Greater Western Sydney Giants

[Boba Fett]

GWS average attendance at games in Sydney not including the home derby was 7,914 in its first season and grew 4.6% to 8,281 in its second season. After 5 games this season it is down to 8,244. 2 of those games, however, were rain affected.

To fill the stadium in 20 years they need 6% pa growth in attendance which is triple the population growth rate in Western Sydney.


FYI. Their latest Financial Report is linked below.

http://www.worldofwookie.com/afl/annual ... Report.pdf

Why would you not include the home derby? Does anyone do that for any other team? Of course not. It's just an excuse to bash GWS for some perceived under-achievement as a start-up AFL team.

[gyfox]

GWS average attendance at games in Sydney not including the home derby was 7,914 in its first season and grew 4.6% to 8,281 in its second season. After 5 games this season it is down to 8,244. 2 of those games, however, were rain affected.

To fill the stadium in 20 years they need 6% pa growth in attendance which is triple the population growth rate in Western Sydney.


FYI. Their latest Financial Report is linked below.

http://www.worldofwookie.com/afl/annual ... Report.pdf

GWS reports 18.2 million in AFL distributions and match receipts. The 2013 AFL annual report shows GWS' club distribution was 10.7 million. That leaves us with 7.5 million in match receipts. This strikes me as rather high. It's likely Canberra and the Swans fixture is generating the lion's share, as my experience is the free tickets only surface for non Swans games in Sydney.

The AFL Annual Report mentions a reduction of $3m in the direct allocation to the Suns and Giants. Elsewhere it mentions an identical reduction in the allocation to "New Markets" which is separate from the distribution to clubs. The total spend on these new markets in 2013 was $11.5m. If the Suns and Giants are the major beneficiaries of the expenditure then the $7.5m you mention would drop to near the $2.58m in Gate Receipts the Financial Review mentions in its article on the Giants 2012 Financial Reports.

http://www.afr.com/p/lifestyle/sport/gi ... isOy2PiQ2J

[yob]

GWS average attendance at games in Sydney not including the home derby was 7,914 in its first season and grew 4.6% to 8,281 in its second season. After 5 games this season it is down to 8,244. 2 of those games, however, were rain affected.

To fill the stadium in 20 years they need 6% pa growth in attendance which is triple the population growth rate in Western Sydney.


FYI. Their latest Financial Report is linked below.

http://www.worldofwookie.com/afl/annual ... Report.pdf

GWS reports 18.2 million in AFL distributions and match receipts. The 2013 AFL annual report shows GWS' club distribution was 10.7 million. That leaves us with 7.5 million in match receipts. This strikes me as rather high. It's likely Canberra and the Swans fixture is generating the lion's share, as my experience is the free tickets only surface for non Swans games in Sydney.

The AFL Annual Report mentions a reduction of $3m in the direct allocation to the Suns and Giants. Elsewhere it mentions an identical reduction in the allocation to "New Markets" which is separate from the distribution to clubs. The total spend on these new markets in 2013 was $11.5m. If the Suns and Giants are the major beneficiaries of the expenditure then the $7.5m you mention would drop to near the $2.58m in Gate Receipts the Financial Review mentions in its article on the Giants 2012 Financial Reports.

http://www.afr.com/p/lifestyle/sport/gi ... isOy2PiQ2J

This would suggest that absent any subsidy operating above and around the standard club distributions available to all clubs, the club runs at a loss of around 7 million. That's my on my phone waiting for Gloria jeans to hurry up my coffee fukn analysis.

[Boba Fett]

Sigh...

For the millionth time - the AFL is taking a a very long term view with the Giants. Yes, giveaways aren't sustainable, but this is not a (historically) typical expansion team where there is a existing demand. The Giants with the assistance of the AFL are building demand with the introduction, development and growth of the team. This will take 10-20 years.

Not sure how many times this needs to be explained to people...

For the millionth time in my counter-punch is that crowds were below what they expected in season one, so if they are below expectations in season 1 how can the AFL have any expectations for 20 years time?

Because anyone who knows the slightest bit about professional sport in Sydney would know that once a team starts winning, the crowds follow. Doesn't matter what sport you choose, it's always the same.

So I would expect the Giants' crowds to be poor for the next year or so, but once they start winning more games than they lose and make the finals a couple of times, watch the crowds jump significantly. In fact, here's a prediction - crowds will be at least 15k average in 2016.

[gyfox]

GWS average attendance at games in Sydney not including the home derby was 7,914 in its first season and grew 4.6% to 8,281 in its second season. After 5 games this season it is down to 8,244. 2 of those games, however, were rain affected.

To fill the stadium in 20 years they need 6% pa growth in attendance which is triple the population growth rate in Western Sydney.


FYI. Their latest Financial Report is linked below.

http://www.worldofwookie.com/afl/annual ... Report.pdf

Why would you not include the home derby? Does anyone do that for any other team? Of course not. It's just an excuse to bash GWS for some perceived under-achievement as a start-up AFL team.

Its quite a common technique when analysing data to exclude data that skews the results. What I am looking for is the underlying trend in support for GWS. Season on season there was a 4.6% increase in support in the Sydney market and weather permitting that underlying trend will continue this season but will probably be a bit smaller due to 2 bad weather events in a 7 game data set.

I could include the home derbies if you want but that shows a 12.8% reduction in support season on season and there will be a further reduction this season. The problem with that analysis is that the reduction reflects Swans fans and theatre goers deciding not to attend and as such is not a measure of the Giants core support.

Your suggestion that I am GWS bashing by eliminating the derbies from the analysis is laughable.

