The NBA Playoffs are under way, and I thought it would be fun to use some analytics to talk about some things I am noticing. I'll continue to update this notebook throughout the playoffs with different projects looking at some NBA playoff stuff:
All the code for the projects below can be found here:
Boban and Zone
Posted: 6/07
The most drastic and interesting adjustment that any coach has made this postseason was Rick Carlisle putting Boban Marjanovic into his starting lineup and running a 2-3 zone. This move certainly had some critics on both sides, but I wanted to look at the impact of the move from a defensive standpoint:
Remember, lower defensive rating is good. When Carlisle made the adjustment in game 4, the game was already basically over. Whether or not we give any credence to that sample is up for interpretation. Game 5 is the first game Boban started, and the zone was not quite as effective as people think it was. To me, the Mavs were able to drive and kick and get good looks, although it did disrupt the rim attacks that worked well in games 3 and 4. The defense was slightly above their average in game 6, which is interesting. In game 7, the Clippers offense was going at all cylinders. By that point the Clippers had a few go to sets and reads to make out of the zone offense, and they were just constantly getting good looks. Some of it was also because a zone defense is not something NBA players are used to playing, and a ton of rotational mistakes were constantly being made.
My own personal conclusion is that the zone felt like a last resort to stopping a much more talented team. Yes, it may have thrown the Clippers off, but not enough to generate a series win. At no point did I think the zone was effective enough to stick with it for as long as Carlisle did, but at the same time it might've been the best option he had.
Kawhi Leonard's Two-Way Impact
Posted: 6/07
I've already written about how good Kawhi Leonard's offense is, and I feel like that's also the primary focus of what everyone is talking about. I want to talk about how insane his defense was. As the series went on, Kawhi was the main guy on Luka Doncic. They tried different combinations though the first few games that clearly weren't working, but eventually Kawhi took the matchup head on:
What I originally wanted to look at was how much more time Kawhi spent on Luka as the series went on, but unfortunately I could not find game by game data on that. So, we're just going to have to trust the eye test here. The fact that Kawhi had such great offensive games in games 6 and 7 while guarding Luka seemingly the majority of the time is mind-boggling. Yes, Luka was still able to score against him, but take a look at the efficiency. Under 50% True Shooting is not good, and that is what Kawhi held Luka to. Leonard may not still be at the same level of defender as he was six years ago, but he is still an elite level on ball disruptor.
Luka's Creation Load
Posted: 6/07
A lot of people are talking about the load that Luka Doncic has to carry, and how that affects his play in the second half. I wanted to touch on that a little, so here is a graph showing Luka Doncic's shots created by quarter:
Looking at efficiency by quarter can be tricky because you're effectively taking an already small sample and splitting it into fourths. I wanted to look at distance traveled data, but that was not available on a quarter by quarter basis. I picked shots created because that feels like a decent enough measure of Luka's ability. It is tough to create a shot, and if you are fatigued you might delegate some creation to others. The tricky thing here is that the defense generally ramped up to an extent as the games went on. The Clippers would figure out coverages and individual matchups as the game went along, so some of the dips could be attributed to that. The fourth quarter rise could also be partially attributed to Luka trying to force shots with his team losing. Based on this data alone, I would say it's inconclusive how much fatigue affected Luka. The only way to really know would be by asking him or by using more sophisticated tracking data. The best answer for us is to trust our own eye test, but I would warn against any conformation biases that may happen in this situation.
Is the Mid-Range More Important in the Playoffs?
Posted: 6/05
NBA Defense is all about prevention and risk assessment. You're going to give up something because of how far along offense is, so trying to control what you give up is the name of the game. There have been a lot of studies done that suggest common ideas such as 3-point percentage defense isn't as much as a thing as we think. We also know that the rim is the place where points are scored most efficiently, so that leaves most defensive schemes daring teams to take those midrange jumpers. In the playoffs, these calculations get even more advanced and players can execute them to a greater extent. I wanted to explore what playoff defenses give up compared to regular season defense:
The most notable dip here is in attempts at the rim. This makes sense, as playoff level defenses will do more in terms of preventing those shots. There is also less transition and a higher median of point of attack defense. General mid-range shots also see a slight increase, but not a massive one. To me, the mid-range does become more important in the playoffs, but only for some. For example, you're not going to ask Jae Crowder or Mikal Bridges to even create a mid-range shot. On the other hand, on-ball creators such as Devin Booker and Chris Paul are going to be forced into taking more self created mid-range shots to score with the defense more tilted to stopping them. For the on-ball creators you rely on to create points, the mid-range game is important to have as a counter to what the opposing team is taking away from you. I was also curious in how the efficiency is affected by zone in the playoffs:
There's not really any major differences, which goes back to the beginning point of defense being about prevention and risk assessment. Playoff defenses are more slanted towards taking away the rim and closing out on shooters. Your best players need a counter to that.
