keyboard_arrow_uptop

Television ratings are a funny thing. The spin that can come out of the numbers can drive reports in wildly divergent directions. In sports, ratings can be spun to say that the popularity of a given league or club is high or low, depending on those feeding the information. Of course, leagues and clubs love to tout growth, while detractors can spin numbers negatively. For Major League Baseball, ratings have been used to show that the game’s popularity is on the rise, while others have pounded keys to say that it’s a “dying sport.”

So, which one is it? As is often the case in data analysis, the truth can lie in the middle. Before we get started, let’s give a quick primer on what the ratings numbers mean.

Nielsen ratings are a measurement put together by Nielsen Media Research seeking to measure the television audience size for programming in the United States. It does not cover Canada or other international markets where the programming might reach. While it’s the industry-standard, the system has its detractors. It’s not a wholly random sampling of U.S. television viewership. Nielsen picks a small number of viewers in comparison to all television households, and from there, extrapolates how many are viewing. Because of this, counts for communal viewing places like sports bars and college dorms are not factored into the numbers. Based on data provided by Nielsen (but no longer through their site, as the link off Wikipedia is now dead), “in 2009 of the 114,500,000 U.S. television households only 25,000 total American households (0.02183% of the total) participated in the Nielsen daily metered system.”

The system works like this: The numbers are split into two metrics that are joined together. One is the ratings number, the other denotes the “share.” An example would be this: 4.4 household rating/8 share. Nielsen presents this as 4.4 percent of all television households (in other words, homes that have TVs) tuning in to watch programming, while 8 percent of the people in the homes watched. This metric split makes sense: not all people in a household enjoy the same television programming. The other number that is often cited is total viewers, an indication of how many tuned in irrespective of percentage.

Was the 2012 All-Star Game a Ratings Hit or Miss?
The morning after the 83rd All-Star Game from Kansas City was played, MLB sent out a press release with the subject line that read in part, “All-Star Gameon FOX Enjoys TV Ratings Growth…” The press release went on to say that despite the 8-0 blowout by the National League, the “telecast on FOX drew an 8.1 metered market rating according to data from Nielsen, representing a +3% increase over the 7.9 from last year’s game in Phoenix.”

So, this was good, right? That depends. First off, you’ll note that I said, “the morning after.” That’s because Nielsen releases what they call “fast overnight” numbers. This is before a complete analysis is finalized, which gets released later in the day.

As to that three percent increase, yes, it’s truly an uptick, but consider this: last year was the lowest rated MLB All-Star Game in history. When accounting for that, the game was the second-lowest rated ever by the overnight numbers.

When the final numbers were released, here’s what the All-Star Game pulled in: 6.8/12 with 7.8 million viewers. That is the lowest ASG rating number ever (last year was 6.9) and marks the seventh consecutive year that ratings have dropped. As for share, it ties with last year’s number. Total viewership was down 149,000 from last year—the lowest decline in the last decade. In the last ten years, just three All-Star Games have seen total increases, with 2003 marking the first after the infamous 2002 tie in Milwaukee. The last ratings increase was between 2005 and 2006 (up 1.2).

So, based on this, MLB is a dying sport, right? Not so fast.

For FOX, the question is, how did the network fare against its rivals? Did FOX win the ratings battle for the night? The answer is yes. While the ratings were the lowest ever, the game still won the night for FOX, posting numbers 7.7 percent higher than any other program, according to Nielsen’s numbers. In that, FOX is happy; ultimately, a network is looking to win as many days in a year as possible.

The Good Old Days
Since Nielsen began tracking ratings numbers for the All-Star Game in 1967, the highest number posted is 28.5/54 in 1970. The most total household viewers were 16,670,000, or 8.87 million more viewers than this most recently watched game. Here’s where detractors go on the warpath, saying MLB is a dying sport.

To say as much is not taking into account the dramatic change in television channel offerings, the explosion of the internet as a competing media platform, the computer gaming revolution, and other changes in society that compete for time. In 1970, there were just the “big three” over-the-air networks in ABC, CBS, and NBC. In larger markets you may have had a PBS affiliate and possibly a UHF channel offering, but that was it. As a modern day comparison, DirecTV currently offers more than 300 channels, and more than 100 of those are in HD.

To add, back then the NFL wasn’t nearly the media juggernaut that it is now. Pete Rozelle, the former commissioner of the NFL and the man most credited with making the NFL what it is today, got ABC to begin broadcasting Monday Night Football in (you guessed it) 1970. Rozelle, in addition to creating the Super Bowl, was the first to sell media rights to CBS and NBC, which fueled competition for media rights not only within the NFL, but with MLB as well.

