Name: NBA Analytics Lookaround: Taking an offense from good to great, LeBron’s free throws, Harden’s usage and more
Brand: theathletic.com
SKU: theathletic-com-1
Availability: OnlineOnly

critic

NBA Analytics Lookaround: Taking an offense from good to great, LeBron’s free throws, Harden’s usage and more

preview

Nov 25, 2019In this week’s installment, I want to quickly touch on a few topics from the last week.First, I want to revisit the concept of marginal contributions on the team level. Using a framework of marginal gains or losses is extremely helpful across the gamut of topics for analysis, as when we are trying to determine the “value” of a given player, team or strategy, it’s always relative. There are 1,230 wins to go around in an NBA season — . I’m perhaps too much of a traditionalist to be supportive of the sort of omnibus approach to schedule changes that has been floated, but discussion of possible impacts is probably best left to a time when we see more concrete proposals.But, given that fixed number of wins, if the absolute level of production rises, for example , this means a win is “more expensive” than it used to be, hence the need for a baseline of that cost to measure against. I’d be remiss to not credit as an inspiration for this line of thinking.In discussing the , I used marginal points added or saved from high-value shot areas to illustrate how rim defense offers greater and more reliable gains than does focusing a defense on the arc. But what about on offense? In one of those weird asymmetries between offense and defense, while I found the spread between the most and least effective teams in terms of allowing fewer points was substantially wider on rim attempts than from beyond the arc, in most years since the full adoption of the 3-pointer as a primary weapon, offenses have shown equal dispersion across the two zones. On the chart below, each dot represents a team-season from the 2013-14 season to present, with performance this season labeled by team:&nbsp;A couple off call-outs here. First is that though I strongly suspect this will regress towards the mean a fair amount, there are five teams (Minnesota, Orlando, Atlanta, Golden State and after Sunday’s 10-44 chuckfest, Houston) that would have cost themselves more points than has any team over a full season in the tracking data era. Given that Orlando and Golden State are 29th and 30th and Minnesota is 24th&nbsp; in what I’m now calling average shooter quality () while Atlanta and Houston (more on the below) feature high volume-high difficulty shooters as lead guards, it’s not terribly surprising for those teams to go through some slumps. However, I think it is unlikely that all or any of these squads will shoot this&nbsp;poorly over a season.On the positive side, the and are putting up near historic seasons in terms of rim attacks. The only team that added more value from the restricted area over this time span was the 2013-14 , which is notable because three of the top five rim-attacking teams of this time period have featured , with the 2017-18 as the third.In terms of interpreting the generally higher positive numbers for rim attacks, I think the way I’d describe it on a general level is to say that a high volume of 3s can take a team from bad to average just by moving inefficient midrange attempts to higher-value areas, while attacking the rim can take an offense from good to great.Speaking of LeBron and attacking the rim, over the weekend The Athletic’s despite the amount of time he spends in the paint. Complaints about officiating are nothing new, but there are some things we can look at to determine the degree to which these complaints are purely sour grapes or if something is going on.In this case, I think there is plenty of evidence that LeBron is not getting as many whistles as he once did. He is getting to the rim at near career-best rates, yet is shooting fewer free throws relative to rim attempts&nbsp;In general, I have found that declining foul draw rates, especially on rim attempts, can be an indicator of athletic decline, but qualitatively that doesn’t seem to be the case this season, as James is by pretty much by every account playing with more pep and vigor than in recent regular seasons, . Also, and that was four seasons, three NBA Finals and one championship ago for James, so I’m not ringing that bell again without definite proof.Some other metrics that have proven helpful in identifying players who might be on the career downslope, either because of injuries or simply getting older, include rim finishing, dunks, rim-attempt rates,&nbsp; and-1 rates and frequency of getting blocked at the rim. In LeBron’s case, this suite of factors doesn’t tell a coherent story.• His finishing at the rim is currently the lowest since his rookie season — though, of course, less kind officiating would show up here, both in terms of attempts while getting hammered are harder, and those rim misses where he does get hit count against his field goal percentage, whereas they are not scored as field goal attempts if shooting fouls are called.• He’s dunking at similar rates to recent careers and getting blocked at normal rates.• On a small sample, he’s actually finishing and-1s at higher than career rates (20.7 percent of his shooting fouls drawn have resulted in an and-1 over his career, while this year that number is above 28 percent). Though it’s not much of an effect on this volume, unusually high rates of completing and-1s can make it appear&nbsp;a player is getting a tougher whistle, because of the second free throw a player doesn’t&nbsp;attempt after scoring through a foul.• As shown above, he is getting to the rim at the second-highest rate of his career on a per possession basis.So what to make of this? I’m not sure. A full examination of LeBron’s rim attacks shot-by-shot could shed some light as to whether he’s attempting shots in similar situations and simply getting fewer calls or if his attempts are on average less conducive to drawing fouls, but that study is a bit beyond the scope of public data. This is still something worth watching, as the issue is unlikely to go away so long as James isn’t getting to the line as much as he’s used to.Over the weekend, made some waves discussing how much he gets double teamed:Rockets’ James Harden on the attention he’s seeing: “The whole season they’re running doubles teams at me. I’ve never seen that in an NBA game where you’ve got really good defenders &amp; someone else running at the top of the key. Y’all let me know the last time you’ve seen that.”