Worst interstate ever

[NE2]

Probably "roads the GIS data used included".

[DBR96A]

I-97 is five times worse than I-99. At least I-99 can theoretically be extended. The DE 1 tollway only to Dover would be far better for an I-97.

Personally, I think there should be an I-101 from Wilmington, DE down to near Savannah, GA.

[briantroutman]

I wonder what the orange lines *are* supposed to represent, other than perhaps "roads the cartographer likes".

I'm curious as to what the selection criteria were. I was thinking NHS routes or at least NHS freeway routes, but neither of those seem to apply.

Just using PA as an example, the map includes the expected Interstates and...

• US 219 and 22 freeways from Somerset to Altoona–but not any of the other 219 or 22 freeway sections...and none of the US 22/322 freeway
• US 15 from the Maryland line to Harrisburg, including the non-freeway section around Dillsburg, but none of the 15 freeway from Williamsport to the New York line
• PA 61 (!) from Frackville to Reading–not a freeway by any stretch of the imagination

And it oddly omits the section of the NE Ext. between Allentown and the Lehigh Tunnel.

[roadman]

Apart from the illogical number (really should be a 3di off of I-76) , my chief complaint about Interstate 99 is this:  Why does US 220 still need to be co-signed with it?  Now that 99 more or less provides a direct connection between the PA Turnpike and I-80, it seems to me it would make sense to put 220 back on the local roads.  For one thing, this would eliminate the countless Business 220 loops along the way.  And they could do away with the "ALT 220" (original 220 routing years ago) between Milesburg and State College.

[Scott5114]

Apart from the illogical number (really should be a 3di off of I-76) , my chief complaint about Interstate 99 is this:  Why does US 220 still need to be co-signed with it?  Now that 99 more or less provides a direct connection between the PA Turnpike and I-80, it seems to me it would make sense to put 220 back on the local roads.  For one thing, this would eliminate the countless Business 220 loops along the way.  And they could do away with the "ALT 220" (original 220 routing years ago) between Milesburg and State College.

The problem is with an AASHTO rule that says, essentially, "U.S. routes must follow the best routing between two points". This dates back from the 1920s and was meant to avoid someone benefiting from circuitous routings (like auto trail organizations or corrupt politicians). Unfortunately, these days it basically only prevents any US route from being moved off a freeway.

[Grzrd]

I wonder what the orange lines *are* supposed to represent, other than perhaps "roads the cartographer likes".

NLCOG created the graphic around 2008, although TxDOT later used it in presentations to the Texas legislature.  Probably more like "roads the graphic artist thought would fit in well in the background."

[NE2]

The problem is with an AASHTO rule that says, essentially, "U.S. routes must follow the best routing between two points". This dates back from the 1920s

Does it? How was US 55 between Dubuque and Davenport (ca. 1932) allowed?

[briantroutman]

...it seems to me it would make sense to put 220 back on the local roads.

I disagree, because even though the two routes overlap from Bedford to Bellefonte (and eventually, to Williamsport as well), they do diverge there, and US 220 serves a different corridor going to Towanda and Sayre. And southward, 220 goes on to Cumberland, MD and further through WV.

If US routes are still supposed to represent viable paths of travel between points and not be merely old-timey curios (like historic 66), I think they should follow a logical course that people would rationally use. If you were driving from Cumberland to Sayre, would you actually exit the freeway at Bedford and drive through 25 mph town streets in Bedford, Altoona, Tyrone, etc.? I don't think so. And in places like Altoona, which aren't exactly tiny towns, the local roads may be overburdened if town motorists had to share the streets with significant volumes of through traffic.

Now if the bypassing freeway doesn't adequately serve a significant community along a corridor–in this case, let's say if Milesburg was an important destination on 220–I think that would be a valid argument for having 220 break off the freeway and continue up the old alignment in Eagle Valley.

And if two routes simply overlap and truly don't serve different corridors, I think that's perfect justification for decommissioning one of the two or at least truncating it to the point of overlap.

[NE2]

The decision was made in the 1950s to use a different route system for Interstates. Thus you follow the U.S. Route if you want to stick to the surface roads. Interstates added to the system later, such as I-99, screw with this expectation.

[agentsteel53]

The problem is with an AASHTO rule that says, essentially, "U.S. routes must follow the best routing between two points". This dates back from the 1920s and was meant to avoid someone benefiting from circuitous routings (like auto trail organizations or corrupt politicians). Unfortunately, these days it basically only prevents any US route from being moved off a freeway.

explain how US-80 makes a weird jog to go to Phoenix, or US-395 makes an even weirder jog to go halfway to Susanville, CA.

[NE2]

explain how US-80 makes a weird jog to go to Phoenix, or US-395 makes an even weirder jog to go halfway to Susanville, CA.

These are explainable by those being the better road at the time (and by Phoenix being a major destination). But US 55 was a worse alternate to existing US 61.

And what about US 64 from Morganton to Statesville, NC, moved off US 70 in the 1980s?

[Alps]

I'm just seeing this thread for the first time. Ignoring the numbers and focusing purely on the roadway at hand:

In terms of design standards: I-278, NY. From the Gowanus to the BQE to the Bruckner, I can't think of any section (even for 50 feet) that's actually up to full Interstate standards.

In terms of engineering: I-480, CA. Way to put a double decker freeway in an earthquake-prone area.

In terms of traffic: Let's go back to I-278. Honorable mention to I-95 through NYC, but I-278 is notable in that backups routinely occur pretty much everywhere on the route.

