Impact of Stadium Size: For Your Consideration

[RonCo]

In my various meanderings around the OOTP world, I’ve often wondered if stadium size itself is an attractor of attendance. Or, putting it more succinctly, does merely the act of adding seats create more attendance? If so, that would be a bit wonky, but I can see “gamey” reasons why it would be put in code that way. I’ll not bore you with my mad wanderings there, but simply leave you with the thought.

If this is true, for example, it means that attendance isn’t like water—meaning that (given fan interest and winning) you just pour it into the stadium until the stadium is full, and when that happens you need to get more seats. As business folks, that’s how we think of it, right? We look at our 40,000 seat stadium and see we’re only drawing 24,000, so we don’t need to add more right now. But, again, if my little supposition is true, then simply adding seats would draw more fans.

Let me state again: I do not know this is true.

I can only guess.

Though, there’s this to ponder …

We are now one sim into the off-season, meaning season tickets have been on sale for long enough to get a small snapshot of what attendance might eventually look like. During this period, all the factors are essentially static: ticket price, fan interest, loyalty, and stadium capacity did not change from the beginning of the sim to the end. In addition, as has been noted, 100% of the BBA’s Market Size has been adjusted to be identical (for whatever that matters).

So I took our data, and ran a correlation study to see how tickets sold (attendance) correlated to Ticket Price, 30-Day Fan Interest Changes, Fan Interest, and, yes, Stadium Capacity. Here are the correlations…


FAN INTEREST (.675)

To probably no one’s surprise, tickets sold were highly correlated to Fan Interest (.675). I think we can all agree that this makes a lot of sense.


30-DAY FAN INTEREST (.068)

This is such a low value that it’s probably fair to ignore it. In other words, if you’re fan interest changed to a great degree over the last 30 days it does not seem to really matter. What matters is what the change resulted in.


TICKET PRICE (-.348)

Also, perhaps of obvious nature, the price of tickets correlated inversely to sales, meaning that as tickets went up, sales went down (or when prices went down, sales went up). This is an obvious correlation. Of, perhaps, a bit more interest is that the magnitude of correlation is moderate (-.349 is large enough to suggest there’s something real in the data to me), and that it’s less than Fan Interest.

So, between the two, this particular set of data suggests that while both Fan Interest and Ticket Price matter, Fan Interest has the slightly bigger sway.

All right then, let’s get to the meat of this little study … which I again caveat in all ways. But here we go:


STADIUM CAPACITY (.495)

So there you go. Stadium Capacity (.495) correlates to tickets sold to a degree that is lower than Fan Interest (.674) , but greater than ticket prices (-.348) . If this one study actually holds water, it means BBA GMs might well be making a great financial investment in adding to seats. Or, put another way, adding seats might be giving teams an actual competitive advantage at the turnstiles.

[shoeless.db]

Bigger is better. Sorry, @jleddy .

[RichY]

I have been thinking about this recently after listening to your BBA finance podcast episode. Especially as I am a bit in the dark finding the ticket price sweet spot as an expansion team with no previous ticket sales history. Interesting stuff. Thanks Ron.

[GoldenOne]

Size does not always matter.

Sorry, too easy.

[RonCo]

Size appears to matter a little less than interest and a little more than how much you paid?


Hmmmm....

[Fat Nige]

The first year in EP I sold 22,700 season tickets in a 55,000 stadium at $16 each. About 7,000 up on the final year at Wichita at nearly double the price (with around 90 FI). I ended the season a wee bit shy of a 10,000 increase on the last Wichita average attendance.

This year I’m offering them out at $12 with 94 FI to see if I can break the 30,000 mark

[Lane]

very interesting, Ron. Looks like my competitive advantage of having the BBA's largest stadium will be short lived!

[CTBrewCrew]

Madison is hung like s tuna can!

[allenciox]

Is this a stepwise regression, or did you just do raw bi-variate correlations?

