Messing with statistics

[FourCamFord]

From Sports Business Daily

The Izod IndyCar Series saw its viewership drop 22% for 19 race telecasts across ABC and NBCSN compared to 15 telecasts last season. The series averaged 953,000 viewers, down from 1.2 million viewers last season. While IndyCar did see a drop in average viewership, an increase in the number...

https://www.sportsbusinessdaily.com/Daily/Issues/2013/10/24/Media/IndyCar.aspx

Sorry, you need a subscription to read the rest and I don't have one.

But here is the problem: They are averaging across a bunch of races, of which one dominates the total viewership. So the average is roughly inversely related to the number of races averaged, because they are dividing the same total viewership by a different number of events. They were dividing by 15 telecasts last year and 19 this year. The difference between 15 and 19 (relative to 19) is (ta-da) 21%!

[turbopanzer]

From Sports Business Daily

https://www.sportsbusinessdaily.com/Daily/Issues/2013/10/24/Media/IndyCar.aspx

Sorry, you need a subscription to read the rest and I don't have one.

But here is the problem: They are averaging across a bunch of races, of which one dominates the total viewership. So the average is roughly inversely related to the number of races averaged, because they are dividing the same total viewership by a different number of events. They were dividing by 15 telecasts last year and 19 this year. The difference between 15 and 19 (relative to 19) is (ta-da) 21%!

Knock out the Indy 500 data and divide by 18 and you get your new average. It still isn't good.

[Z28]

In any field if you use some statistics and omit others you can change the perception of what's going on. I only learned arithmetic but last year's 15 telecasts were the single event weekends while this year included the doubleheaders to raise it to 19. Those doubleheaders had viewer totals from both days which if they were both less than the single race last year dragged the average down twice whether you include the 500 or not.

[MM658]

Average viewers per race, by year, on Versus/NBCSN:

2009 - 315,000
2010 - 366,000
2011 - 402,000
2012 - 292,000
2013 - 276,000

FYI - that yearly average was 778,000 when the series was on ESPN/ESPN2/The Ocho for it's cable package.

[turbopanzer]

Average viewers per race, by year, on Versus/NBCSN:

2009 - 315,000
2010 - 366,000
2011 - 402,000
2012 - 292,000
2013 - 276,000

FYI - that yearly average was 778,000 when the series was on ESPN/ESPN2/The Ocho for it's cable package.

Based on this years numbers and the ESPN number, you have lost roughly 2/3rd's of your audience. At the rate they are going...it will be gone in roughly three to four years.

[Jakester]

Ninety percent of this game is half mental!

[FourCamFord]

Based on this years numbers and the ESPN number, you have lost roughly 2/3rd's of your audience. At the rate they are going...it will be gone in roughly three to four years.

But it doesn't look like a trend but rather occasional downward jumps. I can understand the effect of switching from ESPN to Versus/NBCSN, but what happened between 2011 and 2012? Danica departing? Wheldon crash? Ugly car? All of the above?

The 6% drop between 2012 and 2013 could easily be explained by the double header dilution effect Z28 mentions.

[turbopanzer]

But it doesn't look like a trend but rather occasional downward jumps. I can understand the effect of switching from ESPN to Versus/NBCSN, but what happened between 2011 and 2012? Danica departing? Wheldon crash? Ugly car? All of the above?

The 6% drop between 2012 and 2013 could easily be explained by the double header dilution effect Z28 mentions.

No question. But you are seeing a slow but steady decline. If you had momentary upward spikes you might be able to say it is a bottoming of ratings (similar to stock market stocks bottoming) but you are just seeing presently a consistent drop. If it continues, you basically just lose whats left of your base. 2014 better change or your just sinking further into oblivion.

[Ken]

Has it finnally reached the point it truly can't get any worse

[Indyknut]

Has it finnally reached the point it truly can't get any worse

I'm sure there's a number at some point where even a fledgling sports network like NBCSN would come to the realization that just about anything they stuck in those time slots would draw those kinds of numbers... just about anything.

I guess the question is what that number is and how long before ICS gets there.

[Red Byrd]

Has it finnally reached the point it truly can't get any worse

Where have you been the last 17 years?

It can always get worse.

[B.]

Has it finnally reached the point it truly can't get any worse

I remember watching John Andretti that night and thinking, "the point of which all you can think you have done to believe in so much through the faculties of all you ever could have imagined being will never be reached again."

I thought correctly.

[lkchris]

I'm sure there's a number at some point where even a fledgling sports network like NBCSN would come to the realization that just about anything they stuck in those time slots would draw those kinds of numbers... just about anything.

I guess the question is what that number is and how long before ICS gets there.

They can for sure find things that cost less to produce. Indycar will have to produce bigger numbers simply because of its cost. As GoDaddy stated, it's expensive advertising.