Freelance Friday: 1s & 2s

Freelance Friday is a regular series at Nylon Calculus featuring guest posts from around the web. If you have an idea or submission for a future Freelance Friday post, get in touch on twitter, Facebook or email TheNylonCalculus at gmail dot com. This week’s post is a semi-serious response to Kirk Goldsberry’s plan to fix pickup basketball from Michael Hartnett. Michael is a freelance basketball writer from Syracuse, NY and can be reached on twitter @mikehartnett45.

Pick-up Basketball Scoring isn’t Broken, It Actually Encourages Good Basketball

Earlier this week Kirk Goldsberry had a fascinating examination of the proper scoring system to use in pick-up basketball.  This represents a necessary and long overdue introduction of #analytics into the pick-up scene and should resonate deeply with anyone who has ever called “next” on the hardwood.  

Goldsberry makes the case that based on the fact that the expected point value for a shot outside the arc is so much more valued (33% increase versus a 0% increase) when we play by the “1’s & 2’s” system that dominates casual basketball, pick-up players should use the “2’s and 3’s” system employed in organized basketball to better accommodate players of different skill sets.  The argument is reasonable as a first pass, but important aspects of  pick-up hoops may have been missed.  Anyone who has played pick-up extensively will tell you that games are much more fun when multiple people touch the ball on offense, players are finding the open man, and the lane isn’t clogged with 6’5” 270 lb. Roy Hibbert wannabe’s waiting to flagrantly foul unsuspecting drivers.  

In other words, it’s more fun to play like the Spurs and the Warriors than it is to play like the 2015 Finals-model Cavaliers.  While winning in pick-up basketball is a major goal [1.This is a gross understatement.], getting out there and having fun is a substantially more valued in pick-up games than it is in the NBA. Most casual players don’t have millions of dollars on the line if they lose [2. Despite the occasional seemingly life-changing argument, to hear the volume and stridency of opposing viewpoints as far as who touched it last, who called “first” or most often, “what’s the count?”] and don’t have to deal with hordes of angry fans on twitter if they switch teams or commit a costly turnover.   So the question then becomes, “what coring system best encourages ball movement and ‘good basketball?’”

“Wide-Open” three pointers are usually the result of extra passes, picks, and unselfish play so I’m going to use these quality looks as indicators of “good basketball”.  A scoring system which rewards “open threes” while at the same time discouraging  tightly contested threes should lead  to fun basketball for all involved.  

Using  SportVU[1. http://stats.nba.com/league/team/shots/#!/] data from the  2014/2015 NBA season, I sought to determine to see how much more accurate  players as a whole were on wide open versus tightly contested shots.  On aggregate, NBA players shot 35% overall from three point range. This figure includes 38.3% on “wide open[3. as defined by no defender within 6 feet at the time of release.]” shots, but only 26.4% on “tightly contested[4. as defined by guarded within two feet]” shots.  Dividing the situational three point percentages by overall three point percentage, we can extrapolate how much three point percentages change when shooters are wide open or heavily contested. A wide open shot is about 10% more accurate than average, while a heavily contested one only goes in 3/4ths as often as do the average attempt. Of course, teams shoot more open threes than contested threes and some of these heavily contested looks come at the end of the shot clock, a non-factor in pick-up basketball –  

Applying these ratios to the data from UNC pick-up ball collected by Samuel Reis-Dennis[4. Cited in the Goldsberry’s original article.] here is the expected scoring of wide open and contested threes under each scoring system[3. Adjusted Expected value= value of shot* situational shooting percentage.]:  

So  playing by 1s and 2s actively encourages wide open “outside the arc” shots over most other shots, while playing by 2s and 3s  discourages contested “outside the arc” shots without placing much emphasis on sacrificing shots inside the arc for open shots “outside the arc”.  

The distinction between encouraging good shots as opposed to discouraging awful shots is important when we look back to trends in pick-up basketball.  [1. This is all anecdotal but until we get SportVU cameras in every YMCA or program fitbits specifically for pick-up basketball, anecdotes are what we get.]

So  playing by f1s and 2s actively encourages wide open “outside the arc” shots over most other shots, while playing by 2s and 3s  discourages contested “outside the arc” shots without placing much emphasis on sacrificing shots inside the arc for open shots “outside the arc”.  

The distinction between encouraging good shots as opposed to discouraging awful shots is important when we look back to trends in pick-up basketball.  [1. This is all anecdotal but until we get SportVU cameras in every YMCA or program fitbits specifically for pick-up basketball, anecdotes are what we get.]

Enjoyable pick-up games are not the games Reis-Dennis experienced at Cornell playing by ones only,[3. Ed. I would like the record to reflect that I prefer only ones, even as an inveterate long-range chucker. There is something about the physical grind of “every bucket matters” that appeals to me. Plus, if it’s good enough for Randy Moss, it’s good enough for me.]  where unskilled defenders were free to clog the lane, crash the glass, and foul hard without competitive consequences.  However shifting to more standard “game” scoring  might not fix this as much as Goldsberry would hope.  

Playing by 2s & 3s, rec-league Matthew Dellavedova [3. Which is probably the equivalent of having NBA Matthew Dellavedova in your rec league.] is free to leave his man at arc crash the boards and commit “intelligent[4. But never dirty. Only scrapy.] fouls, knowing that he’s surrendering 0.08 expected points.  [5. Less than half a percent  of a Game to 21.]  if someone kicks it to his man.  Playing by 1’s & 2’s, defenders are forced to leave lanes open knowing that if they drift too far they sacrifice a whole 0.19 additional expected points if the ball gets kicked, 4 times as much value in a game to 11 by 1s & 2s[6. Again, relative to the score target.]  Pick-up players being as intrinsically rational and analytically savvy as they are will internalize this difference and stick closer to their men, leaving lanes open. Open lanes means fewer fouls, more opportunity for skilled players to excel, and more fun for everyone.

Another infuriating occurrence of pick-up basketball is getting paired up with YMCA Dion Waiters, the reckless chucker who takes the inbound pass after a made basket as a license to heedlessly drive, leaving  his teammates looking like, well, Dion Waiters[4.  I love you Dion, when you blocked Darius Johnson-Odom in the 2012 Marquette game it was perfection #315forever #humbleandhungry].

The same math legislating against rec league Delly applies to this ball-stopping gunner. Playing by 2s  & 3s , the penalty in terms of expectation [1. gaining 0.19 expected points by passing as opposed to sacrificing 0.08 expected points by gunning.] for not making the extra pass as it would be playing 1s & 2s. Even YMCA Waiters can do that math and might even surrender the joys of getting BCKTS for the additional points.  It’s a big difference.

So despite what the haters [1. @Grantland, @kirkgoldsberry] might say, pick-up basketball scoring isn’t quite broken.  In fact, it may actually be encouraging good basketball habits, passing, and spacing by placing increased emphasis on “wide open” threes over any other shot.  Of course without detailed ball movement and player fatigue data, these results should be treated as preliminary. But until we get better data and pick-up players everywhere should stick to 1’s & 2’s.