let's get back to the Magic, the most intriguing case of the bunch. What I find so fascinating is they really think their defense is a problem, when in fact it's a slight strength Orlando's Defensive Efficiency is nearly a point better than the league average. It's not an esoteric point, either the Magic made a key in-season trade based on this perceived need when they swapped Cuttino Mobley for Doug Christie.

    Unfortunately, the gaping difference between perception and reality is having a real impact on the team and its front office. Orlando's offense is no great shakes (the other side of the same coin its fast pace also means we need to let some air out of that gaudy scoring average), but the Magic are spending much of their time fretting about defense. They've made deals to shore up that perceived weakness and devoted huge chunks of practice time to it, instead of focusing on an offensive attack that requires at least as much attention.

    Orlando's only crime is thinking that a team's ranking in points allowed per game is an honest appraisal of its defense. Using tools like Pace Factor and Defensive Efficiency, however, we can see that points allowed per game's relevance to team defense is kind of like Chris Andersen in the dunk contest. Sometimes it's in the general area, but it's rarely on target, and more often than not it isn't even close. As a result, the notion that Indiana has a better defense than Orlando isn't such a slam dunk after all.

    John Hollinger is the author of "Pro Basketball Forecast 2004-05." He has joined ESPN Insider as a regular contributor.

  2. #2
    Administrator/ The Real Jay ChicagoJ's Avatar
    Join Date
    Jan 2004
    Location
    Chicago
    Age
    45
    Posts
    17,000

    Sports Logo Sports Logo Sports Logo

    Default Re: A new way judge a team's defense

    Some of us have tried to make the arguement on here that the pace of the game (and over-controlling coaches) has more to do with decreases in scoring than actual defense.

    Opponents FG% is a much better indicator of defensive quality than Opponents PPG.

    More important than any of those stats is whether or not a team can make stops/ get defensive rebounds at crunch time.

    Some of us just know a good (bad) defense when we see it, and it doesn't matter what the stats say. And in this case, it seems to me Orlando gives up more possessions because their opponents can quickly find a good shot. Especially over the past couple of weeks, we've seen Pacers' opponents scrambling to beat the shotclock with any shot, good or not.
    Why do the things that we treasure most, slip away in time
    Till to the music we grow deaf, to God's beauty blind
    Why do the things that connect us slowly pull us apart?
    Till we fall away in our own darkness, a stranger to our own hearts
    And life itself, rushing over me
    Life itself, the wind in black elms,
    Life itself in your heart and in your eyes, I can't make it without you


  3. #3
    Formerly PacerFanInAZ Cactus Jax's Avatar
    Join Date
    Jan 2004
    Location
    Tucson AZ
    Age
    30
    Posts
    4,186
    Mood

    Sports Logo Sports Logo

    Default Re: A new way judge a team's defense

    Did he just watch the last Pacer/Magic game, or not?

  4. #4
    Thanks Kesty! Pig Nash's Avatar
    Join Date
    Feb 2004
    Location
    Indianapolis, IN
    Age
    26
    Posts
    6,478
    Mood

    Sports Logo Sports Logo Sports Logo

    Default Re: A new way judge a team's defense

    Quote Originally Posted by The Article
    This quote is hidden because you are ignoring this member. Show Quote
    Using Defensive Efficiency, we can see that Orlando gives up 102.3 points per 100 possessions, while Indiana gives up 102.2. In other words, the Magic actually play defense just as well as the Pacers, despite allowing nearly seven more points per game (and yes, these numbers include Monday night's beatdown at the O-Rena).
    yes.

  5. #5
    flexible and robust SoupIsGood's Avatar
    Join Date
    Jun 2004
    Location
    Lappy Go Hucky
    Age
    26
    Posts
    17,540

    Default Re: A new way judge a team's defense

    Why make an entire article to explain such a simple concept? I felt like a child being taught how to add and subtract.

    The premise of the entire thing is flawed however. If you slow the pace, doesn't that indicate that the team probably has good transition and half-court defense? Simply milking the clock does not make you the third slowest team in the league. It means other teams are struggling to get easy buckets against you.

    He also really should have picked another team. As the Pacers have been this season, there have been about five different units playing. The pre-brawl unit, the directly post-brawl unit, the post-brawl unit with Reg, Jeff, and others, the Jermaine unit, and then the unit were are fielding now. Some of those lineups played horrible defense, and if on average the Pacers have played equal intensity defense to Orlando this season, that means our standard lineup is significantly better.

    If one is to judge a defense with only stats, FG% allowed, and how often the defense sends someone to the line is probably the way to go.
    You, Never? Did the Kenosha Kid?

  6. #6
    Huge Member heywoode's Avatar
    Join Date
    Nov 2004
    Location
    Indiana
    Age
    46
    Posts
    6,334

    Default Re: A new way judge a team's defense

    I gotta go with Soup on this one, but I loved the insight to a stat angle I had never thought of....I just hope the trend of good defense continues.

    We have some more big tests ahead that could easily let us get some of our swagger back....There are several teams that are good that we owe for a beatdown the last time...I can't wait for the Sonics game.....The soon-post-brawl shot pretty lights out to keep us close but we couldn't hang at the end. I think JJ had his career high of 33 that night?

    We owe them and Golden State too.....

  7. #7
    White and Nerdy Anthem's Avatar
    Join Date
    Jan 2004
    Posts
    23,747

    Default Re: A new way judge a team's defense

    Well, I'm with Jay about FG%. This stat, though, is nutty.

    82games has us at .445 vs. Orlando at .440 for the season, which seems about right. For the season, our stats are similar. With our players coming back, though, we're putting our defense back together. I expect our defensive fg% to be among the league leaders for the remainder of the season.