The best way to view MLB’s place in history as it pertains to television viewership is to say that baseball held a collective consciousness over other sports that it does not enjoy today. That is a byproduct of the game but also of its long and storied history. When you realize that the National League was formed in 1876, the American League was formed in 1901, and the World Series began in 1903—67 years before the creation of the Super Bowl—you have a leg up on your competition in terms of recognition. The pace of society has changed as well. Baseball’s pace was far more accepting with a younger demographic than it was prior.

What to Make of It
Times have changed. Ratings have waned. The numbers still win the day for the network hosting the broadcast. As emphasis on capturing an audience with live broadcasts continue to rake in major dollars (a byproduct of an explosion in media rights), ratings for baseball will continue to be a debated topic. One thing is certain: even though ratings have slipped, MLB will garner a sizeable increase in rights fees when the contracts with FOX, ESPN, and TBS expire no later than the end of next year. In that, networks still see incredible value in baseball as a sports programming staple. As to those saying that baseball is a “dying” sport, attendance will be up considerably (currently up between seven and eight percent), and viewership for the playoffs should be increase with the addition of two more Wild Card teams. But one thing is absolutely certain: spinning the ratings numbers will never cease.

Below are historical numbers for the MLB All-Star Game from 1967 to 2012 along with a graphical representation of ratings, share, and total viewers:

Year

Network

Rating

Share

Households

Diff Rating

Diff Households

2012

FOX

6.8

12

7,800,000

-0.1

-149,000

2011

FOX

6.9

12

7,949,000

-0.6

-686,000

2010

FOX

7.5

13

8,635,000

-1.4

-1,531,000

2009

FOX

8.9

15

10,166,000

-0.4

-275,000

2008

FOX

9.3

16

10,441,000

0.9

1,083,000

2007

FOX

8.4

15

9,358,000

-0.9

-890,000

2006

FOX

9.3

16

10,248,000

1.2

1,370,000

2005

FOX

8.1

14

8,878,000

-0.7

-626,000

2004

FOX

8.8

15

9,504,000

-0.7

-633,000

2003

FOX

9.5

17

10,137,000

0

114,000

2002

FOX

9.5

17

10,023,000

-1.5

-1,219,000

2001

FOX

11

19

11,242,000

0.9

1,061,000

2000

NBC

10.1

18

10,181,000

-1.9

-1,747,000

1999

FOX

12

22

11,928,000

-1.3

-1,106,000

1998

NBC

13.3

25

13,034,000

1.5

1,588,000

1997

FOX

11.8

21

11,446,000

-1.4

-1,213,000

1996

NBC

13.2

23

12,659,000

-0.7

-601,000

1995

ABC

13.9

25

13,260,000

-1.8

-1,530,000

1994

NBC

15.7

28

14,790,000

0.1

240,000

1993

CBS

15.6

28

14,550,000

0.7

830,000

1992

CBS

14.9

27

13,720,000

-2.5

-2,480,000

1991

CBS

17.4

32

16,200,000

1.2

1,260,000

1990

CBS

16.2

33

14,940,000

-2

-1,510,000

1989

NBC

18.2

33

16,450,000

-2.2

-1,620,000

1988

ABC

20.4

33

18,070,000

2.2

2,160,000

1987

NBC

18.2

37

15,910,000

-2.1

-1,530,000

1986

ABC

20.3

35

17,440,000

-0.2

40,000

1985

NBC

20.5

36

17,400,000

0.4

560,000

1984

ABC

20.1

35

16,840,000

-1.4

-1,070,000

1983

NBC

21.5

39

17,910,000

-3.5

-2,470,000

1982

ABC

25

44

20,380,000

4.9

4,740,000

1981

NBC

20.1

36

15,640,000

-6.7

-4,810,000

1980

ABC

26.8

46

20,450,000

2.4

2,270,000

1979

NBC

24.4

45

18,180,000

-1.7

-850,000

1978

ABC

26.1

47

19,030,000

1.6

1,590,000

1977

NBC

24.5

45

17,440,000

-2.6

-1,240,000

1976

ABC

27.1

53

18,680,000

5.6

3,950,000

1975

NBC

21.5

41

14,730,000

-1.9

-760,000

1974

NBC

23.4

44

15,490,000

-0.4

70,000

1973

NBC

23.8

45

15,420,000

0.9

1,200,000

1972

NBC

22.9

43

14,220,000

-4.1

-2,010,000

1971

NBC

27

50

16,230,000

-1.5

-440,000

1970

NBC

28.5

54

16,670,000

13.4

8,060,000

1969

NBC-d

15.1

42

8,610,000

-10.7

-5,840,000

1968

NBC

25.8

49

14,450,000

0.2

400,000

1967

NBC

25.6

50

14,050,000

25.6

14,050,000

Source: Nielsen Media Research

Legend:

