I used the raw data over at dougstats.com to compare the defenses of each NBA team so far this season. The results were fairly surprising, and counter to the media preconceptions of which teams are great defenders, and which teams suck at defense. The media generally looks at two stats when commenting on a team's defensive performance: points allowed per game, and opponent's field goal percentage. Here are how the teams rank, based on these categories: <font face=""Courier New"">Points allowed -------------- 1 MemphisGrizzlies 84.85 2 HoustonRockets 88.74 3 MinnesotaT-wolves 88.83 4 IndianaPacers 89.53 5 SanAntonioSpurs 89.90 6 OrlandoMagic 90.37 7 UtahJazz 91.50 8 DetroitPistons 92.00 9 DallasMavericks 92.45 10 MiamiHeat 93.90 11 NYKnicks 94.79 12 LAClippers 95.00 13 NOrleansHornets 95.05 14 LALakers 95.45 15 GSWarriors 95.55 16 PortlandTrailBlazers 95.90 17 DenverNuggets 96.05 18 NJNets 96.32 19 PhoenixSuns 96.95 20 ChicagoBulls 97.05 21 AtlantaHawks 98.32 22 ClevelandCavaliers 98.53 23 SacramentoKings 99.19 24 CharlotteBobcats 99.33 25 BostonCeltics 99.70 26 MilwaukeeBucks 99.72 27 WashingtonWizards 100.32 28 Philadelphia76ers 102.33 29 TorontoRaptors 103.57 30 SeattleSupersonics 104.79 </font> <font face=""Courier New"">FG% allowed ------------ 1 LAClippers 41.10% 2 SanAntonioSpurs 42.18% 3 HoustonRockets 42.20% 4 IndianaPacers 42.34% 5 MinnesotaT-wolves 42.51% 6 MemphisGrizzlies 42.64% 7 NJNets 42.91% 8 PhoenixSuns 42.95% 9 UtahJazz 43.28% 10 MiamiHeat 43.42% 11 NYKnicks 43.42% 12 GSWarriors 43.64% 13 DenverNuggets 43.65% 14 ChicagoBulls 44.03% 15 DetroitPistons 44.19% 16 DallasMavericks 44.23% 17 LALakers 44.65% 18 OrlandoMagic 44.88% 19 WashingtonWizards 44.94% 20 NOrleansHornets 45.02% 21 BostonCeltics 45.48% 22 PortlandTrailBlazers 45.65% 23 MilwaukeeBucks 45.85% 24 Philadelphia76ers 45.98% 25 CharlotteBobcats 46.04% 26 SacramentoKings 46.38% 27 ClevelandCavaliers 47.05% 28 AtlantaHawks 47.44% 29 SeattleSupersonics 48.84% 30 TorontoRaptors 48.99%</font> Both of these stats only give a partial understanding of how effective a team's defense is. Points allowed per game doesn't account for pace, while field goal percentage allowed doesn't account for three pointers and free throws. The best figure to look at is, ideally, is points allowed per possession. Unfortunately, for some bizarre reason the NBA doesn't keep official tallies of possessions in a game. But, this can be estimated as: possessions = fg attempts + turnovers - off. rebounds + 0.44*freethrows attempts The last term estimates the number of possessions that end with free throws. Unfortunately, dougstats doesn't include team turnovers in its raw data, so that will further skew the final value. Still, we get a good idea of which teams are the "slowest", and which teams are the "fastest", even if the actual figure of possessions per game is one or two off. <font face=""Courier New"">Possessions/game (for opponent) ------------------------------- 1 MemphisGrizzlies 87.61 2 PortlandTrailBlazers 88.75 3 HoustonRockets 88.99 4 OrlandoMagic 89.18 5 UtahJazz 90.12 6 DetroitPistons 90.23 7 MinnesotaT-wolves 90.66 8 NOrleansHornets 91.50 9 DallasMavericks 91.56 10 NJNets 91.88 11 IndianaPacers 92.27 12 SeattleSupersonics 93.02 13 AtlantaHawks 93.29 14 MiamiHeat 93.75 15 SanAntonioSpurs 93.76 16 ClevelandCavaliers 93.95 17 NYKnicks 94.05 18 LALakers 94.20 19 MilwaukeeBucks 94.34 20 LAClippers 94.83 21 SacramentoKings 94.85 22 TorontoRaptors 94.98 23 BostonCeltics 95.56 24 ChicagoBulls 95.66 25 GSWarriors 96.05 26 WashingtonWizards 97.08 27 DenverNuggets 97.11 28 CharlotteBobcats 97.38 29 Philadelphia76ers 97.51 30 PhoenixSuns 100.16 </font> And, finally, here's the final stat we're looking for. Points allowed, per 100 possessions (I multiply by 100, since they're roughly 100 possession in a game). <font face=""Courier New"">Points allowed per 100 possessions --------------------------------- 1 SanAntonioSpurs 95.89 2 PhoenixSuns 96.79 3 MemphisGrizzlies 96.85 4 IndianaPacers 97.02 5 MinnesotaT-wolves 97.98 6 DenverNuggets 98.90 7 GSWarriors 99.48 8 HoustonRockets 99.71 9 MiamiHeat 100.16 10 LAClippers 100.18 11 NYKnicks 100.79 12 DallasMavericks 100.97 13 LALakers 101.33 14 OrlandoMagic 101.34 15 ChicagoBulls 101.45 16 UtahJazz 101.53 17 DetroitPistons 101.97 18 CharlotteBobcats 102.00 19 WashingtonWizards 103.33 20 NOrleansHornets 103.88 21 BostonCeltics 104.33 22 SacramentoKings 104.58 23 NJNets 104.82 24 ClevelandCavaliers 104.87 25 Philadelphia76ers 104.95 26 