Mavericks Size Adjustments
Posted: 6/05
A very obvious adjustment made by Tyronn Lue about halfway through game 3 was to fully embrace small ball. This was something I had been calling for and expecting all season long, and it took him long enough to make the change. The Clippers had a ton of success with the small lineups, as it opened up multiple coverages on Luka, quicker rotations, and more offensive scoring versatility. My expectation for Dallas was that they were going to do the same. Bench Porzingis, embrace Kleber at the 5, and go from there. Rick Carlisle saw it completely differently than I did. Instead of matching the small ball, he decided to go with massive ball. He was going to play a zone with Boban Marjanovic in the middle, Porzingis on a wing, and put three wing sized players next to them. That move messed up the Clippers rhythm for about a game and a half until they figured out how to play the massive ball lineup on both ends. Here is a visualization of the Mavericks and Clippers adjustment:
With game 7 on the horizon, to me it feels like the massive ball is worn out. I think it was smart to throw some junk coverages and lineups against a far more talented team, but where does Carlisle go now? Does he go small ball? Does he go back to the rotations from the first few games? Maybe he stays the course and hopes for the best? Whatever he does, game 7 of this series is loading up to be one of the more interesting intellectual games of the season.
Kawhi Leonard Game 6
Posted: 6/05
Last night's game featured another ridiculous display of scoring, this time from Kawhi Leonard. He hit seemingly every shot down the stretch of the game to make sure the Clippers would see a game 7. The Mavs were throwing different coverages and bodies at Kawhi, but it didn't really matter. Here's a table on how Kawhi Leonard scored his points last night:
The biggest factor here to me is how little he did in the paint. He shot far fewer of his shots within five feet than usual, and far more beyond the three point line. Towards the end of the game, Dallas was really loading up to take away any sort of diving lanes for Kawhi. His response was to go the a variety of step backs from deep. I'm surprised Dallas didn't send two bodies at him more often, but at the end of the day if Kawhi is making 27 foot step backs then you have to throw your hands up. I was curious to see how this shot making game stacks up to the rest of the great performances we've seen so far:
Of course, that insane game from Lillard is at the top in terms of shot making. I'm a little surprised that Booker's game is higher than Kawhi's in terms of shot making, but only by a marginal difference. Either way, shot making of over 30% is out of this world, even for a very small sample. It's exciting to watch a playoffs with such insane offensive creators on seemingly every team left.
Devin Booker Shot Making
Posted: 6/04
Last night, Devin Booker put on a ridiculous display of shot creation and shot making. Since a lot of what I talk about here is those two elements, of course I had to highlight how insane Devin Booker was. Booker scored 47 points on 89.02% True Shooting. He was also doing it against a defense that was slanted towards stopping him. Frank Vogel threw a lot of different coverages at Booker, including drops, hard shows, blitzes, top locks, denials, 1-2-2 zones, etc., but Booker expertly read them all. To show how good he has been offensively for most of the series, I wanted to look at his individual shot making by game:
As you can see, Devin Booker put on a scoring display in the last two close out games of the series, He did not shoot as well in games 2-4, some of which can be attributed to the Lakers adjustments. They threw more doubles his way and made an effort to not let him go right. Booker eventually adjusted to that by working more off the ball and getting to pull ups off his left hand. He also did damage in early offense, getting to his spots before the defense was set to stop him. Booker is a great player, and if the Nuggets don't get Barton and Dozier back then I don't see how they contain Devin Booker.
Portland Trail Blazers Defense
Posted: 6/04
The Portland Trail Blazers just lost in 6 games to the Denver Nuggets, a team without their 2nd best player and two key rotation guys. The Nuggets started Facundo Campazzo and Austin Rivers as their backcourt to match up with Damian Lillard and C.J. McCollum. Obviously, something went wrong. The Blazers defense was awful, posting a 125 defensive rating per CleaningTheGlass, which is only better than the Celtics (who were injured and played arguably the best offense ever). I wanted to explore why the Blazers defense was so bad, and thought I could find the answer through shot location data:
It's not like anything is glaring here. The Blazers gave the Nuggets a little more shots at the rim and from 3 than average, but not by a shocking amount. To me, the biggest reason their defense was so poor was because of lack of discipline and game plan understanding. The looks Denver was getting were just too easy. The Blazers should not have gotten burned on as many cross screens and simple hand offs as they did. They should not have had as many breakdowns defending pick and rolls. They also shouldn't have fouled nearly as much or overhelped as much as they did. There are issues here that go beyond personnel, even though their defensive personnel is relatively weak. The Blazers have a lot of self evaluation to do this offseason, both about their coach and their roster. It'll be interesting to see how the roster looks on the first day of next season.
