It's time for another one of my analysis threads. Can't go too in depth for now cause it's finals week, but I hope to expand on this analysis once I'm out of school. Previous analysis threads: Analysis of Floor Positioning Analysis of 3pt shooting Analysis of Shane Battier Analysis of Transition Defense Everyone here and in the media raves on and on about the Rockets defense, myself included. After watching the Minnesota game, I was a litte surprised at how close the game was at half time. I know the Rockets made a lot of mistakes and missed shots, but they looked pretty darn good overall in the 1st half, especially on defense. I did however notice that the T-Wolves were getting a lot of points in transition. So I went back and tracked the game to see if my hunch was correct. I broke down the T-Wolves possessions into 2 categories: -10 seconds or less -11 seconds or more I did not include end of quarter heaves, technical freethrows, nor intentional foul free throws at the end of games because I don't think the Rockets Defense should be statistically penalized or rewarded in these situations. I also did not include possessions where Minnesota would get offensive rebounds and quickly put up another shot, for that is a result of the Rockets half court defense even though the action takes place with less than 10 seconds spent off the shot clock. What I'm trying to get here is a measure of the Rockets transition defense, secondary transition defense, and defense vs early shots taken out of non set plays. It is my theory that while the Rockets have one of the best half court defenses around, their transition may be suffering. I only have time to do one game now, but will do more later. Conditional 10 or fewer seconds possessions Mike James: 5/6 fg (2/2 treys), 1 fta, 13 points Ricky Davis: 0/1 fg, 4 fta, 4 points Tre Hassell: 1/2 fg, 2 points Ke Garnett: 0/0 fg, 2 fta, 2 points Mar. Blount: 0/0 fg, 1 turnover traveling Tr. Hudson: 1/2 fg (1/1 treys), 3pts, 1 turnover passing Marco Jaric: 0/0 fg, 2 fta, 2 points Craig Smith: 1/1 fg, 2 points Randy Foy: 0/0 fg, 1 turnover passing Totals: 19 possessions, 8/12 fg 66.7%, 9 fta, 3 turnovers, 28 points Conditional 11 or more seconds possessions Mike James: 3-8 fg (0-3 treys), 2 turnovers, 6 points Ricky Davis: 1-6 fg (0-1 treys), 2 fta, 4 turnovers, 2 points Tre Hassell: 2-4 fg, 1 fta, 1 turnover, 4 points K. Garnett: 7-12 fg, 6 fta, 19 points Ma. Blount: 6-13 fg, 3 turnovers, 12 points Tr. Hudson: 3-8 fg (0-2 trey), 6 points Marco Jaric: 1-3 fg, 2 turnovers, 2 points Craig Smith: 4-6 fg, 8 points Randy Foye: 0-2 fg, 0 points Eddie Griffin: 0-2 fg, 0 points Totals: 80 possessions, 27/64 fg 42.2%, 9 fta, 12 turnovers, 59 points Conclusions Well, I guess my hunch was correct for this game. The Rockets played great defense in the half court, when everyone was set. But when the Wolves were able to get shots off before the Rockets had all 5 guys back and ready, the Wolves were very efficient and successful. 28 points in 19 possessions vs 59 points in 80 possessions is very telling. In the half court, KG did his thing, Blount hit some jumpers, and Smith scored on Jho and Hayes. But there really wasn't that much success from anyone else though. Even though James hit a couple of clutch drives late, he missed many more. Mike James' real success came from fast breaks and early shots not in the Wolves' sets. That's not saying James can't play halfcourt ball, for I know he can, but in this game, the Rockets half court D was pretty stiffling on pretty much everyone including James. We knew they couldn't hold KG, and we knew Blount would be shooting jumpers to draw Yao away. The Smith production is was a dissapointing surprise though, even though he has been playing well, the Rockets could have defensed him better. So while offensively the Rockets were terrible in many aspects this game, their defense is far from perfect...still great though in my opinion. Once i'm out of school and working for an NBA team I can provide more stats and analysis on these. Feel free to help me out though if this thread appeals to any of y'all. Go back and track previous games and keep the conditions in mind. So what can the Rockets do? I think 1st and foremost, they need to cut down on turnovers. Ahem, Yao, T-Mac, Jho, Rafer, Span, and general spacing guys. Most of the 10 seconds of less possessions come from turnovers. Secondly, Yao and the rest of the team may be a little fatigued on the 2nd night of a back to back, which could have caused poor transition D. So perhaps Van Gundy should spread out the minutes more. Thirdly, the Wolves took a lot of these shots without a number advantage. So I guess they knew that since they were being crappy in the half court, guys should try to put up shots fast 3 on 3, or even 4 on 4. Lastly, the Rockets just need to practice more and play better in terms of getting back. They failed to matchup on many occasions leaving early open jumpers.
