Results of the attendance prediction contest
DMZ · December 18, 2007 at 11:42 pm · Filed Under Mariners
A long, long time ago, we asked readers to send in their guesses on what the high and low attendance at a game last season would be. The actual results:
Lowest attendance: 16,555 (May 2nd, hosting the White Sox)
Highest attendance: 46,377 (August 5th, hosting the Red Sox)
Going back through our email, I found that the best guess on the high *and* the low was by Perttu Soininen, who guessed 46,512 and 17,203 — only missing the high by 135 and the low by 648.
Congrats, Perttu. Good guessing.
Comments
14 Responses to “Results of the attendance prediction contest”
Leave a Reply
You must be logged in to post a comment.
Is it possible to check the ‘wisdom of crowds’ theory and determine how close USSM readers were as a whole to guessing right on?
On the other hand, Perttu was extraordinarily close, probably closer than any algorithm could predict, so hat’s off there.
Well, the high is going to be a sellout, and while that should be a well-defined number the fuzzy math the M’s use ensures that no two “sellouts” have ever been the exact same number (and no attendence figure seems to have anything to do with actual humans sitting in actual seats, as those empty seats at “sellouts” attest). Nevertheless, knowing that the M’s will have at least one sellout during those Red Sox or Yankees series, and knowing the rough number that Safeco is supposed to hold, you can get pretty close on the high end.
Guessing the low end is pretty darn tough, though. And I wouldn’t have guessed a White Sox game at the start of May, either.
So yeah, nailing both is pretty darn good. A little lucky, but also very good.
Guessing total attendence for the year might be fun to add to the contest for ’08, though. That’s wildly variable, since so much of it depends on how the team does on the field.
Guessing the total for 2008 should be reasonably simple assuming the current offseason trends. 2008 ~= 2007 on paper talent-wise, so 2007 attendence – 5% (less luck/regression to true talent) – 5% (higher ticket prices).
Not at all regretting giving up our full season plan, when 2007 looks to to be high water mark for the foreseeable future.
OK. Thanks. Wow. I never even bothered to check back what I had guessed and how it measured up during the season.
IIRC, the “algorithm” was to look back at the ’06 season and adjust the numbers down some…
Amazing what a difference a change of sox can make . . . 😉
What’s the prize, a 50% off coupon to the site?
Guessing the total for 2008 should be reasonably simple assuming the current offseason trends. 2008 ~= 2007 on paper talent-wise, so 2007 attendence – 5% (less luck/regression to true talent) – 5% (higher ticket prices).
The best predictor of full season attendance is the previous season’s winning percentage, so I expect we’ll draw significantly more fans in 2008 than in 2007.
Plus, I don’t see a lot of evidence of price elasticity in professional sports attendance, so I’d throw out your second -5% as well.
I demand a USSMariner “Vote for Perttu” T-Shirt.
Incidentally, it’s been a while since I’ve seen such an obviously Finnish name outside hockey, strongman competitions, or Nightwish liner notes.
I think the attendance was that low for the white sox game because there was a time change so they had to play during the day rather than night or something like that, and it was a tuesday or wednesday, so I could understand if people don’t go out of there way to make it to safeco field and deal with rush hour traffic after the game.
I won’t be surprised with the higher ticket prices and the mediorce team, if attendance dips that low.
Incidentally, it’s been a while since I’ve seen such an obviously Finnish name outside hockey, strongman competitions, or Nightwish liner notes.
Try reading some OSS source code….
Incidentally, my master’s thesis will be on OSS software markets…
The best predictor of full season attendance is the previous season’s winning percentage, so I expect we’ll draw significantly more fans in 2008 than in 2007
You know, this reminded me of something I looked into a while back; I think I posted something about this on an earlier thread but I’ll throw it in here:
Let’s assume that statement is true. We can then calculate something we might call “Expected Win-Driven Attendance” by dividing one year’s attendance by the previous year’s win total. For the M’s, we see something interesting:
Year....Wins...Attendance.....EWDA
2007.....88.....2,672,409....34,262
2006.....78.....2,480,717....35,952
2005.....69.....2,689,529....42,691
2004.....63.....2,940,731....31,621
2003.....93.....3,268,509....35,145
2002.....93.....3,542,938....30,543
2001....116.....3,507,326....38,542
2000.....91..........