Luka Doncic Creation
Posted: 6/03
Yes, I picked the Clippers in 5. Yes, I overrated Tyronn Lue. No, I do not think the theory of my pick was bad. I think something that I didn't see coming when I made my pick was how much Luka Doncic would demolish this Clippers defense. In theory, the Clippers have four big wings who should be able to hold their own against Luka. However, in the series that hasn't particularly been the case. Luka might have more counters than anyone in the league when going against variety of coverages. His ability to create shots in this series for himself and his teammates feels absurd, so I wanted to take a look at it:
Above is a scatter plot of self creation, all stats being per 36 minutes. As you can see, Luka rises far above anyone else in terms of shot creation for himself. We know he is an elite passer, but being this elite of a scorer against a very good defense has been unbelievable. If the Clippers are switching, he's going to pick his matchup and destroy that player. If they trap, he'll make the cross court kick out. If they drop, he'll keep the defender in jail and make the right read. I honestly think that number of shots created for others could be a lot higher if he wasn't having so much success as a scorer (there's only so many shots to create). A limitation of this is it doesn't tell us about efficiency or total creation, so I wanted to look at that here:
No surprise, but Luka is at the top. We also have to give a quick shoutout to Trae Young, who was exceptional against the Knicks. The efficiency of Luka is insane for how many self created shots he's taking. Some of that is obviously due to variance, but some is simply because of how many ways he can get to his shots against any matchup or coverage. If you really thin about it, Luka is creating over 38 shots per 36 minutes. That's insane. I think Luka is going to be widely regarded as the best player on earth sooner than some may realize...
Clint Capela's Rim Protection
Posted: 6/03
The Hawks just defeated the Knicks in 5 games, and Trae Young is taking the bulk of the credit for that. While he deserves a ton of credit for the creation load he carried throughout the series, the Hawks won with defense. I wrote before about how they were forcing everything with Julius Randle towards Clint Capela, but I have not written about Capela's side. To me, Capela is one of the best defensive centers in the league. I wanted to look at how much of an impact at the rim Capela had in the playoffs so far:
Before we start talking about this chart, I have to say that we are dealing with a small sample. However, Capela's numbers match the eye test, and he also had great regular season numbers. The Knicks could not generate any good looks at the rim when he was in the game. Some of that has to be attributed to the lack of spacing from New York, as they had very little shooting throughout their roster. They also played with a traditional center at all times, allowing Capela to roam around the paint. He did a fantastic job all series, and was the biggest reason on Atlanta's side as to how they had such a good defense.
Dame Time
Posted: 6/02
Damian Lillard put on an insane performance last night. He is a fantastic player, and it always feels like he makes every shot down the stretch. NBA.com has publicly available "clutch stats" that are defined as a game within 5 points during the last five minutes of the game. Using these stats, I wanted to see how much Dame Time really exists:
Above are clutch stats from the regular season, sorted by points per 36 minutes. Near the top of the list is none other than Damian Lillard. He is scoring a ridiculously high number of points in the clutch while shooting an extraordinary True Shooting percentage. However, I am very cautious when using these numbers, and here's why. First of all, the sample size is tiny. 125 minutes is about how much Dame would play in four games, and we would never base any statistical argument on four games of play. Second, the definition of clutch is a little bit questionable. If the Blazers are winning a game down the stretch, the other team has to foul. Dame is going to get to the free throw line way more often than usual because of this, artificially raising his True Shooting percentage. When looking in the prior years of the clutch data, Dame's efficiency frankly hasn't been very good in the "clutch." Getting overexcited based on data from a major sampling error may be a little silly, but at the end of the day sports are built for fun and entertainment. Fans love Damian Lillard tapping his wrist after he hits big shots, and I love it too. However, does Dame Time actually exist? To me, the answer is maybe, but certainly not definite either way.
Passing
Posted: 6/01
People like to talk a lot about passing and how that can help generate offense. But what does that mean? Passing feels like a term that is glorified amongst the general population, but there is a lot more that goes into offense than just passing. To look at this, I decided to make a scatter plot of total passes made vs. offensive rating:
As you can see, there is a slight negative linear correlation between passing and offensive rating. On first thought, in theory more passing should lead to more offense, right? However, more passing generally means you aren't actually creating advantages, and it probably also means you have longer possessions. Making the right pass at the right time is much more important than just the act of meaningless passing along the perimeter, which is something I feel can be forgotten at times. To think about how the "right pass" can impact playoff offense, I looked at potential assists vs. offensive rating:
Strange, right? There is a strong negative correlation between potential assists and offensive rating. This gets back to the idea that every team is a snowflake offensively. They all succeed or fail in different ways. The Nets have the best playoff offense in the league with three unbelievable shot creators who don't need assists to generate efficient offense. The Jazz have an awesome offense, and they do it with multiple pick and roll ball handlers alongside one good individual shot creator. While passing is a good thing, you don't need it to have a successful offense. Plenty of teams can really bend opposing defenses without the perimeter passing we generally associate with "good offense," and that's okay. Next time you're watching a game and hear a commentator talk about ball movement/passing, focus less on the passing itself and more on advantages the offense can create.
