State with the largest average distance between interstate exits

[oscar]

Montana:  average of about 5.0 miles between interchanges (239 interchanges, not double-counting interchanges between Interstates or on the I-15/I-90 duplex), about 1196.2 total Interstate mileage (not double-counting about eight miles on the I-15/I-90 duplex). 

Utah:  about 3.8 miles between interchanges.  Utah has some stretches of I-70 and I-80 with very, very low interchange densities, but its average gets killed by the interchange-dense Ogden-Salt Lake City-Provo area.

There may be other candidates in the mountain states and upper Midwest.  Montana seemed the most likely since it has no major metro areas to drive up the interchange count, and lots of sparsely-populated rural areas to drive down the interchange density.  South Dakota, Idaho, and Wyoming would be worth checking out as well.  (North Dakota has too many metro areas; Nebraska has lots of interchanges per mile in the Omaha and Lincoln areas, and similar for Kansas.) 

I double-checked Vermont which was mentioned above (easy to check the interchange count -- just for this exercise, sequential exit numbering is better than mileage-based), and somewhat to my surprise it has Montana beat.  Who'da thunk?

[vtk]

I think interchanges where two Interstates cross should indeed count double.  A 3-way Interstate interchange (terminus of one route, or one end of a multiplex) should count as one-and-a-half.  A five-leg interchange, assuming all five legs carry Interstates, would count as 2½.

Or, if we're really concerned with exits then freeway-freeway interchanges shouldn't count at all.

[oscar]

I think interchanges where two Interstates cross should indeed count double.  A 3-way Interstate interchange (terminus of one route, or one end of a multiplex) should count as one-and-a-half.  A five-leg interchange, assuming all five legs carry Interstates, would count as 2½.

Or, if we're really concerned with exits then freeway-freeway interchanges shouldn't count at all.

For the states likely to have the lowest interchange densities, these aren't going to make much of a difference.  Vermont has only three Interstate-to-Interstate interchanges out of 53; Montana has five out of about 240; and Utah has nine out of about 200.

[hbelkins]

But why would you not want to allow that 11 miles without an exit to "skew the difference"?

Because that's not between exits within the state.

[bassoon1986]

Ok. Here's my best attempt at what I found for Texas:

Total miles of Interstate: 3,532.39
Total number of interchanges: 2,074

1.703 miles between exits. Which is blown way out of proportion by I-35 and the huge metro areas in Texas...

[Scott5114]

My attempt at Oklahoma: 936 interstate miles / 354 interchanges = 2.64 mi/ic. This is somewhat lower than I was expecting. I expected the I-44 turnpikes to pull the average up, but the 3dis pulled it down considerably.

[english si]

The UK has 2173 motorway miles and 667 interchanges (a rough count), giving us 3.26 mi/ic

Given the population density (661.9/sq mi) is more than all but 4 US states (NJ 1205, RI 1016, MA 852.1, CT 741.4. Fifth is MD at 606.2), the mid-western interchange density does point to a difference in design philosophy over here.

[Brandon]

The UK has 2173 motorway miles and 667 interchanges (a rough count), giving us 3.26 mi/ic

Given the population density (661.9/sq mi) is more than all but 4 US states (NJ 1205, RI 1016, MA 852.1, CT 741.4. Fifth is MD at 606.2), the mid-western interchange density does point to a difference in design philosophy over here.

One of the main differences is urban freeway versus non-urban freeway.  Urban freeways (such as I-94 and I-290 through Chicago) have a very dense interchange to mile ratio.  In fact, some of them are flipped.  The ratio for I-90/94 for a one mile segment between I-290 and Hubbard's Cave is 0.125 miles/interchange.  From what I can see, urban freeways tend to be rare in a lot of other countries, including the UK.  That alone may account for the difference in the ratio numbers.  When I ran the numbers for Illinois, it was 3.54 miles/interchange for the state as a whole.  When I took out the densest interstates (I-90. I-94, and I-290), the ratio jumped to 4.20 miles/interchange.  How different the ratio might be if London had several urban motorways through the central city?

[bassoon1986]

And Arkansas:

If you count the FHWA interstates (30,40,55,430,440,530,540,630) its 647.32 total miles, 242 interchanges, 2.67 miles an interchange.

If you include:
unsigned I-130         -  660.02 miles, 250 interchanges, ratio of 2.64
future I-49 (AR 549) -  686.32 miles, 256 interchanges, ratio of 2.68
future I-555             -  735.22 miles, 276 interchanges, ratio of 2.66

The rest of I-49 one day between Fort Smith and Texarkana would add some miles to this ratio for sure.

[english si]

How different the ratio might be if London had several urban motorways through the central city?

Probably about the same, as if you managed to get the Ringways through, then you'd pick up a few more urban routes (Belfast, Glasgow and Tyneside particularly) that add a lot of junctions per mile, but would have also got rural motorways like the M67, Strensham-Solihull, M31, M64, M100, etc that would have had above average gaps.

London would have been about 1.5 miles between junctions, if not more. The M100 would add a similar distance, with about 6 miles between junctions. It cancels out...