Asses in seats

[classof60]

Might have been the Wyoming game at least 15 years ago...I think the power went out also...my wife and I were lucky to find a B/B

[InnervisionsUMASS]

Stats are from 2008 to present, so that isn't it.

[Old Cage]

The chart says 15 inches on the ground. It doesn't suggest that there was any snow on game day.

[InnervisionsUMASS]

^

That would make MUCH more sense. We've all see first hand that weather during the games is a factor. Hell, one year people almost couldn't drive to a game due to the statewide travel ban.

[McKinney]

Since you have the numbers do you have or can graph out season win totals v attendance?

Sure, though I think these graphs (this is most of what I put together for @CurryHicksSage) tell a more accurate story about the relationship between attendance and record.



Here's attendance for conference and non conference games during the weekend and the weekdays since the 1994 season (opening of Mullins Center). The lines are a 7 game moving average. I just chose the period of 7 games somewhat arbitrarily, but it looked to me to be a good happy medium between being slow enough to block out statistical noise and fast enough to show the effects of a winning/losing streak. Note: the start each series is delayed from the previous series, but the moving averages are accurate (I've checked manually). It's just a bug with the graphing software I'm using.

Some things to take away from this graph:

• There was a time in the late '90s and early '00s when our weekend non-conference games averaged higher attendance than our weekend conference games. What does this say about interest in the A-10 at that time versus our OOC schedule?

• There was a massive interest increase in all weekend games and conference weekday games, but a relatively flat weekday non conference during the Chaz era. What does this say about the time/effort required to attend a weekday game versus a weekend game?

And here's the model output graphs. These show a linear regression of attendance versus a 30 game rolling record. 30 games was chosen since teams can schedule 29 regular season games, or 27 regular season games and up to four games in one multi-team tournament for a maximum of 31 games. 30 seemed like a nice round number to represent one entire season, and because it's a rolling record it shows offseason momentum.



As I said I came up with these a few weeks ago, so I haven't updated it since the Minnesota game I believe (but with ~600 games in my dataset it's unlikely to have a large impact).

[McKinney]

Good stuff McKinney. Which game was the sellout with 15 inches of snow?

Us old timers can tell stories of trekking through dozens of blizzards to wait in line for two hours to make sure we got into the Cage for games .
Actually, there were a handful of adventures.

That was vs. VCU on February 21st, 2014.

[McKinney]

^

That would make MUCH more sense. We've all see first hand that weather during the games is a factor. Hell, one year people almost couldn't drive to a game due to the statewide travel ban.

I do realize that the weather model isn't the best since our alumni base has one of the longest gameday commutes for D1 FBS schools (from an analysis I did for football). This is probably only really accurate for locals (Amherst/NoHo, but maybe even the Greater Springfield area) and in the case where a storm impacts statewide. I encourage suggestions though on how I could account for difficult travel. Maybe take the average of the weather data from Amherst and the 5 biggest cities in New England? Really not sure how to approach that yet.

[InnervisionsUMASS]

^
Oh yeah, there was def snow on the ground that day, but it was a pretty sunny/clear day in the sky if I recall.

[umass2001]

Since you have the numbers do you have or can graph out season win totals v attendance?

Sure, though I think these graphs (this is most of what I put together for @CurryHicksSage) tell a more accurate story about the relationship between attendance and record.



Here's attendance for conference and non conference games during the weekend and the weekdays since the 1994 season (opening of Mullins Center). The lines are a 7 game moving average. I just chose the period of 7 games somewhat arbitrarily, but it looked to me to be a good happy medium between being slow enough to block out statistical noise and fast enough to show the effects of a winning/losing streak. Note: the start each series is delayed from the previous series, but the moving averages are accurate (I've checked manually). It's just a bug with the graphing software I'm using.

Some things to take away from this graph:

• There was a time in the late '90s and early '00s when our weekend non-conference games averaged higher attendance than our weekend conference games. What does this say about interest in the A-10 at that time versus our OOC schedule?

• There was a massive interest increase in all weekend games and conference weekday games, but a relatively flat weekday non conference during the Chaz era. What does this say about the time/effort required to attend a weekday game versus a weekend game?

And here's the model output graphs. These show a linear regression of attendance versus a 30 game rolling record. 30 games was chosen since teams can schedule 29 regular season games, or 27 regular season games and up to four games in one multi-team tournament for a maximum of 31 games. 30 seemed like a nice round number to represent one entire season, and because it's a rolling record it shows offseason momentum.



As I said I came up with these a few weeks ago, so I haven't updated it since the Minnesota game I believe (but with ~600 games in my dataset it's unlikely to have a large impact).

So it seems the true measure of attendance is simply record. In an almost linear fashion. Surprise surprise. lo

[PreecherJenkins]

To summarize:

There is a strong correlation between winning and attendance.

however, winning will only bring attendance up to a certain level during the week compared to the weekends. So a marketer needs to best spend their money on max crowd size on the weekends and let the weekdays be what they may.

A better chart of data that none of us will ever have is the student attendance over the same period. Has student attendance dropped at a greater clip than the other people who attend games.

Flat out 2004, is a PATHETIC number, and shows the season ticket base to be at probably 1500 (which includes coach and employee comps). It is funny because the ticket rep told me everyone but me renewed, they were full of it.

[McKinney]

FROM http://www.umasshoops.com/newboard/view ... 25#p467825

The only times we packed the Centrum and Garden were the same times we were packing the Mullins. Attendance has more to do with winning than anything else. I agree with those who say maybe we should play one marquee opponent in those venues each year to boost statewide interest, but we need to get good first.

I think that's correct. Also, while playing marquee opponents is good for building the brand/interest, don't expect it to have as major of impact on attendance as winning (or even stronger, our RPI/AP rank). Here's attendance vs opponent RPI (or AP Poll rank where available) since 1993.



At most we're talking about a ~18% correlation, and that's usually a best case scenario: playing big time opponents (typically ranked BE+P5 teams) on the weekend. An attractive opponent will put asses in seats, but not at the rate of an attractive home team (which I think is pretty common sense).

[natwam2547]

I know it will probably never happen, but I'd like to see 1 game in Springfield and 1 at the Cage as part of a throwback night every year.

[InnervisionsUMASS]

It is funny because the ticket rep told me everyone but me renewed, they were full of it.

What was the person supposed to tell you? "Oh yeah, everyone is jumping ship. What a shit show this year is going to be!"

[69MG]

I know it will probably never happen, but I'd like to see 1 game in Springfield and 1 at the Cage as part of a throwback night every year.

I also like that idea. However, I have been to all of the recent cage games and those stands are freakin' uncomfortable. But I'll put up with it for the atmosphere. Regarding Springfield, as mentioned in the McCall thread, 1 game while the students are away would be OK. You would get most of the same Western Mass fans who go to home games and some locals. Try for a "name" opponent to draw the casual fan. I still think back to the atmosphere of the BYU game. The Mass Mutual Center was rocking.

[McKinney]

Also regarding playing at "home away from home" over the winter session: I think UCONN plays more games at the XL Center over break so it's not unheard of.