Will Barton and PJ Dozier's Offensive Impact
Posted: 6/01
Since I picked the Nuggets to win the series before I knew Will Barton and PJ Dozier would miss most if not all of it, I feel like I need to talk about their impact. First of all, I think the side of the ball where those two would be the biggest upgrade is on defense. Both are above average for their position, and they both have more size and length to defend Dame/McCollum/Powell. Since defense is so tough to quantify, I wanted to look at their impact in terms of scoring:
Both Barton and Dozier have been playing for the Nuggets for multiple years. Campazzo just came over this off season and Rivers was signed about a month ago. It's just impossible for either of those two to have the knowledge of the system or the understanding of how to play with Jokic as players who have been their for multiple years. The most obvious place where this is seen is with the scoring off cuts. Barton and Dozier are both very good cutters, which fits perfect for the Nuggets flow motion offense. The bigger missing piece here is Barton, so I'll touch more on him. Barton has been a sixth man for this team in the past because of his ability to create shots off the dribble. With Murray out, the Nuggets just don't have anything in terms of secondary shot creation. I've written before about how the Blazers are playing Jokic, and it's just hard to generate better offense than the Blazers playing the way they are. For the Nuggets to pull out this series, they are going to need a herculean offensive effort from Jokic.
Ja Morant in the Paint
Posted: 6/01
During the Memphis Grizzlies playoff games so far, it feels like all the commentators talk about is Ja Morant's ability to get into the paint. Despite this, I feel like they never talk about how he gets there or why he's so effective when he does. There are a few skills Ja Morant possesses that allows him to get to the paint so often. The most important one to me is the handle. Yes, Ja is a special athlete, but that would be meaningless if he didn't have such a good handle. He has a herky-jerky ball handling style that is really tough to stay in front of, and then the short distance explosion just amplifies his effectiveness. Morant is also incredible at using screens. This is an important skill that doesn't get talked about very much, but Morant really uses the screen well when his defender is chasing over the top.
In terms of effectiveness, Morant is really good because he is elite at three things: finishing, floaters, and passing. The finishing is the most obvious, as Ja has other-worldly body control and explosion. He can finish a million different ways, and is completely unafraid of the defender. He is an elite passer, as his knowledge of angles and where help is coming from is really high. He also is really good with his delivery, doing it in a variety of different ways with either hand. The skill that sets him apart to me is the floater. The floater is a great counter to the big dropping back really far, and the threat of it opens up a lot for rim finishing and playmaking. During the first 3 games, Ja was excellent with the floated, but the Jazz made an adjustment in game 4:
First of all, there had to be some regression. No one can sustain 65% shooting on floaters. However, the Jazz didn't just sit back and wait for the regression. They played a little more aggressive with their big, not dropping him as far as they were earlier in the series. If a floater was going to come, it was going to be more contested. While this takes away the floater and some of the rim finishing to an extent, this does open up passing lanes for easier drop downs or corner skips. I'll be watching to see how Ja personally adjusts to this coverage in the win or go home game 5.
Shooting Fouls
Posted 5/31
In the Playoffs, there is a notion that fouls get called far less often than in the regular season. To see just how true this is, I decided to make a chart showing regular season and playoff foul rates for both 2 and 3 pointers:
As you can see from the graph above, there is a slight decrease for two-pointers and a raise in 3-pointers. Based on the data as a whole, can we really see a major difference in how the games are called in the playoffs? It doesn't appear so, and the foul rate in the playoffs so far has actually been higher than the foul rate in the regular season. There are a couple reasons I think there may be a perception of less fouls in the playoffs. First, maybe the playoffs are just more physical. There are less open shots at the rim, rotations are cleaner, and maybe that is going to lead to fouls. The player pool is also better in general, and better players draw more fouls. Second, maybe we just remember the non-calls more. That's always the case, but maybe even more so in the playoffs when we are likely to attribute non-calls to "playoff basketball." I have also heard people talk about how players that draw more manipulative fouls in the regular season may not be able to sustain that in the post season. Of course, I wanted to look at this:
The table above shows the foul rate on 2's and 3' for the 15 players with the most playoff FTA. When I think of "B.S." foul drawers, the first names that come to mind are Trae Young, Jimmy Butler, James Harden, and Joel Embiid (I didn't include Chris Paul or LeBron James because of injuries). Young, Butler, and Embiid all have lower foul rates, but I wonder how much of that is due to context. I've written a lot about how Butler has been guarded, Embiid is just bullying the entire Wizards team, and no one on the Knicks can guard Trae Young. They are all different circumstances that may or may not lead to lower foul rates. James Harden's foul rates are way up, and I think that's because of how weak Boston's point of attack defense is outside of Marcus Smart. Another obvious factor here is sample size. 3-4 games of action is just not enough to make sweeping conclusions about what is going on. I certainly think how fouls are called in the playoffs is an interesting discussion, and not one that is as set in stone as people think it is.
