Statistical Analysis. Humor. Knicks.

Sunday, October 26, 2014

KnickerBlogger’s Anti-Tank Idea

The NBA’s dirty little secret is out, and everyone knows that teams are intentionally losing games down the stretch. Franchises that have been eliminated from the playoffs and held on to their pick (sorry Knick fans) can reward themselves by losing games down the stretch. And I can’t say I blame them. Athletes are trained from day 1 that winning is the ultimate goal (right Herm?) and a lot of players will resort to just about any means that accomplishes that goal. I’m sure Knick fans aren’t outraged when Malik Rose gets a handful of jersey when he performs his “pull the chair out from under the guy” routine. While an illegal move, if he can get away with it, Rose would be foolish not to keep it in his repertoire. The same goes for the league’s franchises. Would Milwaukee or Memphis or Boston be doing their team a disservice by trying to win down the stretch, when they can put an inferior lineup on the floor? Yes, as long as they can get away with it.

There has been some discussion in the media about possible solutions. One idea, which I think Mike Wilbon of PTI fame has been touting, would be to give all non-playoff teams an equal chance at the lottery (or the “one team one envelope” rule). The downside to this solution is that teams that really need help may not get it, which is antithetical to the draft’s purpose. Imagine if the Clippers or Pacers landed that #1 overall pick this year, while Boston or Memphis sat at #14. A team could finish in last place for 3 straight seasons, and would only have a 51% of getting one top 3 pick (for those scoring with a calculator at home that equation is 1-[11/14]^3). Not only would this solution cause an imbalance in the league, but it would give conspiracy theorists something else to harp on. To this day there are people convinced that Patrick Ewing to the Knicks was an NBA orchestrated event.

Bill Simmons has proposed a tournament where the top 6 teams in each conference are guaranteed playoff spots, and everyone else plays for those last remaining playoffs spots. It’s an interesting concept, but it’s just as easy to circumvent. No one in their right mind would think that if Boston or Memphis won a mini-tournament, they could go on and take the Pistons or Mavs in 7. So this doesn’t really address the problem. Why would a team risk losing a franchise player like Durant or Oden in order to have the privilege of getting spanked by the first or second seed? Teams will be tanking games in the tournament just as they would if it were a regular season game. In fact they would only have to purposely lose one game with this method.

Other solutions include handing out fines to teams that tank, shortening the season, and eliminating the lottery altogether. David Stern’s office could fine teams that are throwing games, but this would be a hard rule to enforce. Often teams have players fake injuries, and disproving something like knee tendinitis would be impossible (right Steve?). And an eliminated team could say they’re trying to give extra playing time to their end of bench guys. Shortening the season would take revenue from both the players and owners, so that option is out the window. And removing the lottery would just exacerbate the problem. In fact that’s what the lottery was created for in the first place, so that teams wouldn’t tank down the stretch.

So what’s a league to do? Here is a fool proof solution: set the lottery order earlier in the season, like at the All Star Game. In other words take a snapshot of the standings at the the All Star break and use that as a basis for the lottery. Obviously only the teams that fail to make the playoffs will participate in the lottery. The only teams that this might give an advantage to are teams like the Sixers who have a good second half. But then again, that’s what we want bad teams to do, win games down the stretch (and Philly was trying to rebuild with the Iverson trade). No team is going to start the season losing, because attendance is linked to winning percentage. And also they might have a Cinderella team in the making (2005 Sonics, I’m looking at you), which would net them profit due to a playoff series (7 games series means that both teams get at least 2 home games).

Below is a chart with the lottery team’s All Star Game ranking (ASG), and their end of season ranking (EOS).

ASG Rank EOS Rank Team
11 20 Indiana Pacers
16 24 Minnesota Timberwolves
17 17 Los Angeles Clippers
18 18 New Orleans Hornets
21 22 New York Knicks
22 21 Sacramento Kings
23 25 Portland Trail Blazers
24 27 Atlanta Hawks
25 26 Seattle SuperSonics
26 23 Charlotte Bobcats
27 28 Milwaukee Bucks
28 19 Philadelphia 76ers
29 30 Memphis Grizzlies
30 29 Boston Celtics

7 comments on “KnickerBlogger’s Anti-Tank Idea

  1. Sly Williams

    The problem with this ‘solution’ is 2-fold:
    1) it attempts to fix a problem that has not been sufficiently identified as an issue
    2) it wouldn’t change much anyway

    The only presented evidence of ‘tanking’ is a reduction of playing time for the best players on the team in games towards the end of the season. Teams also use this strategy after they solidify their playoff position (see Dallas, Denver, SA, etc). The last game Denver vs. SA had no: Iverson; Anthony; Nene; Camby; Duncan; Parker; or Ginobli. If this type of strategy occurred with a non-playoff team, it would be considered evidence of tanking. No – it would be considered the only evidence of tanking. Clearly there are other causes.
    It is also a good strategy (!) to rest those players, regardless of draft position. It helps identify those players that may be competing for a spot on the roster the next year. I would love to have seen how Randolph Morris played in the last few games, rather than pushing Curry over 40 min/g. It also reduces the risk of major injury to the most important players.

    Knickeblogger, you and some of the rest (sports law; Hollinger) are the writers I most respect, and I have always felt there was tanking. But the more I read on this subject, the less I am convinced of a tanking problem. It is a long season. Maybe all teams (other than those coached by Thomas) want some rest.

  2. Ted Nelson

    A problem with setting the lottery just over half way through the season could be a team that starts 0-10 or so just mailing in the rest of the season. Or, in a draft as hyped as this one, teams tanking from game 1 to the All-Star Game then bringing back their “injured” players and trying to get things turned around in the second half to build some momentum going into year 1 of the Greg Oden, Kevin Durant, or whoever era.

  3. Hudson River

    I think this is a fantastic idea, maybe a little bit later than the all-star break, maybe after 60 games. David Stern should seriously consider this proposition.

  4. Joe C

    I like this idea. To reframe it a little, the point is to continues to identify the bad teams so they can be helped, while eliminating the obvious opportunities for teams to game the system.

    Under the current system, when you’re mathematically eliminated from the playoffs, it’s “obvious” that the best outcome you can attempt to influence is your lottery status.
    When you haven’t yet been eliminated (either mathematically or realistically), it’s a lot less “obvious” that you should be playing for a better lottery pick — you still have the chance to make the playoffs, and isn’t that a far better outcome for your team than a high draft pick?

    I believe that’s the crux of your idea – pick the lottery set point when most teams haven’t been eliminated yet.

    How about you add one more element to the system to remove other points where gaming may be the obviously better strategy. For example, if you know that the order is going to be set at the All-Star break (or some other fixed point), there may be scenarios where losing one or two right then is worthwhile.
    My solution would be to also randomize the point at which the lottery is set – and wait until after the season is completed to pick that point. You would want to keep it in a reasonably narrow band, say games 45 – 55.

    That way gaming becomes much less compelling, because you don’t know when the right time to tank is. It also would add some uncertainty add the end of the season – if a team is deciding between playing for the 8th seed or their lottery draw, they may not know exactly what their draw is.

  5. BrandonM

    I was going to make a post on Knicksonline.com on Simmons “tourney” idea b/c I really think it would be great for the future of the NBA. Even when the Knicks were in the post season the playoffs always just dragged and they still do. Even rivalries are hard to watch b/c it is just too much, and the NBA historically has been “sleeper weak” which always leads to good theatre. If we had an opportunity to play one game tourneys at the beginning, that would be something to watch. They should really consider a rendition of the NCAA tourney because it is so much more worth watching, even though the quality of play is not up to par with NBA.

Comments are closed.