AtlantaHawks 105.38 27 MilwaukeeBucks 105.71 28 PortlandTrailBlazers 108.06 29 TorontoRaptors 109.05 30 SeattleSupersonics 112.65 </font> Not surprisingly, San Antonio is at the top. But here are some other observations that the are of surprise: - Phoenix is the second best defensive team in the league so far this season. If you just look at them by points given up, they're 19th. But adjusting for pace, they become one of the best defensive teams in the league. - Golden State and Dallas are solid, above average defensive teams, despite the reputation of being gunning, offense-only teams - Detroit has actually been a below-average defensive team so far under Flip Saunders. So why do they have the best record? They've been the best offensive team (by a surprisingly wide margin) in terms of scoring per possession. - Chicago has only been average on defense after being the second best defensive team last year - The Lakers, under Phil Jackson, have improved to being average on defense after they were one of the worst defensive teams last year
<div class="quote_poster">Quoting durvasa:</div><div class="quote_post"> - Phoenix is the second best defensive team in the league so far this season. If you just look at them by points given up, they're 19th. But adjusting for pace, they become one of the best defensive teams in the league. </div> this blows me away. definitely shows you what a little "stat manipulation" can do. i'll be reading the fine print a little more closely. anyway, interesting stuff. thanks for sharing it.
Very interesting stuff, durvasa. Good job finding it. Some of these charts definitely do show how misleading stats can be, especially when you're talking about defensive stats. For me, seeing the Bucks in the middle of the pack on most of these defensive statistics is laughable-they're a terrible defensive team. Just going by stats won't give you a true measure of a great defense. When you look at stats, you don't see thing accounted for such as who the teams' opposition has been (a team might be playing great D on paper, but what if they've played some real cupcakes so far), or if opponents have had injuries to star players (the Rockets have probably made teams look great defensively...without their #1 scorer) and so on. I think that while you can use stats more to determine who has a top offensive team, it's very tough to quantify great defense. That's something you get more from actually seeing a team play. It's tough to actually see a suitable cross section of every team's defense all year, but it's about the only way you can really tell who's the best on defense. If you asked me, I'd say that the top five defensive teams are the Spurs, Grizzlies, T'Wolves, Pacers, and surprisingly, the Magic. Kudos to them for upping the defensive intensity. I think they're poised for a run when Grant Hill comes back, assuming they keep it up on the defensive end. Cool stuff man, thanks for posting it.
Stats like these aren't everything. F For example the Pistons don't play lock down defense until the 3rd and 4th quarters and play a faster pace offense through most of the games. You will find that the Pistons is still very strong but they tend to let teams score early and shut them down when It matters. Personal wise I don't think there is much debate that the Spurs and Pistons are far away ahead of the rest of the league on the defensive end. May not be showing It by the stats but in reality they are still the top two teams.
<div class="quote_poster">Quoting AllNet:</div><div class="quote_post">Stats like these aren't everything. F For example the Pistons don't play lock down defense until the 3rd and 4th quarters and play a faster pace offense through most of the games. You will find that the Pistons is still very strong but they tend to let teams score early and shut them down when It matters. Personal wise I don't think there is much debate that the Spurs and Pistons are far away ahead of the rest of the league on the defensive end. May not be showing It by the stats but in reality they are still the top two teams.</div> I wouldn't say that the Pistons are far and away at the top of the league, but they're definitely up there. Like I was saying about watching a team, not looking at their stats, Detroit has that innate ability to know when they need a stop, and to actually get it. On paper they're not the best defensive team, but they sure are crafty, and when it comes down to it, they're right there with the best of them.