Suns Game 4 Offensive Adjustment
Posted 5/31
I wrote after game 3 about the Suns offensive struggles. They then came out in game 4 with an offense that was much more effective. A big reason as to how they were able to do this was through their work in transition and early offense. Playing against a defense that isn't set is always much more efficient, and the Suns were really able to attack this. The Lakers play big and attack the offensive glass, which leaves them susceptible to weak transition defense. Monty Williams and his staff were smart enough to exploit this in game 4:
As you can see, the Suns did much more of their work in game 4 during the earlier stages of the shot clock. The Suns shot the ball in transition on 25% of their possessions, which is in the 100th percentile per CleaningTheGlass. League average eFG% was 53.8% in the regular season, so anything above that is a value add to your offense. When pushing the ball in transition or early offense, the Suns had a much better than average eFG%. As I wrote about earlier, the pace normally slows down in the playoffs. Frank Vogel is an excellent coach, and he is going to adjust for the ever important game 5.
Jimmy and Bam
Posted 5/31
Jimmy Butler and Bam Adebayo both really struggled offensively during the Bucks sweep against the Heat. As a team, the Heat posted a really bad offensive rating of 98.9 in the series, per CleaningTheGlass. The would've been the worst mark in the regular season by about 6 points. A lot of the reason for why the Heat struggled was because they didn't have an answer for the Bucks packing the paint. Giannis defended Butler and went under every screen, daring Jimmy to shoot a pull up from anywhere. Lopez defended Bam, and just hung back in the paint trying to bait Bam into shooting anything outside of five feet. With both players not being very good scorer outside of around 5 feet, the strategy paid off. Below are tables of both players shot attempts by distance and their effectiveness:
What stands out about the Jimmy table is that he couldn't get to the rim nearly as much. Even when he got their, he just wasn't a very effective finisher. A bulk of the credit should go to Giannis, but the backside help from the Bucks was also excellent. Also, check out how poor the efficiency was from 5-19 feet. Those are the shots the Bucks wanted.
Looking at the Bam table, again the lack of shots at the rim stands out. What also stands out is how much he was baited into shooting 10-14 foot jumpers. Like I said earlier, anything outside of 5 feet the Bucks would live with, and neither Bam nor Jimmy had much of a counter.
Erik Spoelstra to me is the best coach in the league at manipulating half court space through actions and player placement, and the fact that this Bucks defense did what it did to the Heat definitely raises some questions. Here's some food for thought: Can Jimmy Butler and Bam Adebayo not coexist offensively? Was the supporting cast just too weak? Is this Bucks defense just that good? To me, the answer is a little bit of everything...
Spacing
Posted 5/30
Lineup construction is really put under the microscope during the playoffs. Every lineup decision a coach does or doesn't make is highly scrutinized, and for good reason. Lineup decisions is the easiest place to judge a coach, and they also have a massive impact in the box score. A lot of people talk about lineups needing to have spacing, and in theory the easiest way to do that is to have players that "can shoot." The explore this, I used violin charts to look at offensive success of lineups with differing numbers of players that "can shoot":
Shot quality is probably the better metric to use, as the offensive ratings in small samples are ridiculous. Overall, since this is using playoff data, all the samples are so small it's tough to make meaningful conclusions based off of them. However, if we want to look at the trends, both offensive rating and shot quality seems to increase with more "capable shooters" on the floor.
You've probably noticed at this point that I keep putting shooters in quotes, and that's kind of the point of this one. There's no line between a someone that can or cannot shoot the ball. If you really wanted to look at it, there would be a bell curve of shooting, but even that would be flawed. Some players get their shot off quicker than others. Some utilize a one-two step while others use a hop. Some can shoot off of different types of screens and movement patterns while others can't. Different players have different release timings and release points. There is a ton that goes into how the defense treats the gravity of an opposing player that goes far beyond 3-point percentage and attempts. Yes, every team needs to find ways to maximize space. But how each team does that is unique. Look at how the Heat did it last year compared to how the Mavericks do it this year and it's two completely different yet effective ways of space creation. Lineup spacing is a very complicated concept that cannot be boiled down to shooting percentages. Pay attention to how the defense is playing from 1-2 passes away or from the weakside corner. Pay attention to the closeouts. Also, at the end of the day you need a player who can really punish the spacing you create. I could keep going, but just remember there are a ton of factors that go into what we perceive as spacing.
Nikola Jokic Creation
Posted 5/30
The Portland Trail Blazers strategy for defending Jokic has been interesting. They are bringing a lot less help than some other teams would, and are trusting their players to do a good enough job at defending Jokic straight up. It's a strategy that I think is paying off. The chart below shows shots created by Jokic on passes and his own shots, in both the regular and post season:
Jokic is creating a lot more shots for himself, but that's what the Blazers want. The way Jokic really bends opposing defenses is with his passing versatility. If you stay home and prevent him from doing that, all of a sudden it's a lot easier to defend him. Jokic is creating about six less shots per game through passing in the post season. Some of that is because of the context of who's on the floor. Chicago and Zipper HO actions with Austin Rivers is completely different than those same actions with Jamal Murray. Same can be said for all of their pop and inverted screening pet plays, or the simple slips and wide pins. Even not having WIll Barton and PJ Dozier are big blows for who Jokic is passing to.