  8. #8
    Administrator Unclebuck's Avatar
    Join Date
    Jan 2004
    Posts
    32,912

    Default Re: A new way judge a team's defense

    Quote Originally Posted by Jay@Section204
    This quote is hidden because you are ignoring this member. Show Quote

    Opponents FG% is a much better indicator of defensive quality than Opponents PPG.

    More important than any of those stats is whether or not a team can make stops/ get defensive rebounds at crunch time.

    Some of us just know a good (bad) defense when we see it, and it doesn't matter what the stats say. And in this case, it seems to me Orlando gives up more possessions because their opponents can quickly find a good shot. Especially over the past couple of weeks, we've seen Pacers' opponents scrambling to beat the shotclock with any shot, good or not.


    Jay, you and I often disagree, but in this case I agree with you 100%. great ponts.

    I think the most important team stat is defensive FG%, followed by defensive rebounding %. Of course what a team can do in the last 5 minutes of a close game in these two areas is huge.


    As I watch a game if the defense appears to be "suffocating the offense" That is the barometer I use. The Magic allowed the Pacers to do whatever they wanted get any shot they wanted and allowed the Pacers to move the ball at will. The Heat were much more suffocating.

    Another baromerter I use is if it seems like there are 6 defenders on the court. The Pitons last yar and the Bulls in the 90's seems like they had 6 defenders

  9. #9

    Default Re: A new way judge a team's defense

    This is roughly similar to what Dean Smith had always done at North Carolina. He called it points per possession. He used it on offense and defense. If I remember correctly his goal was 1 point per possesion offensively and .75 ppp defensively. His formula for this took in offensive rebounds, 1-1 FT's , everything. As a matter of fact most basketball stats software has this as one of the things that is kept. I believe there are a few pro teams using a software called Cybersports. It is a gametime software. Easy to use if you have a spotter to help you.

  10. #10
    Member
    Join Date
    Jan 2004
    Location
    Frankfort, IN
    Posts
    9,136

    Default Re: A new way judge a team's defense

    I agree with just about everyone.

    In essence, the Pacers defense so far this season hasn't been up to what it was last year in a game-in game-out basis. Part of that's you've turned into a horrible rebounding team, part of it's not having Artest and part of it's the team D just hasn't been up to snuff.

    If I was going to put together a fairly simple formula to calculate team defense I'd be looking at these factors:

    Defensive FG %
    Defensive Rebounding %
    Turnover Margin
    Opponents FTA per FGA

    Between those factors you'd get how many more/fewer possessions your team gets and how well you do defending those possessions.

    Hollinger's formula is OK - it uses most of the above factors. But nothing is a substitute for watching a team's games. And I'd say that if you'd REALLY want to figure it out with stats you'd have to take whatever number you come up with and do a full statistical analysis using standard deviations, validity, etc. If two teams give up an average of 90 ppg, but for one team 68% of their games give up between 87 and 93 while for the other it's between 80 and 100, I'd prefer the first team.

    Unless of course the second team is very young, or has injury problems and everyone will be in for the playoffs, or any of a dozen other factors that numbers alone can't explain.
    The poster formerly known as Rimfire

  11. #11
    Fear my small avatar Gyron's Avatar
    Join Date
    Jan 2004
    Location
    North Carolina
    Posts
    6,366

    Default Re: A new way judge a team's defense

    Man I swear I've seen this before......Wasn't there something very similar to this last season?

  12. #12
    Administrator/ The Real Jay ChicagoJ's Avatar
    Join Date
    Jan 2004
    Location
    Chicago
    Age
    45
    Posts
    17,000

    Sports Logo Sports Logo Sports Logo

    Default Re: A new way judge a team's defense

    Quote Originally Posted by Unclebuck
    This quote is hidden because you are ignoring this member. Show Quote
    Jay, you and I often disagree, but in this case I agree with you 100%. great ponts.

    I think the most important team stat is defensive FG%, followed by defensive rebounding %. Of course what a team can do in the last 5 minutes of a close game in these two areas is huge.


    As I watch a game if the defense appears to be "suffocating the offense" That is the barometer I use. The Magic allowed the Pacers to do whatever they wanted get any shot they wanted and allowed the Pacers to move the ball at will. The Heat were much more suffocating.

    Another baromerter I use is if it seems like there are 6 defenders on the court. The Pitons last yar and the Bulls in the 90's seems like they had 6 defenders
    I was going to add that I thought the late-1990s era Pacers were a great example of a team that, while not excellent at either defense nor rebounding, could always be relied on to get a defensive stop (McKey) and rebound (DD) when it needed one. Even Jalen Rose, Mark Jackson, and Reggie Miller could contribute "clutch" defense on those teams. When we lost a close game, it was either an opponents' offensive player hitting a great, contested shot (and great offensive players can score, no matter how good the defense), or we couldn't convert at our end of the court.
    Why do the things that we treasure most, slip away in time
    Till to the music we grow deaf, to God's beauty blind
    Why do the things that connect us slowly pull us apart?
    Till we fall away in our own darkness, a stranger to our own hearts
    And life itself, rushing over me
    Life itself, the wind in black elms,
    Life itself in your heart and in your eyes, I can't make it without you


  13. #13
    _PD_
    Guest

    Default Re: A new way judge a team's defense

    I mostly agree with the opps FG% as a good barometer of D. And I do like the points scored/possession concept.

    While there is some truth to the posted article, let's not use the pacers, but the colts as the comparison. The colts being a run and gun team must have good D because that D is on the field so much longer. I'd be willing to bet that points allowed per time of possession was pretty low for the colts D, therefore they must be one of the top defenses in the nfl. gaaaaaa

    Ever notice how many more 24 second violations there are during the playoffs? And fewer possessions? And lower scores? Basketball is one of the few, maybe only, sports where allowing your opponent to have more time of possession works in your favor.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •