08 August 2012

Just a Graph: Orioles' 2012 Attendance

Around game 5-7 is when attendance hit its peak.  This was June.  Home attendance has been shaky since.  Maybe summer vacations or weather is taking its toll.

Ideas?

EDIT: I have corrected the data below.  Some of the days within a month were flipped.



10 comments:

The Oriole Way said...

Wow, that graph is a really tough way to look at attendance.

The most obvious question is: who did the Orioles play? In June, the club had home series against the Phillies (sell-outs), Pirates, Nats, Angels and Indians. July brought Detroit, Tampa and Oakland. So far in August it's been the Mariners. Other than the Phillies and Nationals, that's a whole lot of teams that aren't big draws.

Jon Shepherd said...

Right...things that probably affect attendance: school, weather, current success, historical success, opposing team fan base, trades, promotions...

What would also work is to batch these into average by week then go against previous years.

Bret said...

As of now they are averaging 4K more per game than they did in 2011. Given they have 6 left at home against Boston, 4 against the Yankees you can be assured that number is going to increase, dramatically increase if they can hang a couple more weeks and play meaningful games through the White Sox homestand into the NY/TB one.

From a business and baseball perspective I see one downside to bringing Dylan Bundy up. He gets rocked, has all of his confidence shot, gives up baseball and decides to play guitar. Here is the upside.

1. I think he is better than anything else they got.
2. They will absolutely sell out if he pitches a home game, add 20K extra attendance for 2-3 games compared to their average and that probably makes up for a year loss of service time and earlier FA.
3. His innings will be around 85 after tonight. They want him at 125. So let him go 7 games at 6 innings apiece. That gets you very close to the end of September when hopefully you have a fully healthy Hammel etc. Or you could use him to give everyone an extra day given Chen's heavy workload and space him out in a 6 man rotation for a couple turns.

If we keep throwing Hunter and Gonzalez out against AL East teams down the stretch I don't think it ends well - he would be a huge enery surge with the fanbase and give teams a different look. And Machado is on fire. I am better than Wilson Betemit, will send you my American Legion fielding statistics if you want to verify. Why not put the best team on the field and really go for it? This chance isn't coming again in 2013.

Tom Watson said...

The first Friday game was opening day with 46,773 fans. I think you have several other data points in the wrong order as well, unless I'm misunderstanding the graph. The spike in weekends 5 and 6 were due to matchups with PHI and then WAS as the first commenter pointed out. The series with DET was a nice rebound after a weak showing for CLE.

Jon Shepherd said...

Thanks Tom...fixed now.

Matt P said...

I can answer that one. It's interleague games.

We got 37,241 on average for the interleague games. The only two teams who we had 40k fans when they came to town were Washington and Philadelphia. The reason why we got significantly fewer for the Pittsburgh series was because that happened during the workweek unlike the Wash/Philly series's which were on the weekend.

I've tried looking at attendance. The problem is that it's time consuming to really build a dataset because the data isn't collected in one spot. Plus there are so few games in a season and so many variables that need to be controlled. Which means you need a large dataset to really do an analysis and it takes a long time to get it.

Still, I've found that there are two major factors that impact attendance for the Orioles; the team that the Os are playing and the day of the week.

It makes sense; fans of opposing teams come out for interleague games because its rare for them to have a chance to see their team play. Also, fans come out on the weekend when they're off from work rather than catch a weekday game.

If you factor in those two variables, we see that attendance has stayed relatively static through the whole season with the exception of games played against non-rivals on weekdays. Since June, there has been about an 8.5k fan increase from the April and May numbers.

The Oriole Way said...

Doing a quick linear regression on the following variables:

Opening Day
Attractive Opponent (NYY, BOS, PHI, WSN)
Weekend (Fri, Sat, Sun)
Summer (Mem Day to Labor Day)
Day Game

Yields a 15% y/y increase in attendance.

Actual year-to-date attendance: 26,073
Expected YTD attendance using 2011 model results: 22,428

The Oriole Way said...

Doing a quick linear regression on the following variables:

Opening Day
Attractive Opponent (NYY, BOS, PHI, WSN)
Weekend (Fri, Sat, Sun)
Summer (Mem Day to Labor Day)
Day Game

Yields a 15% y/y increase in attendance.

Actual year-to-date attendance: 26,073
Expected YTD attendance using 2011 model results: 22,428

Bret said...

I'm not sure why this matters other than the fact that payroll might need to go up a bit in order to keep their own players over the next 3-4 years. They will easily get 400,000 more fans this year than last and that is the floor. This is what I was talking about last week when I said having the payroll go up 5-10 million next year would be a drop in the bucket considering the attendance increase and ratings spike for MASN (also for Nats who they are doing a nice job of Leon Trotskying).

Anyone expecting 35-40K fans on a weeknight in Baltimore is delusional pennant race or not. Much of their fan base is DC suburbs and people have to work. Yeah, in 1997 that was a nice thing but in 1997 the economy was booming, Camden Yards was still relatively new and the O's were the game in town baseball-wise. It is clear the fans are coming back compared to the past couple years in a big way and that is only going to accelerate down the stretch if they stay in it during the first September home stand.

Vlad Guerrero doesn't increase attendance. Winning does. And it has.

Matt P said...

The data seem to indicate that if Baltimore had a winner we could expect an increase of about 10k fans during the weekday games and the non-interleague and NYY/BOS games.

When I state a winner, I don't mean that they're winning for a few months. That has a miniscule impact on attendance. I mean a team that has actually shown it could get to the playoffs.

If this team doesn't fade, then it will be interesting to see the impact.

As for how it impacts the team, well, most fans come out for the interleague games. Next year, every team will play each other. The obvious question is whether the opposing fans will still come out each year to watch their team play at Camden Yards because they only have one shot that year or whether teams will decide that they can always do it next year and not show up.

If it's the former, we can expect the Orioles attendance to jump considerably. If its the latter, it will hurt the Orioles attendance.