[gyfox]

The AFL Annual Report mentions a reduction of $3m in the direct allocation to the Suns and Giants. Elsewhere it mentions an identical reduction in the allocation to "New Markets" which is separate from the distribution to clubs. The total spend on these new markets in 2013 was $11.5m. If the Suns and Giants are the major beneficiaries of the expenditure then the $7.5m you mention would drop to near the $2.58m in Gate Receipts the Financial Review mentions in its article on the Giants 2012 Financial Reports.

[/quote]This would suggest that absent any subsidy operating above and around the standard club distributions available to all clubs, the club runs at a loss of around 7 million. That's my on my phone waiting for Gloria jeans to hurry up my coffee fukn analysis.[/quote]

Bingo. Maybe a bit less than that.

[Boba Fett]

Its quite a common technique when analysing data to exclude data that skews the results.

*facepalm*

[yob]

Well, yes. It's important to have a full data set to start with, but almost always when you're using data to produce information, that process involves performing operations on that data, and it often involves excluding part of the data set.

To put it simply - in the A-League, Melbourne Heart hosts one home derby one season, two the next. If their average non derby crowd drops, but two sizable derbies push their average above the year before, then you're left with two indicators pointing in different directions. Average is up, they're more popular. Base line average is down, maybe not so.

Gyfox is quite sensible to consider base line averages. It's not about drawing concrete conclusions. It's about direction and magnitude. Opinions form over time, and rarely settle.

[Boba Fett]

But the minute you exclude data from a data set you skew the result. Which is the opposite of what gyfox said. Using that logic, you can twist data analysis to essentially become meaningless.

For example, I'm sure Eddie McGuire would love to exclude that Sunday night game Collingwood had a couple of weeks ago when listing the average crowd figure for 2014.

[yob]

But the minute you exclude data from a data set you skew the result. Which is the opposite of what gyfox said. Using that logic, you can twist data analysis to essentially become meaningless. .

Nope.

You need to soften the idea of what the result is. We don't read data, we read information. Information is produced from operations on data. It always involves judging what data is relevant, and what data is not.

A real world example is the census the ABS performs. If you were to see the raw data _before_ it's been operated on - namely for confidence (where you literally multiply a number by its confidence interval and almost *always* make the number smaller) - you'd think every single person in Australia rides their bike, skateboard, dogs to work, 55% of us are feminists and we all speak a second language. It's bullshit. People lie because they want their interests to be better recognised by the government. The truth is though that those numbers are combined with the confidence with which we have those respondents speak the truth, and it *always* involves that number being smashed to a fraction of the raw data.

By all means have a complete data set. But if you're producing information for a human being to digest, and if you're looking to form conclusions, then you do make choices of which data is relevant and which is not. Lest you want to have the real global warming debate, where we use absolutely nothing but raw data. Prepare for some impassioned debate every time a mountain goat pisses in a rainfall gauge.

[Rob]

But the minute you exclude data from a data set you skew the result. Which is the opposite of what gyfox said. Using that logic, you can twist data analysis to essentially become meaningless.

For example, I'm sure Eddie McGuire would love to exclude that Sunday night game Collingwood had a couple of weeks ago when listing the average crowd figure for 2014.

It depends what you're trying to measure. If it's the popularity of Collingwood games on Saturdays, then it would be more than reasonable to not include games played on Sundays.

If you want to measure GWS popularity, it would seem silly to include games against the Swans, because how many Swans fans show up has nothing to do with how popular GWS are.

[Simmo79]

Its quite a common technique when analysing data to exclude data that skews the results.

*facepalm*

Seriously, dude? Have you never studied stats? One of the most basic concepts when crunching numbers to make them talk is to exclude outliers, especially when you know the outlier is answering a question you're not asking.

If you straight want to know what GWS' attendances are, then include the Swans game because it's repeated every year. It's reliable info.

But, if you're asking what GWS' own fanbase size is, exclude Swans fans from the count. Get rid of the derby.

(Same rules apply when asking the same questions for WSW and SFC - we got 40k when they visited, doesn't mean our fanbase got commensurately bigger)

edit: what Rob and Yob said

[yob]

But the minute you exclude data from a data set you skew the result. Which is the opposite of what gyfox said. Using that logic, you can twist data analysis to essentially become meaningless.

For example, I'm sure Eddie McGuire would love to exclude that Sunday night game Collingwood had a couple of weeks ago when listing the average crowd figure for 2014.

It depends what you're trying to measure. If it's the popularity of Collingwood games on Saturdays, then it would be more than reasonable to not include games played on Sundays.

If you want to measure GWS popularity, it would seem silly to include games against the Swans, because how many Swans fans show up has nothing to do with how popular GWS are.

It would also be silly to include Canberra crowds if you want to infer from the data how popular GWS is in Sydney. And vice versa.

[yob]

Just to muddy the waters a bit:

You don't have to entirely exclude an outlier - you can assign a level of confidence to it and include a portion. In the case of derby crowds, you may not exclude *100%* of it - because there will be Sydney FC fans that attend only 1 game per year, the derby. But they are a small fraction. So you would consider a very small proportion of the derby crowd to still be a SFC fan, even though the WSW fans caused the obvious bump. It's not necessarily 0 or 100% in this instance. Maybe only 1% even.* But if you really wanted to sharpen your pencils you'd go to that extent.

Play on.




*getting off track, but how would you guys do it? Maybe subtract the average from the derby crowd to get the derby uplift, then assign a level of confidence based upon a sample survey of attendees asking how many games they have/intend to attend in the season, and what team they support.

[yob]

BTW this is the only time it's ever been cool to know statistics, so fukn make hay while the sun shines boys.

And this thread officially jumps the shark when someone mentions z scores.