<div class="quote_poster">Quoting Schaddy:</div><div class="quote_post"> Just going by stats won't give you a true measure of a great defense. When you look at stats, you don't see thing accounted for such as who the teams' opposition has been (a team might be playing great D on paper, but what if they've played some real cupcakes so far), or if opponents have had injuries to star players (the Rockets have probably made teams look great defensively...without their #1 scorer) and so on. </div> You're right in that stats don't do a good job in accounting for opposing competition. Over the course of a season, this generally doesn't factor in very heavily, since the competition level evens out for all the teams. Early in the season, this /qucan certainly skew results. That's not to say that it can't be accounted for. I've seen NBA statistical ratings that do account for competition level. But it requires some fairly sophisticated methods I don't have much knowledge of. <div class="quote_poster">Quote:</div><div class="quote_post">I think that while you can use stats more to determine who has a top offensive team, it's very tough to quantify great defense. That's something you get more from actually seeing a team play. It's tough to actually see a suitable cross section of every team's defense all year, but it's about the only way you can really tell who's the best on defense.</div> It seems to me that the argument can be used equally for judging defenses and offenses. Clearly, a team who only faces poor defensive teams will look much better offensively. <div class="quote_poster">Quote:</div><div class="quote_post">If you asked me, I'd say that the top five defensive teams are the Spurs, Grizzlies, T'Wolves, Pacers, and surprisingly, the Magic. Kudos to them for upping the defensive intensity. I think they're poised for a run when Grant Hill comes back, assuming they keep it up on the defensive end.</div> I haven't got to watch the Magic this year. What have they done to make you think they're an improved defensive team? Statistically, they are about where they were last year (in terms of points given up per possession). The big difference is they were a very faced-paced team last year (third behind Phoenix and Philadelphia), and this year they are one of the slowest.
<div class="quote_poster">Quoting "AllNet":</div><div class="quote_post"> Stats like these aren't everything. F For example the Pistons don't play lock down defense until the 3rd and 4th quarters and play a faster pace offense through most of the games.</div> But after accounting for pace, the Pistons are a below average defensive team. Actually, the Pistons play at a fairly slow pace, so they look pretty good defensively if all you look at is points per game given up. <div class="quote_poster">Quote:</div><div class="quote_post">You will find that the Pistons is still very strong but they tend to let teams score early and shut them down when It matters.</div> I think a coach would cringe if that's the case. Every point matters, regardless of whether it's the first quarter or fourth quarter. Great defensive teams play consistently good defense. I am curious if the Pistons really do play better defense in cruch-time situations. It would be difficult to verify without going through the play by plays, unfortunately. <div class="quote_poster">Quote:</div><div class="quote_post">Personal wise I don't think there is much debate that the Spurs and Pistons are far away ahead of the rest of the league on the defensive end. May not be showing It by the stats but in reality they are still the top two teams.</div> Alright. What the stats do show is that the Pistons have been far and away the best offensive team in the league, while being pretty average on defense. All their key defensive stoppers from last year are still on the team, so I wonder why there's such a huge dropoff. Last year, they were third best in points allowed per possession, and just below average in offense. This year, it's reversed. Something is clearly different in the way they're playing defense. I'll use Knickerblogger's Stat Page for some more insight. There are four general factors to a team's success (in order of priority) -- shooting from the field, turnovers, offensive rebounding, and free throws. A good defensive teams should limit their opponents ability to be successful in those categories. This is how the Pistons ranked in each those categories last year: 2004-2005: <font face=""Courier New"">opponent's shooting: 5/30 opponent's turnovers: 18/30 opponent's off. rebounding: 5/30 opponent's free throws (ftm/fga): 3/30</font> And so far this season: <font face=""Courier New"">opponent's shooting: 8/30 opponent's turnovers: 23/30 opponent's off. rebounding: 26/30 opponent's free throws (ftm/fga): 5/30 </font> This biggest dropoff defensively for the Pistons, so far, has been defensive rebounding (i.e., limiting their opponents offensive rebounding). They were one of the best last year, and this year they're one of the worst. There's a general dropoff in every other category as well, but rebounding is the biggest difference.