I do think the strategy is paying off. Great scorers can elevate their own efficiency, but great passers can elevate everyone else's. If you can limit Jokic to just being a great scorer, than you are significantly lowering the ceiling of this Nuggets offense. For the Nuggets, I would continue to try and expand on the actions I use Jokic in. Use him in sets where he's moving around the floor and off ball cutters and screeners are forcing more difficult communications. It'll be interesting to see how Michael Malone adjusts going forward, especially if Barton and Dozier don't come back.
Donte DiVincenzo Rebounding
Posted: 5/29
The loss of Donte DiVincenzo for the Bucks is a very significant one. DiVincenzo is part of the Bucks closing 5, providing them with perimeter defense, shooting, and tertiary creation ability. DiVincenzo is a big part of the Bucks lineup construction, as his ability to space the floor and handle a variety of defensive assignments allows him to slide in with any unit. His loss means that players like Pat Connaughton, Bryn Forbes, and PJ Tucker are going to get significantly more minutes. One of my favorite parts of DiVincenzo's game doesn't get talked about a lot, so I wanted to point it out here:
Above is a table showing the guards that grab the most contested rebounds per 36 minutes. Donte DiVincenzo is a really good rebounder, and I think that matters for the group that I thought was the Bucks strongest (especially for the impending series against the Nets). I think the Bucks best closing group features Holiday, DiVincenzo, Middleton, Tucker, and Giannis. That group completely opens up the floor for Giannis, has a lot of switchability and versatility on defense, and is a group I think the Nets could struggle with. Something that's important to note with these smaller groups is that defensive rebounding is usually an issue, especially against a team like the Nets who crashes the offensive glass much more often than most other teams. I think DiVincenzo's loss is going to be felt most with the lineups the Bucks can throw out there in the next series, and it will be interesting who takes on his important role.
Julius Randle Scheme
Posted: 5/29
Julius Randle has struggled offensively so far this series, and a lot of that has been because of the scheme Nate McMillan has used against him. He's doing everything to prevent Randle from going middle, pulling early help from the weakside to help on baseline drives, and not giving him anything easy at the expense of some open looks. The strategy has worked really well, but I wanted to see how much it affected the quantity of Randle's shots in the middle of the floor:
As you can see from the table, Randle is still getting a fairly similar amount of his shots in the middle of the floor. The difference is the quality of the looks. The difference has been how effective he is in the middle of the floor because of players like Capela doing a great job of protecting the rim. Randle is also just not hitting a lot of the same shots he had success with during the regular season. Per PBPStats, Randle's shot quality has actually been better in the playoffs than the regular season. His diet had consisted of a lot of really tough looks that just aren't falling right now. Below is Randle's shot chart, which does a good job of showing how the rim presence is affecting his effectiveness:
However you look at it, Julius Randle needs to find ways to get middle more and get more easy looks. The Knicks need sets to get Capela moving and away from the rim. It's tough with the specific personnel the Knicks have, but Thibodeau needs to get creative here.
Derrick Rose Defense
Posted: 5/29
Derrick Rose is a very important piece to the New York Knicks offense. He is a crucial secondary ball handler and is their biggest pick and roll threat. Rose is also the Knicks best playmaker. Too much of the Offensive scoring burden is already on Julius Randle, so Rose needs to be on the floor 35-40 minutes a night to give the Knicks a fighting chance. The problem with this is his defense. Rose is a really poor point of attack defender at this point, which is a big problem against this dynamic Hawks offense:
When you have a poor defender, you want to hide him away on a non-threat. The problem with that is the Hawks don't have a non-threat in their closing 5 (Young, Bogdanovic, Hunter, Collins, Capela). He's spent a lot of time on Trae Young, but the results have been really bad. He can't stay in front on an island, and any simple drag or double drag immediately creates an advantage for the Hawks. Put Rose on Bogdanovic and it's the same story. Put Rose on any of the other three and he's way too small. The way the Knicks have been able to counter this in previous games is by playing more aggressive coverages with Nerlens Noel, but with him being hurt Taj Gibson can't execute on the same level. Tom Thibodeau is in a really tough spot with how he handles Rose defensively, and I'll be interested to see what he does in the crucial game 4.
Suns Offensive Struggles
Posted: 5/28
The Suns offense has struggled so far this post season during the first three games against the Lakers. During the regular season, the Suns posted an offensive rating of 117.7, good for 6th in the league. So far in the playoffs, their offensive rating has fallen to 105.4, which ranks 14th amongst playoff teams. When watching the broadcasts, all the announcers explanation for this is the Lakers "imposing physical stature." While I do think that is some of the reason for their success on both ends (more so on offense), I really don't think that tells us why the Suns are struggling offensively. To look at this, I decided to look at shot locations for the Suns:
Not to pat myself on the back or anything, but it looks like my intuition was right. Not only are the Suns getting to the rim more often, but they are scoring there at a higher efficiency. Where the Suns have struggled so far is everywhere else (besides 20-24 feet). A big reason for this can be attributed to the lack of scoring ability from Chris Paul. He's the one taking the bulk of their mid range jumpers, and he just hasn't had it in the post season because of that shoulder injury. Another factor is Devin Booker. Since game 1, the Lakers have done a much better job making his life more difficult. Vogel has thrown a multitude of different coverages his way, not giving him any automatic reads. The Lakers are also excellent in rotation, which can explain a little bit of the shooting struggles, but some of that also just comes down to variance. Next time you watch the Suns, try to think more critically than "the Lakers have physically imposing players."
