THE BOOK--Playing The Percentages In Baseball

Why would you want to crowd source stadium shadows?  That seems like a route to bad data that could actually be determined accurately using geometry and the known motion of the earth and sun throughout the day at any given point in the year.

Let me clarify.  My point is not that you don’t want observations.  My point is that you need to build a geometrical model that accounts for how the earth and sun move and some basic positional information about where the relevant components of the stadium are located, even in a simple way, so you know how the shadows change throughout the day and year.

Mike: that’s a great idea! 

If you, or anyone else, does all the hard work, I’ll use that data. Otherwise, what’s my alternative?  Well, I’d rather use some data than no data.

The key point from my perspective is that someone ELSE is doing the hard work of collecting or providing the data.  To that end, I will encourage any and every person to roll up their sleeves here.  We need worker bees.

I just come in at the end to take the glory, like a well-respected Wall Street CEO.

Tango, a nicely worded email to Populous might be worth a shot.  I know that they have shadow studies for the ballparks they have designed, and have the ability to display the light patterns on the field for any date/time combination. 

I’ve seen such a summary for one of the recent parks (I can’t recall which one), which was promulgated as part of a pitch to potential season-ticket buyers, IIRC.  It seems people like to know how often they are going to be baking in the sun vs. chilling in the shadows…

Interesting as a laboratory of sorts. We have some data now on what happens when batters can’t see the spin on the ball.

I would agree though - not good for the game of baseball.

Greg, dude, you are making me work!

Ok, just sent an email…

I’ve tried looking at this a bit already but have run into some issues with processing power and a lack of database skills on my part.  Right now, it is a 150MB Excel spreadsheet that is bogging down my laptop.  I also haven’t been able to figure out when the retractable roofs are open/shut.

The game_events.xml file at the MLB site time stamps the start of each atbat with the current UTC time.  NOAA has a a freely available spreadsheet that calculates the Azimuth and Elevation of the sun based on location (lat & long), date and time.  Using Google Earth, you can pretty easily figure out the direction from home to the mound for each park except Target field and the Rays dome to normalize.

I’ve uploaded the stadium data to this file:
https://docs.google.com/spreadsheet/ccc?key=0Asg_A8vM31yKdFB6OGkxTjBHdWJZbi1RUllIM2VqX1E&hl=en_US

Here’s the link for the NOAA spreadsheet:
http://www.esrl.noaa.gov/gmd/grad/solcalc/NOAA_Solar_Calculations_day.xls

You guys are my heros!

Chris, I might be able to assist you.  I’ll look at the data when I get home.  Feel free to email me:
tom~tangotiger~net

Greg, dude, you are making me work!

Contacting prominent companies and asking them for information is the sort of work CEO’s do

I had never thought about this as a consideration, but these shadows are also an argument against old-school back-to-back doubleheaders.

You basically never want to be playing baseball at ~5-6 p.m. in a non-dome MLB park.

If a team were to schedule a rare doubleheader, it would make sense to have the first one start at 1 p.m. and the second to start at 7, not only for the financial benefit of being to charge fans for separate tickets, but also as a safety issue with these shadows.

Couldn’t they solve this problem simply by turning on the stadium lights?  I’ve often wondered about this with football games, where you often have shadows covering half the field.  Wouldn’t turning on the lights during certain parts of the daytime improve the product?

I remember the Mariners playing a 3pm playoff game in 2000 or 2001 (or both).  It was all, of course, for the convenience of television (or more to make sure the Yankees were always in prime time).  The idea that you play an entire season starting games at 1 or 7 and then when you get to the playoffs you play under odd October late afternoon shadows tells me a lot about the Bud Selig era.

This data is can be generated using Google Sketchup. Almost all of the current stadiums have models available for download (http://sketchup.google.com/3dwarehouse/), and the program allows for geolocation/shadow generation, with inputs for time of day and date of year.

I love the internet:

The data in “Sun Position - Baseball Stadiums” covers regular season games through 9-8-2011.  It is a compilation of the game_events.xml files on the MLB Gameday server.  There are a number of missing games that are identified in “Missing Games List.txt” The Field descriptions are included in “Field Listing - Sun Position data”

The calculations for the solar position were adapted from the file “NOAA_Solar_Calculations_day.xls” The adaptation adjusted the original file to switch from local time to UTC/Zulu time.

The park data is stored in both “Stadium Data” files.  This table comes from Wikipedia (basic list of stadiums, attendance, etc) with location data (Lat, Long, Angle) calculated using Google Earth.

If you have any questions or comments, you can contact me at cmcary at gmail dot com

Thanks
Chris

http://www.tangotiger.net/chriscary/

Tom - Is all the data for each stadium in the spreadsheet (I can’t download it right now)?

There’s all 30 teams there yes.  This is what the first record shows:

RE: stadium lights

Bright sunlight produces over 100,000 lux (lumens per square meter), maxing out at about 120,000 lux on a clear day at high noon. On an overcast day at noon that is reduced to the 10,000-25,000 range. In contrast, a brightly lit office or supermarket is in the 500-750 lux range. I couldn’t find any figures for stadium lighting, but my experience is that it is just barely noticeable on an overcast day. At sunset natural light (skyglow) produces a maximum of about 400 lux and the stadium lights are producing most of the light. My guess is that good stadium lights are responsible for around 1000 lux, maybe a bit more.

When a baseball moves from sunlight to shadow its brightness decreases by 80%-90%. In bright daylight, with pupils contracted to their smallest size, eyes cannot see the ball well under those conditions. The stadium lights wouldn’t change that by much. However, high intensity lighting focused between the mound and home plate might help.