  • Year – Year of game
  • Network – Network broadcasting game
  • Rating – As based on Nielsen ratings number
  • Share – As based on Nielsen ratings number
  • Households – Total number of television households viewing
  • Diff Rating – The difference in ratings, plus or minus, from the year prior
  • Diff Households – The difference in TV housholds, plus or minus, from the year prior
  • d – Day Game
You need to be logged in to comment. Login or Subscribe
TangoTiger1
7/16
Maury, the "control group" would be World Series. So, can you show the ratings of the World Series (say Game 1) for every year, and show the % of that audience that the All-Star game got?
mbrown
7/16
That would be an article for after this year's World Series
Behemoth
7/16
Is there not an issue in that World Series viewing figures will be impacted significantly by the teams involved. You'd presumably get higher viewing for Yankees-Mets or Yankees-Dodgers than, say, Rays-Reds.
mbrown
7/16
Tom, I'd also add that while Game 1 is certainly a good barometer, you add other variables in such as how teams are geographically. A NYM/NYY or LAA/LAD series chokes ratings due to proximity. In other words, in some senses I would advocate that the ASG, while not perfect, actually works as a decent control group. Its format spans all clubs and therefore its interest should not ebb and flow based on the "teams" partisipating.
Behemoth
7/16
But then the intensity of the game has diminished over time, and there isn't the same opportunity to see players you wouldn't otherwise see. This would suggest that viewing figures would, all other things being equal, shrink over time.
mbrown
7/16
I'd say this is a possible variable. But, it would really be in the context of core fans. With ratings, you're trying to gauge interest to average fan. I don't know if the intensity factor is something that is considered. The exception would be if we had a "Pete Rose charging home" type scenario. That would get played (likely repeatedly) on Sportscenter, which could slingshot into the following seasob's ASG ratings.
Behemoth
7/16
I just think that the ASG now, and the ASG forty years ago are sufficiently different that all the data tells us is that the ASG is less popular now. If there's a way round the adjustments required to look at the World Series, and you guys are much more qualified to look at that than I am, then I think it's a much better indicator of overall interest in baseball.
TangoTiger1
7/16
You can adjust for the teams involved, if you insist. But at the very least, the first step is to present the data. Is it the dramatic falloff that we see with the All-Star game (I'd bet no). World Series games still appear in the "top 10" events of the week. The All-Star game used to. Does it still? So, before we start to talk about how we need to adjust (and we do), the first step is to present the unadjusted data. Then, let's worry about how much the adjustment can really impact things.
mbrown
7/16
If you have a method by which to adjust for regional alignment, I'm dying to see it
mbrown
7/16
I'd also add this: ASG does not conflict with other Big-4 sports programming. At this point, NFL, NBA, NHL in slumber.
gweedoh565
7/16
I think the message here is that the Nielsen ratings need a Ratings Above Replacement stat to normalize across eras.
Tythelip
7/16
The tie was in Milwaukee.
bornyank1
7/16
Fixed.
mbrown
7/17
Much thanks, Ben
TangoTiger1
7/16
Maury, can't be too hard, can it? Put teams in various market size, find World Series over the years that have similar markets involved. Dodgers/A's played in both 1974 and 1988. But, Angels/Giants played in 2002. They probaby match up well, market-size-wise. Phillies/Royals in 1980 is smaller market than Phillies/Yanks in 2009. How much difference was the TV ratings? Cards/Rangers v Cards/Tigers v Cards/Royals v Cards/Twins v Cards/Redsox v Cards/Brewers. Did the ratings follow the market size? By how much? I think if you sit down and try different models, you can come up with something reasonable.
TangoTiger1
7/16
Maury, can't be too hard, can it? Put teams in various market size, find World Series over the years that have similar markets involved. Dodgers/A's played in both 1974 and 1988. But, Angels/Giants played in 2002. They probaby match up well, market-size-wise. Phillies/Royals in 1980 is smaller market than Phillies/Yanks in 2009. How much difference was the TV ratings? Cards/Rangers v Cards/Tigers v Cards/Royals v Cards/Twins v Cards/Redsox v Cards/Brewers. Did the ratings follow the market size? By how much? I think if you sit down and try different models, you can come up with something reasonable.
mbrown
7/16
More importantly Tom, there needs to be a means by which to account for changes in media landscape. There has been a massive shift, even since 2002, by which additional entertainment options compete with television media (Netflix, computer gaming, league-owned sports channels, horizontal offerings by networks, continued expansion of cable/satillite operators). We should go offline. There would need to be several variables and a means to normalizing data to allow for a true understanding of ratings against the environment. You know my email address.
mbrown
7/16
I would add this: there's little doubting that the NFL has taken a massive chunk out of interest in MLB. Ratings and total viewership numbers have skyrocketed by comparison. Pace of the game, it's ability to be well defined by time (I.e. you know within a close proximity when games will end) and fact that games aren't played daily, a sign that society has more competing interests than even 20 yrs ago, play into the growth of NFL and decline of MLB
mbrown
7/16
Oh, and a football field is the perfect size for theater aspect HD systems.