Playoff Site Selector

freeagent

Moderator
Just saw this linked on another board ... looks like some egg-head at UT Dallas put together a playoff site locator that takes all sorts of neat things into account (one of which is looking for a sixman field).

Here's the link:

http://www.utdallas.edu/~bjc062000/map/football.html#

Also, I read something the other day on a new take on the coin flip, especially when it has to be done over the phone.

One of the two coaches chooses a city/town in Texas and the other coach guesses odd or even on the Zip Code for that city. For example, if I chose Seguin and you guessed Odd, you'd be right, since the Zip Code here is 78155.
 

Blue Bird

Well-known member
I don't know how often sixman fields are updated as far as capacity is concerned. But, Robert Lee, with their 2 year old turf also has the capacity to hold 1500 at least.
 

elenaran

Member
Granger - works fine for me on Firefox. It does look better on larger resolutions, but should work even on smaller. Is your FF up to date?

CoachMiller - thanks, I'll update
 

granger

Founder
apologies. It worked fine when I reloaded it. The team drop-down boxes were not populating originally.

What does the score mean? (I assume you are trying to get close to 1) Any info on distance?

Have you ever seen the NCAA distance calculator? The use it to create regionals to minimize teams that have to travel over 400 miles (which requires them to provide air transportation, rather than ground) for national championships.

It is very easy to use as an API.
https://web1.ncaa.org/TES/exec/miles
 

elenaran

Member
Right, the score is basically a ratio of the combined distance from school 1 and school 2 to the specified stadium to the straight-line distance between the two schools, all multiplied by the ratio of the further school's distance to the stadium to the closer school's distance:

d1 = School 1's distance to stadium, assuming School 1 is further away than School 2
d2 = School 2's distance to stadium, assuming School 1 is further away than School 2
stln = straight-line distance between School 1 & School 2

score = (d1+d2)/stln * d1/d2

So, in an ideal world, the two schools would meet at the exact midpoint along the straight-line between the two, which would mean d1=d2 and d1+d2=stln, resulting in a score of 1.

Unfortunately I'm restricted to using straight-line distance (or technically, great circle distance). I would really prefer to use travel time, which makes a lot more sense, especially in West Texas, where there is no straight-line road between two schools. I used to use Google Maps API to do that, but they've since implemented much stricter usage limits, so it's not really possible to calculate drive-times only the fly for thousands of stadiums and schools anymore. My only idea for a workaround is to pre-calculate travel times for every school/school and school/stadium combination in the state and store them in a database somewhere, but that won't be easy either.
 

sixmanfan2007

Active member
elenaran":iitxe7cm said:
Right, the score is basically a ratio of the combined distance from school 1 and school 2 to the specified stadium to the straight-line distance between the two schools, all multiplied by the ratio of the further school's distance to the stadium to the closer school's distance:

d1 = School 1's distance to stadium, assuming School 1 is further away than School 2
d2 = School 2's distance to stadium, assuming School 1 is further away than School 2
stln = straight-line distance between School 1 & School 2

score = (d1+d2)/stln * d1/d2

So, in an ideal world, the two schools would meet at the exact midpoint along the straight-line between the two, which would mean d1=d2 and d1+d2=stln, resulting in a score of 1.

Unfortunately I'm restricted to using straight-line distance (or technically, great circle distance). I would really prefer to use travel time, which makes a lot more sense, especially in West Texas, where there is no straight-line road between two schools. I used to use Google Maps API to do that, but they've since implemented much stricter usage limits, so it's not really possible to calculate drive-times only the fly for thousands of stadiums and schools anymore. My only idea for a workaround is to pre-calculate travel times for every school/school and school/stadium combination in the state and store them in a database somewhere, but that won't be easy either.

Jayton is playing Follett in Highland Park High School's Hornet Stadium in Amarillo, Texas

Google maps shows distance from Jayton to stadium is 184 miles with a drive time of 2hr50min or 170 min

Follett to stadium is 138 miles with a drive time of 2hr10min or 130 min

Straight line distance is 322 miles. Selector gives a score of 3.6551

Just wonder how it got that score?

formula score = (d1+d2)/stln * d1/d2 gives 1.9339
 

elenaran

Member
sixmanfan2007":2p3tz9cr said:
Jayton is playing Follett in Highland Park High School's Hornet Stadium in Amarillo, Texas

Google maps shows distance from Jayton to stadium is 184 miles with a drive time of 2hr50min or 170 min

Follett to stadium is 138 miles with a drive time of 2hr10min or 130 min

Straight line distance is 322 miles. Selector gives a score of 3.6551

Just wonder how it got that score?

formula score = (d1+d2)/stln * d1/d2 gives 1.9339

Hey somebody's using the selector!

Great question. So in my explanation above there's a bit of handwaving with the formula. In actuality, I raise the first part of the formula to the power of 3, to give the distance exponentially more power over the results than the ratio, like so:
score = (d1+d2)^3/sch1_2dist^3 * d1/d2

(Where d1 is the larger of the two distances between schools & stadium and sch1_2dist is the distance between the schools)

So the travel times I have calculated are:
d1 = Jayton->Highland Park = 200min (3hr 20min)
d2 = Follett->Highland Park = 132min (2hr 12min)
sch1_2dist = Jayton->Follett = 248min (4hr 8min)

score = (200+132)^3/248^3 * 200/132 = ~3.65

The calculated travel times I use (e.g. Jayton->HP=3:20) aren't as accurate as Google Maps, but it's the best I've got.
 

elenaran

Member
I've considered giving the user some sort of slider to let them adjust their preference between Distance and Neutrality (i.e. the exponent mentioned above), but I don't know if it would make things too complicated.
 

granger

Founder
I love posting ridiculous matchups in here and seeing where they are potentially played... This is also good for teams on opposite sides of the state to find a neutral site
 
Top