Three Point Shooting

I've been mulling an in-depth post about three-point shooting for awhile.  Citizen John R's diary last week about 'risky' strategies in basketball dovetails nicely with some of my thoughts, and with a couple days off in the game schedule, now seems like as good a time as any.

Watching the Suns dismantle the Cavs last night, and in light of their massive success for the last two and half seasons, one can't help but ask one's self if their style can be duplicated.  Some teams (like the Warriors, Nuggets and now the Grizzlies) have tried recently to varying degrees of success.

And make no mistake, the Suns' recent success has been massive and is very much related to the style of play.  In 03-04, a Suns team featuring Shawn Marion, Joe Johnson, Amare Stoudemire, Leandro Barbosa and Stephon Marbury won 29 games.  Sure, Barbosa was a rookie, Stoudemire was hurt some and Marbury was eventually traded.  But that team had huge talent, and won 28 games.  The following season, they won 62 games, losing in the Western Conference finals with Joe Johnson.  Last season, they won 54 games and lost in the WCF, with Amare hurt the whole year.  

Obviously, a personnel change in the form of Steve Nash has had an impact.  But Steve Nash alone does not account for this.  Marbury, as much as we dislike him, is a talented basketball player.  And Steve Nash (playing with Dirk Nowitzki) only made it to the conference finals once in Dallas.

Still, it's tempting to conclude that the Suns are a unique combination of the right personnel in the right system, that it works for them, but would not work for other teams.  This is what Mike Dunleavy would have us believe.  The Suns style of running, spreading the floor and shooting a lot of threes works for them, but the Clippers style of walking it up, pounding the post and shooting the fewest threes in the league fits their personnel and is equally valid.  

But wait a minute.  It's not just the Suns shooting the three.  BY FAR the three best teams in the NBA are the Suns, Mavs and Spurs.  And they are ranked 1 (Spurs), 2 (Suns) and 3 (Mavs) in 3P%.  They are also all in the top 10 in 3PA, ranking 1 (Suns), 6 (Spurs), and 9 (Mavs).  So these teams attempt a lot of three point shots, make a high percentage of them, and win basketball games.  Interesting.

It goes beyond the big three.  The Rockers have the 4th best record in the NBA, take the third most threes, making 36% which is 9th in the league.  The Lakers have the 6th best record, take the 4th most threes, also making 36%.  Washington is arguably the hottest team in the East, and they're 4th in attempts, 5th in percentage (first in each among East teams).  Even Denver takes a lot (10th in the league), though they don't make them (last in percentage).  It's also interesting to note that it is more about the three ball than the running.  Obviously the Suns and Wizards play up tempo, but Houston under JVG is the epitome of `walk-it-up' basketball, while San Antonio and Dallas are definitely `defense-first' teams.

Of the teams that appear to be challengers, only Utah is in the bottom half of the league (26th) in 3FG attempts, and they make a decent percent (33%), and take almost 4 more per game than the Clippers (13.1 versus 9.3).  I suppose if you run your offense as well as Utah does, you can win without the 3 ball; the Clippers should take note of that.

If you look at the current playoff picture, and the list of teams by most three pointers attempted, the overlap is astounding.  

Note that I am using their conference rank based on record, not the NBA playoff seed.  Hence the Atlantic leader New Jersey has the 7th best record in the East, but would get the 4th seed in the playoffs.

If it were simply a matter of choosing between equally valid philosophies, we would expect the number of 3 pointers attempted to have no impact on a team's record.  MAKING a high percentage of them would of course always be good, but simply jacking up threes should not necessarily equal success.  But as you can see looking at the top half of the league in 3PA, fully 11 of the 15 are currently in the playoffs, 3 more are currently within 1 game of the playoffs, and Sacramento is having the worst luck, all of 2.5 games out of a playoff spot.  And two of the teams out of the playoff picture can certainly be said to be overachieving to even be close (Golden State and Toronto), the implication being that taking a high number of threes is improving their chances.  I want to emphasize this:  we're talking about attempts.  Intuitively, we wouldn't expect attempts to equal success, and about 53% of the teams at the top of the list would be in the playoffs since 16 of 30 teams make the playoffs.  The percentage of current playoff teams in the top half of the league in 3PA is actually 73%.  The percentage in the top 10 is 90%, with the overachieving Warriors trying to crash the party.  Wow.

There's definitely some self-fulfillment going on here.  Teams that shoot it well are going to shoot it more, and obviously shooting it well will yield more success than shooting it poorly.  But based on these numbers, simply using the weapon appears to be worth some wins, in and of itself.

There's a painfully obvious statistic that has always bugged me.  When you look strictly at 2 point field goal attempts, the league wide percentage this season is about 48.4%.  The league wide percentage on 3's is about 34.4%.  Well, 34.4% of 3 (1.032) is bigger than 48.4% of 2 (.968).  Over 6% bigger.  So why not take more 3's?  Why not take exclusively threes?

The answer of course is free throws, which you get a lot more of driving to the basket than standing and shooting jump shots.  But it does make it seem very odd to take a 20 footer.  You're not likely to get fouled taking a 20 foot jump shot, and it's still only worth 2 points, so why take it?  (I guess this is why so many players no longer have an `in-between' game - because it makes no sense to have one.)  Someone smarter than me has undoubtedly factored in shooting fouls (on twos and threes), and free throw percentages to calculate a true expected value for a 2 point attempt versus a 3 point attempt.  If any of you know where we could get those figures, let me know.