— Ben Golliver (@BenGolliver)I’m not going to try to answer whether Harden is right or wrong in saying he gets double teamed more than other players, but I will point out that this defensive attention is either explicitly or implicitly by design. Over the last several years, Harden has always been among the league leaders in time of possession, having the ball an estimated 53.0 percent of the time he was in a game and Houston on offense in 2016-17 (3rd), 49.8 percent in 2017-18 (2nd), 46.7 percent last season (led league) and is currently 3rd at 45.6 percent behind and . (Hey, did you hear they were traded for each other? But I digress.) The question of how will Harden and Westbrook share the ball has been answered. They won’t. After sitting finishing 3rd in possession share at 44.0 percent last year, Westbrook is down to 32.7 percent, in the company of combo and secondary guards like Fred van Vleet, and , and playing him off the ball has been an…imperfect fit given his seemingly impossible 18.2 percent accuracy (6-33) on catch &amp; shoot 3s so far this season.So the defense knows he’s going to have the ball. Perhaps more important in terms of the opposition’s ability to show Harden multiple defenders is that the defense is also going to have a pretty good idea where the other four Rockets are as well.Aggregate level “speed and distance” data available from NBA.com isn’t especially useful for showing things like how hard players are working or anything like that, as the “average speed” metric is simply distance over time, which treats a player walking from point A to B the same as the same player sprinting from A to B and waiting. But because it is simply a measure of distance moved, it can serve as useful window into how much teams move overall on offense or defense. Movement in and of itself isn’t a good or bad thing, as it is a purely descriptive rather than evaluative stat. Even using it as a descriptor requires a little bit of care, because a decent chunk of a team’s overall movement is changing ends of the floor from defense to offense or in reverse, so teams that change ends of the court more frequently, that is play at a higher pace, will see higher numbers in the movement data without necessarily doing more in their half-court sets.That said, over the last three seasons as the Rockets have gone to a fairly extreme “iso-Harden” offensive philosophy, they have been an equally extreme outlier in terms of offensive movement relative to pace:&nbsp;Despite playing at the second-highest pace of the tracking data era (since 2013) heading into Sunday’s game, Houston has had the sixth-lowest average movement per second on offense, with the second- and third-lowest movement seasons coming in 2018-19 and 2017-18, respectively.Again, this isn’t a critique of the effectiveness of Houston’s offense, but it does illustrate how the double teams Harden is seeing largely come with the territory of the team structuring itself so completely around his individual creation ability and statistical accumulation.Finally, I wanted to help surface the best discussion I can remember for why the “Quick 2” strategy in end-of-game situations is a classic “” strategy. is a friend and the brains behind the invaluable . Here’s an excerpt from his recent post “:”To look at it another way let’s say if you shoot a 3 you will make it 25 percent of the time (I am intentionally choosing a low percentage here to demonstrate that even a low percentage 3 is a fine shot). Even if we assume that if you miss the shot you have zero chance of winning (which we know is not true — missed 3s are offensive rebounded ~20 percent of the time), given that teams without the ball in a tie game when the other team can hold for the last shot win 35-40 percent of the time, you win 9-10 percent of the time if you take that 3. Compare this to the 15 percent of the time teams down 1 without the ball and 15-20 seconds left and this basically means that the only 2 point shot that is better than a bad 3 is a good shot right at the rim (a 70+% shot). This is the other mistake people make when saying “you don’t need a 3”, it’s really hard to get a 2 point shot that good.(Emphasis mine).As Darryl alludes to, half of the fallacy of the Quick 2 strategy is that the 2-pointer has to be so high percentage that if a team could generate such a shot on command, why is that team saving it for a time when it has already probably lost? The other part is the “quick” part. It’s tough to fully measure, but how often does a team try to go for the quick two, get its designed option shut down, and end up running 10 seconds it doesn’t have off the clock?One of the reasons I’m somewhat uncomfortable being especially critical of individual coaching moves and strategies is these decisions are incredibly complex and involve a lot of information not readily available to us on the outside. My preferred analogy is to say in most circumstances, coaching is like cooking: you have a general idea of the dish you’re trying to make and have the ingredients in place, but actually putting the meal together involves a pinch of this, a dash of that, salt to taste, etc.However, some situations are much more like baking, requiring near scientific rations of ingredients, temperatures and times to produce a predictable outcome. Many aspects of end-of-game strategy, such as when attempting a 3-pointer is critical, when to foul and perhaps even timeout usage.As Darryl also notes: “another thing I think teams get wrong is not saving their last timeout to advance the ball with under 10 seconds left,” a sentiment with which I fully agree. The quick two is one of those situations where coaches routinely take a strategy that effectively reduces their chances of winning a contest from slim to none. Given how small the initial win probabilities are in situations where a team is trailing late, incorrectly going for two probably doesn’t cause too much of a leak in the average team’s record. But at the same time, fixing it is essentially costless, and pile up enough theoretical tenths of wins and soon enough we’re talking about real money.(Photo: Joe Murphy / NBAE via Getty Images)

Nov 25, 2019In this week’s installment, I want to quickly touch on a few topics from the last week.First, I want to revisit the concept of marginal contributions on the team level. Using a framework of marginal gains or losses is extremely helpful across the gamut of topics for analysis, as when we are trying to determine the “value” of a given player, team or strategy, it’s always relative. There are 1,230 wins to go around in an NBA season — . I’m perhaps too much of a traditionalist to be supportive of the sort of omnibus approach to schedule changes that has been floated, but discussion...

public

Seth Partnow

Get every sports story that matters. Download The Athletic app: https://linktr.ee/TheAthletic

NBA Analytics Lookaround: Taking an offense from good to great, LeBron’s free throws, Harden’s usage and more

No article scores yet.

CRITIC REVIEWS

PUBLIC REVIEWS