I like your analyses and the thought here. But I think a couple of important pieces missing from this are: 1) The NBA min/max/avg for FG% of 10 secs or less. I know you said you'd do more games later.. but other teams are important also in this case. Because it's quite possible that everyone suffers from giving up a high FG% below a certain # of secs purely b/c of the way the game is played.. and that may just be something you have to deal with, and something that naturally increases FG%'s and always has. 2) The avg proportion of possessions of 10 secs or less (compared to the total # of posessions). If this game had an inordinate # of those, then the problem was an anomaly. If you do more Rockets games, you'll be able to figure this one out while you're seeing if this is a continuing trend or not.
Also..it'd be cool if you could somehow track the actual # of secs that ends up being the defining point for all NBA teams. That would take a hell of an analysis.. but I think you know what I mean. Basically tracking 5, 6, 7, 8, etc. through maybe 13 or 14...all as the breaking point as you have used 10 in this case. And then seeing which one gives the biggest difference between above and below. But yeah, too much of a pain there unless it's your job.. ask Morey
I think the key to a good transition defense isn't so much reducing the efficiency on shots taken in transition, but rather reducing the transition opportunities for the other team. We allowed 11 transition possessions for the Wolves. Is that above average?
I have a defensive question. Since I can't start threads Kim's awesome transition defense analysis seems like a good place to ask this. I am a "noob" to the NBA and how the rules are different from college ball. My question stems from the lack of movement on defense. If a team, like the Warriors, are playing a zone, it seems that they park in the middle (which typically defines the reasoning behind playing a zone) but don't get the "defensive 3 secs" penalty. The same is true of teams, like Minnesota, who are double and triple teaming Yao. They are in the paint around him for more than three secounds but again this is not a penalty. But if Yao or Chuck stick for any amount of time in the paint on the man to man help defense they get it called against them. What exactly is the rules regarding what you can and cannot do on defense in the NBA?
Thing is, with that unit on the floor, not many of them are speed guys. Guys like Battier, Yao, Hayes, they all are strictly set-defense guys. And then you have Rafer, who can't do much on defense except gamble and scramble around. So that leaves basically Tmac alone who is capable of running back and disrupting shots or breaking up fastbreaks with steals. But one person alone can't cut it. To play good transition D, you need a team effort, and at least 3 guys who can get back fast and disrupt the fastbreak flow; either with steals or weakside blocks.
Do you really need those numbers to figure out what went wrong last night? Numbers work if you're going to analyze last year's performance, you'll see a trend there. But for last night?
I believe Kim said that eventually the analysis would be done for all the games, not just last night. *Good luck with your schoolwork by the way Kim.* It's a solid theory. In practice your defense works against your own offense, correct? How often does our team run a fast break offense? Hence the lack of concentration in practice on a fast break defense. More games added to the mix will definately let us know if this is a one time thing (which I doubt sincerely) or its a trend.
Fast break points are only a part of the theory, so here they are for the Wester Conference top teams. I wish it included ft attempts due to fast breaks. Fast Break Points per Game: Ranked by Defense 1) Phoenix ---------------- 4) Spurs Offense 17.6 ---------------Offense 09.5 Defense 11.4 ---------------Defense 11.9 10.6% of total pts allowed --13.0% of total points allowed 2) Lakers ------------------ 5) Mavericks Offense 08.3 ---------------Offense 12.1 Defense 10.7 ---------------Defense 12.9 10.9% of total pts allowed--13.7% of total points allowed 3) Houston ----------------- 6) Jazz Offense 06.3 ---------------Offense 09.9 Defense 10.1 ---------------Defense 13.9 11.5% of total pts allowed--14.0% of total pts allowed So it looks like the Rockets are decent amonst contenders in defending against fast break points. They're terrible at scoring on the fast break though, but that's for a separate thread. Among Western conference contenders it looks like only Phoenix is good at fast breaking, and surprisingly they are good at defending the fast break. All the other teams have a fast break pts deficit. Phoenix gives up a lot of points, but the % of points given up are mostly non fast break points, a higher % than all other contenders. 