[DugoutDesperado]

I will admit that I did not give much thought to the seating capacity. I opted for the 55,000 seats in the hopes that I have enough 45,000+ games to somewhat offset the additional $7M cost over the life of the stadium (works out to about 44 sellouts - if that is over a 10 year period then ab out 5 per year - not counting the crowds between 45k and 55k).

The fan interest bump was a big piece of saying that we may be able to fill enough of the seats if we can figure out a way to play above.500 ball. Time will tell - projected season ticket revenue is supposed to be $7.7M more than last year.

[RonCo]

Is this a stepwise regression, or did you just do raw bi-variate correlations?

Raw correlations.

[allenciox]

A stepwise regression would be useful for dealing with correlations between your predictor terms... what "falls out" when you account for codependencies.

[RonCo]

Regression on current numbers:

	Coefficients	Std Error	t Stat	P-value	Lower 95%	Upper 95%	Lower 95.0%	Upper 95.0%
Intercept	-4053.963	2644.421	-1.533	0.136904803	-9479.87	1371.94	-9479.87	1371.94
TktPrice	-355.629	71.470	-4.976	3.24921E-05	-502.27	-208.98	-502.27	-208.98
Capacity	0.154	0.042	3.640	0.001138068	0.07	0.24	0.07	0.24
30-dayFI Δ	-65.435	71.322	-0.917	0.36702677	-211.78	80.91	-211.78	80.91
Fan Interest	178.889	23.848	7.501	4.54159E-08	129.96	227.82	129.96	227.82

If I'm reading this right, Fan Interest and ticket price are primary impacts, with capacity significant but small?

I need to understand the Lower and Upper bits better.

[allenciox]

Pretty much. The three impacts that are meaningful are 1) +Fan Interest (7.5 t score), 2) -TktPrice ( - 4.98 t score, 3) +Capacity (3.64 t score).

Nothing else is significant. It would be interesting to see how it changes if you only include those three predictors. Capacity is almost as important as ticket price but is less than half as important as Fan Interest.

Interesting!

[RonCo]

Updated to current numbers, dropping 30-day fan interest....

	Coefficients	Standard Error	t Stat	P-value	Lower 95%	Upper 95%	Lower 95.0%	Upper 95.0%
Intercept	-4377.540	3716.318	-1.178	0.248741874	-11990.072	3234.993	-11990.072	3234.993
TktPrice	-473.703	98.928	-4.788	4.94636E-05	-676.347	-271.058	-676.347	-271.058
Capacity	0.214	0.058	3.677	0.000992724	0.095	0.333	0.095	0.333
Fan Interest	228.032	31.173	7.315	5.77936E-08	164.177	291.887	164.177	291.887

Looks like about the same story to me.

[allenciox]

So there are two different ways to interpret the "capacity" significance. One is that it could be due to a "ceiling" affect: i.e., teams whose attendance is close to filling up the stadium every (or most) games are of course affected by the capacity.

The other is that it could be due to people wanting to come "fill up" the stadium a little more when there is plenty of space available.

I am wondering, would it be possible to tease these out, by running the regression separately for teams that are less than 75% capacity on average and those that are more than 75% capacity?

[crobillard]

Nerds

[GoldenOne]

So, my stadium size is 40K, all I really need/want to know is whether or not it would be worth the expense to add seats and by how many.

I dont do fuzzy math.

[crobillard]

So, my stadium size is 40K, all I really need/want to know is whether or not it would be worth the expense to add seats and by how many.

I dont do fuzzy math.

There's an offseason cap of 5k seats. That would be 50 ppts and 3.5 mil I believe.

[GoldenOne]

So, my stadium size is 40K, all I really need/want to know is whether or not it would be worth the expense to add seats and by how many.

I dont do fuzzy math.

There's an offseason cap of 5k seats. That would be 50 ppts and 3.5 mil I believe.

Far too late for either of those right now anyway.