Having said all of that, let's look at the Clippers.  The Clippers are last in the league in 3PA (9.3 per game),  and 3PM (3), and 25th in 3P% (32%).  I have been an advocate of more outside shooting for a LONG time.  Going back to the old blog, on July 6 I recommended that the Clippers use their final three roster spots on one big, and two shooters.  They instead began the season with 2 more bigs (Davis and Aaron Williams), zero additional shooters, and an open roster spot.  On October 27, I listed 3 shooters on the waiver wire that the Clippers could pick up to fill that final spot, and at the head of the list was Luke Jackson.  75 days later, the Clippers signed Jackson to a 10 day contract, though he has yet to play.  (My advocacy of long-range shooting pre-dates my blog - I wanted them to go after Brent Barry in summer 2004.)

While the team has been struggling these first few months of the season, Baylor and MDSr have seemed to wake up to this need.  All the talk is of acquiring a shooter for Maggette, and now we have the Jackson signing.  But what took them so long?  Wasn't this problem evident long before the season started? Weren't the Clippers at or near the bottom in ALL the same statistical categories last year?  Why haven't they drafted a shooter in the last two years (sure Korolev is supposed to be a shooter, but he's not ready to contribute by a long shot, and they knew that)?  Why is Diaz in the Czech Republic?  Why didn't they sign or acquire another shooter this summer?  Eddie House signed for $1.5M this summer.  Eric Piatkowski signed for $1.2M.  Would House or Pike get major minutes on this Clippers team?  Maybe not, but at least we could bring in a shooter if we needed one.

MDSr seems to have a willful disregard for the three, which is strange for a guy who led the league in three pointers made and attempted in 82-83.  Of course MDSr's 67 makes that year were more than all but one other TEAM, while 6 players from the Suns made more last season, and that's not including Tim Thomas who only played 24 games for Phoenix.  So the mindset has changed more than a little since MDSr's playing days.  Still, you would think that he would be a pioneer, an early adapter.  

In his other head coaching experience with the Lakers, Bucks and Blazers, his teams were in the middle of the pack in terms of utilizing the three.  So certainly it does seem at least in part to be a function of MDSr using the strengths of his personnel.  But Cuttino Mobley averaged 5.4 3PA per game in Houston in 03-04 and 5.2 in Orlando and Sacto in 04-05.  Compare that to 3.1 with the Clippers last season and 2.3 this season.  As the Clippers only legitimate deep threat for most of last season, and one of two since last February, you'd think he'd be needed to take more three's, not fewer.  But MDSr doesn't want Cat taking threes.  Meanwhile San Antonio and Houston have shown pretty clearly that a strong inside presence is complemented and enhanced when surrounded by effective long range shooting.  It's great to have strong post players, but if you don't have shooters around them, the defense will pack in and suffocate the post.  

In fact, I would argue that the best way to survive without efficient three point shooting is by running, a la Denver.  The combination of running AND great three point shooting you get with Phoenix is simply lethal.  But if you want to walk the ball up the floor and to get your post players involved, you're also allowing the defense to get set.  And without deep threats around the perimeter, the defense's job gets a lot easier.

The post is usually going to draw a double-team, as Elton Brand has found out this season.  At that point, by definition, somebody is open.  The defense tries to rotate to cover the players closest to the ball, and run at those furthest away from the ball.  MDSr seems to think that kicking the ball out for an open mid-range shot is just as effective as kicking it out to the three point line.  It isn't.  Obviously, you get more points with the 3 (I won't go into all the math, but suffice it to say that 3 is bigger than 2).  But more importantly, the extra couple of steps for the defender to recover is significant.  The difference between a 20 footer and a 24 footer isn't just the extra point - the 24 footer may in fact be the better shot, simply because it will be an open shot, before the defender has arrived.  

Obviously, QRoss doesn't have that range right now (though wouldn't that be a great think for Q to be working on, a la Michael Cooper and Bruce Bowen).  And Sam is more comfortable closer in, despite his traditional ball (© Michael Smith) heroics, and Corey is shooting an abysmal 12% right now.  

Here's what I say: so what?  Corey shot 34% on 3's last season, and is at 32% for his career.  Sam can make them.  Singleton can make them.  Who knows if Ross can make them?  He's only taken 14 in his career, most probably desperation heaves at the end of the quarter.  So spread the floor and shoot the thing.  The defense still has to react to it, and good things will happen.  Besides, confidence is a huge factor.  If your coach tells you `Never shoot the three' you won't have a chance to make one.  If your coach says `Shoot it any time you're open' your percentage will skyrocket.  Why was the 2004 Athens Olympic team so much WORSE from outside than their NBA stats would suggest?  Because Larry Brown was waiting to pull them out of the game after a single miss.  Why is Matt Barnes making 40% this season?  Because Don Nelson told him to shoot the ball and not to worry.   And that's just with the existing personnel.  Wouldn't it be nice if we had a couple more shooters?  Like, say, Luke Jackson.

So I guess I don't agree with saying that the fewer threes you take, the fewer risks you are taking.  Obviously, you have to be able to make some of them, which the Clippers certainly can do with Mobley (career 38% on threes) and Thomas (37%) and even Cassell (33%) and Maggette (32%).  But to take so few (the Clippers are the only team in the league that takes fewer than 10 a game, and the league average is 16.7) is also a risk.  The three is a weapon, a significant weapon, in the NBA of 2007.  The best team's use all of the weapons.  Without this weapon, you cannot be an elite team.