82Games.com tracks offensive and defensive points divided up by shot clock use. This isn't exactly what I'm trying to accomplish because the stats don't filter out end of quarter heaves and end of game intentional fouling. They also don't separate put backs and offensive rebound possessions that lead to 10second or less scores, but I believe it should be separated because that's different from transition defense. And they only use effective FG% (combining 2's and 3's), instead of showing 2pt and 3pt percentages. 82games also doesn't account for when opposing teams go to the free throw line. Tracking when free throws are taken and when turnovers occur should be down in my opinion. But here they are anyways. Defensive Points Allowed Broken Down by Shot Clock Usage: Ranked by best defenses by eFG% under 10 seconds 1) San Antonio 0-10 seconds: 30.7 ppg, 37% of shots taken, eFG% .515 11-24 seconds: 45.0 ppg, 63% of shots taken Total: FG% .446, eFG% .476, ppg 75.7 (free throws not included) 2) Utah 0-10 seconds: 31.6 ppg, 40% of shots taken, eFG% .516 11-24 seconds: 43.2 ppg, 60% of shots taken Total: FG% .446, eFG% .491, ppg 74.8 (free throws not included) 3) Phoenix 0-10 seconds: 35.1 ppg, 36% of shots taken, eFG% .521 11-24 seconds: 57.3 ppg, 64% of shots taken Total: FG% .470, eFG% .498, ppg 92.4 (free throws not included) 4) LA Lakers 0-10 seconds: 35.3 ppg, 41% of shots taken, eFG% .530 11-24 seconds: 44.5 ppg, 59% of shots taken Total: FG% .461, eFG% 498, ppg 79.8 (free throws not included) 5) Houston 0-10 seconds: 30.0 ppg, 36% of shots taken, eFG% .547 11-24 seconds: 39.4 ppg, 64% of shots taken Total: FG% .419, eFG% .457, ppg 69.4 (free throws not included) 6) Dallas 0-10 seconds: 33.2 ppg, 37% of shots taken, eFG% .553 11-24 seconds: 45.3 ppg, 63% of shots taken Total: FG% .458, eFG% 485, ppg 78.5 (free throws not included)
So again, I don't like 82Games.com stats that much because they don't track games within context. I'm trying to judge the Rockets early shot clock defense. I don't want to include offensive rebounds possessions because the defense is already down court, but 82games.com includes that. I also want to filter out intentional fouls and end of quarter heaves. I also want to track when free throws are taken and when turnovers occur. Only then, can one get a complete picture of the defense. But anyhow, the data is still nice to have, and even though flawed, can still be used to make some tenative conclusions. The Rockets statistically are considered to have the best defense in the NBA. It isn't the points per game stat that most experts use to back this claim because the Rockets play at a slower pace. It's about freethrow attempts per possession, field goal percentage, and 3pt field goal percentage. And judging by those stats the Rockets have a great defense. And so the fast break stats may be a little deceiving too, for there are fewer fast break opportunities in Rockets games because there are fewer possessions. So back to my original theory and point of this thread, which is defense under 10 seconds. In looking at the 82games.com stats, which again aren't perfect, I notice that the Rockets are not very good compared to the other contenders of the West in 10second defense. Everyone seems to give up the same % of shots taken in 10seconds or less. This means lets say your opponent gets 100 shots in a particular game. Rockets opponents shoot the ball in 10 seconds or less 36 times. San Antonio opponents do it 37 times, and Utah 40. But the main variation comes in eFG%. The Rockets eFG% Defense in 10 seconds or less is .547. That's only better than Dallas amongst the 6 contenders I have reported on. Now, take a look at the total eFG%s. The Rockets are clearly the best amongst contenders. So what does this mean? Well, for me, it means that if Rockets opponents take their shots after 10 seconds have been used on the shot clock, then those opponents are screwed. The strength of the Rockets defense seems to be in a set, prepared, ready half court with Yao cloggin the middle, Battier and Hayes sliding like crazy for charges, and Mac and Rafer running around to cover shooters. I think this overall stat view negates my back to back theory of fatigue that I pondered in the first post of this thread. The Wolves scored 9 fast break points this last game, which is on par with what the Rockets have been giving up. From 82games.com stats, it looks like 0-10 second defense is the Rockets' biggest defensive weakness. If it's not fatigue, then maybe it's just not finding the open man on 2ndary transition. It's like eveyone runs back to try to cover layups and then someone is open for a 3 or something. Maybe it's a mental thing. The point is, overall the Rockets D is the best. But they're even better when their D is totally set against half court set plays, for half court set plays usually take more than 10 seconds, considering the time it takes to get up court. But when it comes to fast break shooting %, 2ndary fast break shooting %, and just early shots taken before the D is set,....when it comes to those things, the Rockets are below average. Once the Rockets learn how to fix this problem, or if they do, then they could become the best defensive team of all time, statistically speaking. Again, I'll come back at a later time to update the stats that 82games.com doesn't account for.
