Idea

[Bo]

What do all of you think about having a redistricting competition here, like at Swing State Project?

[Lunar]

What do all of you think about having a redistricting competition here, like at Swing State Project?

Would you actually do it or are you just the ideas guy?

[Bo]

What do all of you think about having a redistricting competition here, like at Swing State Project?

Would you actually do it or are you just the ideas guy?

I suppose I could be the judge in the competition.

[muon2]

Any competition needs rules, not just a judge. I don't know what SSP has set up, but the Ohio Redistricting Competition last year was fairly detailed. There needs to be a common data set to use for the districts so that maps can be compared.

Even more important is a scoring mechanism that the competitors can use to tune their entries. The scoring is important because it reflects the goals of the map makers. Measures like compactness and partisan balance can work against each other so that the weights as well as the criteria matter.

[RosettaStoned]

Sounds like a good idea.

[The Mikado]

This would be after the Census comes out, right?

[Bo]

This would be after the Census comes out, right?

It could either be before the Census comes out or after. Dave's Redistricting Application already has 2008 census numbers for almost all the states, and those numbers are going to be pretty close to the 2010 numbers, especially consdiering that the fianncialc risis and Great Recession slowed down migration in 2008 and 2009 (and possibly 2010 as well).

[Verily]

Not true. The Census estimates do a very poor job of estimating population changes within counties, both in terms of differential growth rates (one part of a county might be growing very fast while another part is not growing at all or even in decline, yet Dave's Redistricting would not reflect this) and in terms of race and ethnicity (a racial group may be growing rapidly in one part of a county while continuing to have no presence in other parts, yet Dave's Redistricting must assume equal growth county-wide--metro Atlanta is the most notable place where this is a problem).

[Bo]

Not true. The Census estimates do a very poor job of estimating population changes within counties, both in terms of differential growth rates (one part of a county might be growing very fast while another part is not growing at all or even in decline, yet Dave's Redistricting would not reflect this) and in terms of race and ethnicity (a racial group may be growing rapidly in one part of a county while continuing to have no presence in other parts, yet Dave's Redistricting must assume equal growth county-wide--metro Atlanta is the most notable place where this is a problem).

Actually, I used Dave's Redistricting App a lot and the growth there in different precincts within the same county are different. I don't see why population estimates must show that each precinct within a county must grow at the same pace, because Dave's Redistricting App. doesn't show this.

[Verily]

Not true. The Census estimates do a very poor job of estimating population changes within counties, both in terms of differential growth rates (one part of a county might be growing very fast while another part is not growing at all or even in decline, yet Dave's Redistricting would not reflect this) and in terms of race and ethnicity (a racial group may be growing rapidly in one part of a county while continuing to have no presence in other parts, yet Dave's Redistricting must assume equal growth county-wide--metro Atlanta is the most notable place where this is a problem).

Actually, I used Dave's Redistricting App a lot and the growth there in different precincts within the same county are different. I don't see why population estimates must show that each precinct within a county must grow at the same pace, because Dave's Redistricting App. doesn't show this.

The Census Bureau only calculates estimated growth rates based on municipalities (and only based on counties in unincorporated areas, which make up a large chunk of the country), so they don't have the precision to vary population growth across precincts. Similarly, they only calculate estimated demographic changes based on counties.

[Smid]

The accuracy of the basic figures is probably unimportant so long as everyone uses the same information. If it's just a game, it doesn't really matter whether or not Dave's App gives accurate figures or not, that only becomes important if the maps are to be used outside of the Forum - such as lodging as submissions when the redistricting actually happens. Heck, if it's just a game, you could use 1932 figures and it wouldn't really matter! I think so long as everyone's got the same numbers and everyone's judged according to the same set of rules (see Muon's post), it doesn't matter whether or not the figures we're using are absolutely 100% correct.