If we ignore 2005 (I’ll come back to that in a minute), you’ll note how consistent the numbers are. It’s basically 33500 plus or minus a couple of thousand for every win the team had the previous year — which also happens to be a pretty good one-game attendance figure. Essentially, for every game they win one year they get another good crowd the next.
But 2005 is the obvious outlier. What happened there? Well, they were coming off the disastrous 63-win ’04 year. You might be tempted to argue that year drove away the casual fans, and all that were left were the die-hards. But look at 2006: there were even fewer die-hards that year, despite 2005 being an improved year in terms of wins. No, it’s pretty clear something else happened in 2005, and that something else was Richie Sexson and Adrian Beltre. The ’04-’05 off-season, as I’m sure you recall, was the year the M’s finally started acting like a big-market team and went shopping for free agents. Now, did they do that because they finally felt financially comfortable? Because Bill Bavasi persuaded ownership to open the purse? Or did they do that because they did the same calculation I just did here? After all, it’s an easy and obvious calculation to do. And if you were the owners sitting around in the autumn of ’04, knowing that you get about 33,500 fans for every win, knowing that you had just 63 wins, you wouldn’t like where the math was taking you. 63 wins times 33,500 fans is just 2 million, almost a full million less than the year before. That’s ugly. In fact, for a business, that’s terrifying. One third fewer customers? So you open your wallets and make a splash, hoping that will be enough to draw in more fans than your win total gives you any right to expect.
And it worked. In fact, using the average EWDA value (excluding the ’05 season) of 34,344 we can calculate how many fans should have attended, based just on the previous year’s win total:
Year.....Attendance.....Projected.....Variance.....%
2007.....2,672,409..... 2,678,841 ......-6,432.....0%
2006.....2,480,717..... 2,369,744 .....110,973.....5%
2005.....2,689,529..... 2,163,679 .....525,850....24%
2004.....2,940,731..... 3,194,003 ....-253,272....-8%
2003.....3,268,509..... 3,194,003 ......74,506.....2%
2002.....3,542,938..... 3,983,918 ....-440,980...-11%
2001.....3,507,326..... 3,125,315 .....382,011....12%
Yeah, maybe that 116 win ’01 should’ve given more of a bump than it did, but take a look at ’05. The M’s got a cool half million more fans than they had any business seducing to the ballpark. At a rough average $24 per ticket, that’s $12M. Sexson and Beltre together cost $14.5M that year (though if you calculate the real cost of Sexson’s backloaded contract and signing bonus, it’s obviously more). Plus whatever those two contributed to the win total in the following years (it’s worth noting the M’s haven’t undershot their expected attendance since). Was it worth it? I think you can make the case that it was: 2M fans is less than 27K per game; factor in those NY and BOS sell-outs, and ’05 would’ve had a lot of Kingdome-in-the-80s nights, with no way for TV to hide the scattering of fans in the stands and HR balls bouncing around in the bleachers for minutes before somebody can get over to grab them. Ugly indeed. And perhaps the beginning of another vicious circle of dropping budgets and spiraling attendance.
Anyway, that was then, and this is now. And now we can make a prediction: barring some equivalent offseason splash (eg, Santana), the 88 wins this year should push the M’s back to right around 3M attendance next year.
An epic post, to be sure.
Don’t forget to add in parking and concessions revenue. If each fan is paying $24 per ticket, they’re probably paying at least another $5 on average in associated costs (the “four hot dogs and colas, team hat and team pennant” cost-of-attendance BS) then those contracts do indeed pay for themselves.