<div class="quote_poster">Quoting durvasa:</div><div class="quote_post"> I haven't got to watch the Magic this year. What have they done to make you think they're an improved defensive team? Statistically, they are about where they were last year (in terms of points given up per possession). The big difference is they were a very faced-paced team last year (third behind Phoenix and Philadelphia), and this year they are one of the slowest.</div> From what I've seen of the Magic this year (which is not a lot, to be honest ), they're working the tempo of the game better, slowing it down, and doing a better job of hustle defense-getting around screens, forcing tough shots. I think they're also making better use of the defensive weapons they have-forcing guys into the lane against Howard, etc. They're not a great, great defensive team, but I had to mention them because they've improved dramatically without changing a lot of their players. As far as comparing offensive/defensive stats, I just feel like you can get a better opinion of how efficient and proficient a team is offensively from their stats. You can tell if they shoot very well from 3, or if they score in the paint, etc. Defensive stats, IMO can be skewed more by opponents than can offensive stats. That's just the way I see it, though. I might be crazy
Very good analysis. Pts allowed per 100 possessions is a very good grading criteria. Its still early in the season, we need to look at it sometime during the all star break to see if this ranking still holds true, especially Phoenix. It'll also be interesting to look at Pts scored per 100 possessions and see who has the best points differential.
<div class="quote_poster">Quoting dallasdude:</div><div class="quote_post">Very good analysis. Pts allowed per 100 possessions is a very good grading criteria. Its still early in the season, we need to look at it sometime during the all star break to see if this ranking still holds true, especially Phoenix. It'll also be interesting to look at Pts scored per 100 possessions and see who has the best points differential.</div> Yep. In theory, that should give us an approximate rank of best to worst teams overall. Again, it doesn't account for competition level, or "clutch"-ness, but over the course of a season it's typically the best indicator of who's the best and who's the worst (at least better than wins/losses). Here's what my spreadsheet tells me. I'll stress again that my possession figures are slightly off (a little higher than they should be), so this isn't an exact figure but the ordering should more or less be accurate. <font face=""Courier New"">Rank, Team, Off, Def, Diff -------------------------- 1 SanAntonioSpurs 105.12 95.89 9.23 2 PhoenixSuns 105.35 96.79 8.56 3 DetroitPistons 109.90 101.97 7.93 4 DallasMavericks 108.03 100.97 7.06 5 MinnesotaT-wolves 104.29 97.98 6.31 6 IndianaPacers 102.47 97.02 5.44 7 ClevelandCavaliers 109.33 104.87 4.46 8 MemphisGrizzlies 101.29 96.85 4.44 9 LAClippers 104.32 100.18 4.15 10 MiamiHeat 103.17 100.16 3.01 11 GSWarriors 101.54 99.48 2.06 12 Philadelphia76ers 106.47 104.95 1.52 13 LALakers 101.87 101.33 0.54 14 DenverNuggets 99.41 98.90 0.50 15 SacramentoKings 104.77 104.58 0.19 16 WashingtonWizards 103.04 103.33 -0.30 17 HoustonRockets 98.30 99.71 -1.42 18 NJNets 103.19 104.82 -1.63 19 BostonCeltics 102.12 104.33 -2.22 20 CharlotteBobcats 99.63 102.00 -2.38 21 MilwaukeeBucks 103.04 105.71 -2.67 22 ChicagoBulls 98.56 101.45 -2.89 23 NYKnicks 96.72 100.79 -4.06 24 UtahJazz 97.12 101.53 -4.41 25 NOrleansHornets 99.41 103.88 -4.46 26 OrlandoMagic 96.86 101.34 -4.48 27 SeattleSupersonics 106.99 112.65 -5.67 28 TorontoRaptors 101.86 109.05 -7.19 29 AtlantaHawks 98.02 105.38 -7.36 30 PortlandTrailBlazers 95.64 108.06 -12.42 </font>
<div class="quote_poster">Quoting "durvasa":</div><div class="quote_post"> This biggest dropoff defensively for the Pistons, so far, has been defensive rebounding (i.e., limiting their opponents offensive rebounding). They were one of the best last year, and this year they're one of the worst. There's a general dropoff in every other category as well, but rebounding is the biggest difference.</div> To follow up, it appears that Rasheed Wallace's dropoff in defensive rebounding is the key factor. - Last year, he grabbed 17.9% of potential defensive rebounds. So far this year, it's down to 15%. - With him on the court this season, the Piston's defensive rebounding is only 66.4% (well below average). With him off the court, it's 71.9%. - Comparing the starting unit's overall rebounding this year versus last year, it's only 47% this year while it was 50.5% last year. But the rebounding rate of the starters with Rasheed replaced by McDyess is exactly the same this year as last year
To quote Mark Twain, "There are lies, damned lies and statistics." That said, the Suns defense is fairly underrated. It's lightyears better than last season's and the team is still learning to play together (and not just offensively).
Good related article on CNNSI from contributer Kevin Pelton (from 82games.com). He looks at several different team's defense this year and talks about how the pace teams play at can mask how good (or bad) their defense actually is. http://sportsillustrated.cnn.com/2005/writ...ctor/index.html
