Now You’ve Gone Too Far

In today’s 24-hour news cycle, as people like to call it, there’s a certain pressure to be reactionary.  Unless you’re incredibly insightful or far-sighted, it’s too hard to consistently generate novel ideas over the course of a 6 month (or longer) sports season.  Instead a lot of content reacts to what just happened yesterday (if not earlier today) and has a certain lack of perspective.  In other cases people latch on to a bad piece of information just to tell a story.  I’m going to take a quick look at two different such stories today since they both rely on team evaluation.

In a post that came up after the Heat lost to the Celtics last night, Michael Wallace says that the ‘pressure is starting to build’ in Miami since the Heat are only 5-4.  More crazily he says “The Heat now face the very real prospect of falling to .500 on Saturday, because there doesn’t appear to be any reason why the Raptors won’t be capable of exposing Miami’s defense the way the Celtics did on Thursday and the Utah Jazz did two days earlier”.

In a completely different realm, Dave Berri posted that the Detroit Lions are a top NFC team.  This is despite that fact that the Lions are 2-6.  The evidence supporting this comes from  This is a little different from Wallace’s post on the Heat since there’s at least some reasoning involved, but still mistaken in my opinion.

So how do we evaluate teams?  As most people who have looked into sports stats even a little know, it isn’t necessarily their win/loss record.  Instead, point differential or margin of victory is preferred.  If your team tends to win by more points on average, you should end up winning more games; if you win by more points than another team, you are better than that team.  Some people don’t think that’s enough, and they also try to account for a team’s opponents.  You might be winning every game by 10, but if it’s against the worst teams in the league it becomes less impressive.

And so let’s turn to the Heat.  The Heat are only 5-4, but they have a point differential of 9.4, good for second in the league.  They have also played Boston (twice), Orlando, New Orleans, and Utah.  Their losses are by 8, 5, 4, and 2 in overtime.  Yes, they’ve happened to come at the beginning of the season, but they aren’t deal-breakers.  The Celtics, for example, are 7-2 but have played Cleveland, the Knicks, Detroit, and Milwaukee; two of their wins were in overtime.  With a point differential of 5.2, we should expect Boston to slow down and Miami to catch up and pass them as the season progresses and the schedule starts to even out.  Now let’s look at the Raptors.  Toronto is 1-7; they have a point differential of -4.4.  They have also played some tough teams (Utah, the Lakers, Portland), but they mostly aren’t keeping games close.  In terms of them ‘exposing’ Miami’s defense, the general consensus is that you can do that with your big men, but Toronto has the non-scoring Reggie Evans and the criminally bad Andrea Bargnani in that department.  I don’t see anything happening in this game besides the Heat romping to a win.

As for the Lions, the evidence for their awesomeness comes from their positive point differential (tied for 6th in the NFC, but 15th in the league overall) and an apparently very difficult strength of schedule (1st in the NFC, 2nd in the league), which combines to make the Lions 8th in the league overall and 3rd in the NFC according to the simple rating system.  If we take SRS at face value, then the Lions indeed seem to be fairly good.  But we have to think a bit more about this one.  First, teams in the NFL have only played eight games.  Yes, that’s half the season, but it still isn’t many games.  Just as in the NBA, we shouldn’t be thinking too hard about point differential yet.  But, looking at my rankings (linked just below), the Lions have played six teams in the top half of the league (including 1, 5, and 7) and only two teams worse than themselves (Washington and Saint Louis, which happen to be their two wins).  So they certainly are facing quality opposition.

But, by my power rankings, the Lions are number 20 in the league.  They have 2.5 expected wins and 2 actual ones; they’re right on track.  They should win 3 or 4 of their last 8 games, so they’ll end up 5-11 or 6-10; a big improvement over the past couple years, but not great.  If you go by SRS though, the Lions would be favored in each of their remaining games except for the Packers and Patriots.  And here’s where SRS fails: it doesn’t know why the Lions have the rating they do.  It’s driven mostly by the 38 point win over the Rams; if the Lions had won by only, say, 17, their point differential would drop to -.75 and their SRS to 1.8, good for a tie for 13th.  Even then the Lions are probably over-rated: they are below average at passing and running efficiency on offense and defense, only average at turnovers, and they commit too many penalties.  That doesn’t sound like one of the best teams in the league, does it?

And now we can return to Miami.  I haven’t put together an NBA model so I can’t say which stats are predictive of future success, but in their games so far Miami is shooting better than their opponents, both overall and on threes, turning the ball over less, getting to the line more, getting more rebounds, getting more blocks, and they lead the league in points per shot while being third best in that category on defense.  In other words, it sure seems like they’re doing all the right things.  They just happen to have lost a few games early (maybe they should talk to the Chargers).  I certainly wouldn’t panic if I were them, even if everyone in the media is.  So thanks to Mike Wallace for the strong overreaction and Prof. Berri for looking at the wrong stats; it’s good to have a reminder to look for what really counts.

This entry was posted in Uncategorized and tagged , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s