Individual Shot Making
Posted: 5/28
Yes, I am very interested in how shot making can impact an individual series or even the entire playoffs. I have written about it in a couple of articles, but here I wanted to quantify it. To do this, I used the shot quality metric available at PBPStats. As mentioned in an earlier post, shot making is simply eFG% - shot quality. To get a baseline of expected shot quality, I used the players shot making over the regular season because of the larger sample. This is the shot making of the top 15 players in field goal attempts per game:
What I'm interested here is in the +/- expected column. I think some of that can be attributed to skill, but some of that is also luck. Since we are comparing the players shot making to themselves, it automatically adjusts for sampling issues with league averages. What I will say is that playoff basketball is different, as players on both sides are more locked into their own game plan by a significant amount. Having better shot making in the playoffs could be partially attributed to the malleability of certain players skillsets. However, I do think a lot of it can be attributed to luck. I'd be very careful before using efficiency numbers in the playoffs as a point of discussion without using context.
Just for fun, here's the players who are 1-50 in field goal attempts per game:
LeBron James Ankle/Age
Posted: 5/28
Is LeBron James literally slowing down? That's a question I wanted to look at, and luckily NBA.com has awesome tracking data on speed and distance. With LeBron's ankle injury coming into the playoffs, he clearly hasn't been himself so far. I wanted to se how that impacted his activity from a play to play basis, and to do this I decided to look at his distance per game and average speed while on the court. I looked at data sice he was 30 years old to try to be more fair to him:
Is LeBron James running less since he was in Cleveland? Obviously. Not only has LeBron aged, but he is playing a different role with the Lakers. They are not quite as heliocentric as the Cavs were, and that allows LeBron to take on less of the load on both ends. He is used more as a stationary playmaker offensively with higher IQ cutters and better shooters, and defensively he is used more in the weakside corner free safety role (which he's incredible at). The issue with distance data is that some of it can be fairly reliant on how many minutes a player plays, so I wanted to look at some data that didn't rely on minutes played:
It's weird that the average LeBron speed in the playoffs is slower, right? When thinking of why this may be, the only reasons I could think of is because the pace of the game slowing down or his minutes load going up (so he couldn't move as fast the whole time). While LeBron's speed is lower in this year's playoffs than it has been at any other point, it really isn't out of line with the general trend we have seen. The eye test along with some other numbers does tell us that it is still bothering him, but based on the speed data there doesn't seem to be a huge impact.
Team Shot Making
Posted: 5/27
It feels like shot making in this years playoffs has been very evident when you look at the results of the games. There have been a few games so far that feel like they were completely swung by the element of shooting variance. To explore this a little bit, I wanted to look at shot making from each team. Shot making is determined by eFG% - shot quality, and luckily PBPStats has a pretty solid shot quality measure.
Obviously, the teams that are hitting shots are having a lot of success. Something that is eye opening to me is just how well some of these teams are shooting it. The Mavericks shot making so far is above 16%, which is a ridiculous number. That's adding about 30 points per game of value they wouldn't be getting with normal shot making.
There are some drawbacks with the limitations of how shot making can be measured, but I do think this is interesting to keep in mind when watching games going forward. It is important to ask yourself if teams are playing really well or if the shots are just going in, and vice versa.
Playoff Pace
Posted: 5/27
Something I hear (and say) all the time is that pace of play slows down in the playoffs, and teams that rely on transition or early offense tend to struggle in the playoffs. Now that every team has played two games, I wanted to take a look at this. I used PBPStats measures for average possession time over the past 20 seasons. Higher average possession time obviously means the pace is slower.
The first graph above shows the average seconds per possession during all possession types. As you can see, in almost every season the playoffs are significantly slower than the regular season. This makes sense, as when teams are more locked in they aren't going to give up looks in transition or early offense as easily. However, some of the slow down could be explained by better defensive preparation or better shot selection. Come playoff time, teams know each others playbook and tendencies, therefore possessions may take longer to generate an advantage. To try and take this away, I wanted to look at pace after a missed shot or after a steal.
For the most part, we associate transition opportunities with chances off of live rebounds. The past couple of years, pace after rebounds has been much slower in the playoffs. A lot of this is because teams offensive rebounding schemes. Teams nowadays understand the vale of transition offense, so they are willing to take that away by not even going after offensive rebounds. The affect of this is less transition scoring. Steals are a different story, and a lot of those numbers can rely on the context of the steal. Generally, however, pace has been slower after a steal.
Ben Simmons Advantage
Posted: 5/27
Ben Simmons has been awesome in the Wizards/Sixers series. He has bee a huge matchup issue offensively, and I wanted to dive a little deeper into how he's been so effective.