1) It seems like amongst the top West teams, 10 second or less shots are the hardest to defend, because according to the 82games.com formula, they include offensive putbacks, fast break shots, and the like. And the percentages are higher for early shots against any Defense, Rockets or otherwise, but the Rockets are poorer at stopping it than most other Western contenders. 2) Sifting through the 82games.com site, I found the answer to this. I'm not going to sort it all out yet, cause no time...but amongst the Western top teams, the Rockets opponents shooting 36% of their total attempts in 10 seconds or less is pretty low. So generally the Rockets D does a good job in making the other team eat up clock before they shoot, and that's a good thing. And the Minny game was actually a better ratio: 99 possessions and only 19 with actions occuring in 10 seconds or less. If you're only going by the limited stat of Field Goals attempted like 82games.com does, then it was 12 out of 78 fga's in 10seconds or less, which would be 15.4%. The problem with the Minny game was that Minny had an eFG% of .792 !!!!! in shots done 10 seconds or less. So defensively, that was super duper key to that game. The semi-good 82games.com stats already show that the Rockets are great at eFG% overall, and really great at eFG% past 11 seconds, but not so good at eFG% defense in 10 seconds or less...but the Minny game was extremely poor 10 seconds or less defense. So Minny wasn't getting that many shots off in quick times, but the few times they did, they made the most of them, which was the defensive difference maker in this game. Mike James made 2 big baskets late in the half court, but otherwise was crap in the half court. Mike James made his mark getting down court and taking advantage of the fact that Rafer or Luther couldn't find him or guard him without Yao, Chuck, Bat, and TMac backing them up.
I don't know how much time teams spend practicing transition defense. I'm guessing very little. That leads me to believe transition scoring efficiency is much more dependent on the ability of the team on offense. Conclusion: The focus should be getting even better at reducing transition opportunities. Over the course of 82 games, our transition defense efficiency should balance out with the other good teams.
Defenders can stay in the paint for more than 3 seconds as long as they are within arms length of an offensive player. So if the Rockets are on offensive and Yao is next to the lane, the defender can park his ass in the lane for as long as he wants as long as he's within reach of Yao. The reason Yao and Hayes get a lot of defensive violations is because opposing teams like to get Yao away from the bucket by having jumpshooting big men. Other times, they opposing offenses just totally clear the lane area. So since Yao isn't guarding anyone near the lane, he only has 3 seconds to be allowed to stay there before he has to jump out.
So is the same true on a set zone defense, or is that the exception to the rule? On a man to man help defense (what tubby smith calls his "ball line defense") what the interior defenders would do is play the zone between his man and the ball in order to help if an opposing player drives past the defender. So you do have a "zone" you are responsible for, its just a mobile zone that changes with what opposing player has possession of the ball. Obviously with this 3 sec rule you can't take that same sort of strategy into the NBA and its a disadvantage to interior players who help their teammate by coming off their man. But do the set zone defenses get to park without the arms length rule? What about a 1-3-1 match up zone, would that count as zone or man to man? Sorry for all the questions, just trying to get a handle on what is feasible defensively in the NBA without breaking this weird rule (okay maybe just weird to me)
I haven't done or am planning to do the type of statistical analysis that's been done here but IMO fatigue did appear to be a factor. Yao in the second half didn't appear able to take it as strongly to the hoop as we've seen at other times this season and T-Mac appeared very tentative on drives and ended up taking long shots that fell short. OTOH the T-Wolves appeared very agressive in the second half and played with more energy than the Rockets. At the same time the T-Wolves passing game appeared much sharper than the Rockets. While the T-Wolves tried to push the ball up during a lot of the game in the 4th quarter they also slowed it down in their half courts sets too.
Have you considered the effect of clear path/2 on 1 break/3 on 1 break on FG%? Generally, when you have one of the above situations, you expect a higher FG%. So you can't compare to the half court D and say we are good at half court because we allow fewer points compared to our transition D. I'd venture to guess that it's true for all teams. You have to compare how our half court D compares to other teams' half court D.
I mostly agree with this. normally, short jumpers are consistent with "tired legs" syndrome. When push came to shove, the Rockets were out-hustled and out-executed by a fresher, hungrier team. I can understand that...but what worries me is the seeming sense of panic when the shots stopped falling late and the defense slipped on a couple of crucial late possessions. They are going to have to develop the ability to trust in their game plan and stay in control in tough lategame situations. HayesFan...You will notice one way of stayin g near the paint while avoiding the 3 sec rule is how they will step out momentarily adn then back over. The problems occur when the player takes too long to step out...those would count as mental lapses that will hopefully be corrected upon game tape review. great thread Kim.