As you can see from the chart above, a lot of Simmons offense has been coming from rebounds and post ups. The reason he is so effective here is because of the Wizards personnel. They are starting 3 guard sized players while the Sixers only start 1. That means one of the guards have the matchup with Simmons, which is why Philly has used him so much in the post. They are also using Simmons a lot in the dunkers spot. Putting him there makes it so that when the baseline help comes for Embiid, there is a sink from either Neto or Beal on Simmons. That's a huge issue if a shot is missed. Unless Scott Brooks makes some major lineup changes, I think Simmons will keep having this type of success.
Mavericks Offense
Posted: 5/26
The Mavericks offense has been absolutely ridiculous in the first couple of games. They have the best offensive rating of the playoffs with a ridiculous 129.7, per CleaningTheGlass. For reference, the all time regular season record is 119.4, set this season by the Brooklyn Nets. Using the NBA's publicly available play type data, I wanted to explore just how they've been doing this.
Looking at their play type data, a few categories have increased in usage. The most obvious one is the jump in isolations. Besides isolations nearly doubling and the pick and roll ballhandler shooting more often, there doesn't seem to be any huge differences. Where we can see differences is with the efficiency. Spot ups take up almost a quarter of the Mavericks offense, and they are scoring on those attempts at an unbelievable mark of 1.362 PPP. The ball handler in the pick and roll and the roll man are also seeing upticks in efficiency.
To me, this is just confirming what we are seeing on the film. Luka Doncic is an excellent creator in the pick and roll, and he is absolutely killing the Clippers in those actions. Upticks in efficiency in the categories mentioned all stem from Luka Doncic's brilliance in handling the ball. Tyronn Lue needs to scheme better against this Mavericks offense in terms of slowing down the pick and roll success, otherwise they will not slow down this potent offense.
Chris Paul's Shoulder
Posted: 5/26
Obviously, Chris Paul has been really feeling the affects of that shoulder injury he suffered in game 1. There's a ton of ways to show this, but I decided to do it by looking at his usage.
Looking at his playoff usage in the blue, where he is at right now is an all time low. His usage rate right now for this years playoffs is at around .15, which is about where Kenrich Williams was at this season. Whichever was you want to look at it, the Suns need an effective Chris Paul if they want to beat the Lakers.
Impact of the Stars
Posted: 5/25
It always feels like the best players in the league have a far greater impact on winning come playoff time. I wanted to take a look at how much time they spend on the ball during the regular and post season using the NBA's publicly available tracking data. Below is a graph showing time spent on the ball per minute for the 30 players that have the ball the most:
The white dots on the graph show were the player is at in the playoffs, and the tail of the "comet" shows where the player was at in the regular season. A lot of this is expected, but a few are surprising. Obviously, once it gets to playoff time you want the ball in your best players hands. This effect can be seen with a lot of the players above. However, I am surprised to see such a large increase for players such as Jimmy Butler and James Harden because of their teams particular offensive schemes or situation. I am also surprised to see a couple of players in the negative, such as LeBron James, Joel Embiid, and Julius Randle.
Updated chart (5/27):
What's up with Jimmy?
Posted: 5/25
Anyone who has watched the first couple games of the series between the Bucks and the Heat can see that Jimmy Butler is struggling. He just has not scored the ball efficiently at all, and has clearly not had the same impact he had in the regular season. Below are his shot charts for the regular season and the playoffs thus far:
What first stands out to me is the inefficiency around the rim. It's not like he's not getting there at all, but he's just not able to finish. Some of these are simply because of missed shots, but some are also because of the length of Giannis and Brook Lopez. However, with a star players scoring volume down so much, I wanted to quantify the matchups to show how the Bucks were stopping Butler:
As you can see, a ton of the damage is being done by Giannis. He's guarded Butler almost half of the time, and is holding him to a measly 10.7 points per 100 possessions. For reference, that would've ranked 9th to last amongst qualified players this season. Butler also hasn't faired too well against P.J. Tucker either. It's also important to note that basketball is a team sport, and there's more to it than the individual matchups. The help defense against Butler has been good around the basket and at the nail. However, it's impossible not to see Butler's struggles scoring while watching the first two games.
Wings win
Posted: 5/25
Playoff basketball always feels like a completely different animal when it comes to rotations and who can stay on the floor. To look at this, I made a graph depicting player's height and the percent of the minutes played.
When looking at the graph, we can see that players listed at around 6'6" to 6'9" do seem to get a larger proportion of minutes so far in the playoffs. Some of that may be attributed to how some playoff teams are constructed, but also if better teams happen to have more wing sized players than that is saying something. 7 foot players are 84 inches tall, and there does not seem to be a huge drop off in players of around that size.
Updated chart:
If you're interested in learning how to get started in R or learning more about basketball analytics, I would highly recommend subscribing to the F5 by Owen Phillips. His tutorials allowed me to build a great foundation on how to scrape data and make these good looking charts you see in this notebook. Could not recommend it more!
Comments