Future/planned: I-69 through TN. I-55 is right there to the west. Just route 69 that way and over I-155. Bam. Done.

In terms of didn't need the number: I-2.

In terms of wasting money: I-93 through Boston

[agentsteel53]

In terms of didn't need the number: I-2.

I would put 97 ahead of 2 in this category. 

or, to take "didn't need the number" in various other directions... a lot of the tiny three-digit interstates really don't need to be signed.  both I-375s, Florida I-175... all are just glorified off-ramps that should be treated like I-345: an internal reference number, and signed "TO" something or another in either direction.

[Molandfreak]

But US 55 was a worse alternate to existing US 61.

55, I assume, went where it did for the sole purpose of serving Clinton. There are far better ways they could have planned that, I know, but meh... https://www.aaroads.com/forum/index.php?topic=9800.msg230898#msg230898

[Mapmikey]

Apart from the illogical number (really should be a 3di off of I-76) , my chief complaint about Interstate 99 is this:  Why does US 220 still need to be co-signed with it?  Now that 99 more or less provides a direct connection between the PA Turnpike and I-80, it seems to me it would make sense to put 220 back on the local roads.  For one thing, this would eliminate the countless Business 220 loops along the way.  And they could do away with the "ALT 220" (original 220 routing years ago) between Milesburg and State College.

My thought regarding 220 is that it should be replaced with US 15 above Williamsport, dropped from I-99, replaced with US 221 from Roanoke to Bedford, PA then renumbered as state routes Roanoke to Rockingham wherever it isn't part of I-73.

Back in the day before I-73 and I-99 came along I used to think US 220 should be extended south to Cheraw SC then replace SC 9 to North Myrtle Beach.

Mapmikey

[english si]

I say Interstates 39, 41 & 43.  Does Wisconsin really need to waste so many 2 digit Interstate odd numbers?

I-39 was IL's fault for not allowing I-43 for the US51 corridor (they wanted Madison linked southwards). WI had to do something to have the interstate have a unique section in their state... Oh, and I-43 was due to I-55 or I-57 (which I believe was the number WI and AASHTO/FHWA wanted) being blocked by IL.

And didn't I-41 talks start with extending up from Chicago? And when that was blocked WI consulted on I-41, I-47, I-794 and I-243, with I-41 the clear winner.

I mean, why wouldn't IL force the state to the north to use different 2di numbers so that, one day, it can extend them south in its 2di gluttony fest.

In fact, I'm highly surprised that I-92 for Chicago-Detroit didn't happen, so they could claim another. Or that, more recently, they didn't push for an I-44 extension into IL with the renumbering needed for the St Louis new bridge

---

I-45 is ok, because it serves Houston, the fourth largest city in the U.S. That isn't to say that it shouldn't be extended to Tulsa or further, though.

Part of the problem is that the odd US4x routes kinds of puts the squeeze on the I-4x series (really anything odd between I-35 and I-55).

When the interstate system was created, I-45 was the only N-S route between I-35 and I-55. I-37 was added before numbers were final. I-49 in LA was 1970, 12 years later.

Back in the late 50s, Houston and Dallas-Fort Worth had only just started their population explosions, but given that both are top 10 urban areas, you can't begrudge I-45 too much.

---
Quite clearly some criteria needs to be set:
1) substandard roadway - at grade junctions, etc
2) poor use of a decent number - short 2dis
3) lack of traffic function - low AADT
4) lack of network purpose - why is it on a national network?

I'm going to say: I-A3.

Not divided for most of it. No reason why it's not part of I-A1 that it meets end-on (esp given that both are AK1), and the tiny town at the southern end is really not that important.

[Avalanchez71]

I say H-201.  Too many numbers to fit on a shield.

[Mr_Northside]

Now that 99 more or less provides a direct connection between the PA Turnpike and I-80, it seems to me it would make sense to put 220 back on the local roads.  For one thing, this would eliminate the countless Business 220 loops along the way.  And they could do away with the "ALT 220" (original 220 routing years ago) between Milesburg and State College.

I've thought that for years.
Maybe I've been a little conditioned to expect something like that.  A little further west we have US 19 fairly closely paralleling I-79, which seems just fine.

[NE2]

I say H-201.  Too many numbers to fit on a shield.

3 is too many?

[Billy F 1988]

I say H-201.  Too many numbers to fit on a shield.

3 is too many?

Well, that's a judgement call you'd have to make on your own, but it does appear that way with 4 digits. I'd say just redesignate it as H1A or for those "suffixed-route-sensitive" folks, H4.

[hbelkins]

If we're talking about numbering, I-238, hands down, for not having a parent 2di.

I-97 is the runner-up of the choices offered here.

If we're talking about a glorified set of ramps, Virginia's I-381 counts for certain.

[ap70621]

The first one that came to my mind was I-278.

[Alps]

In terms of didn't need the number: I-2.

I would put 97 ahead of 2 in this category. 

or, to take "didn't need the number" in various other directions... a lot of the tiny three-digit interstates really don't need to be signed.  both I-375s, Florida I-175... all are just glorified off-ramps that should be treated like I-345: an internal reference number, and signed "TO" something or another in either direction.

97 needed some sort of Interstate number to get the funding. 2 was already built. The designation does absolutely nothing.

[empirestate]

The first one that came to my mind was I-278.

Same here, not because I have that much against it myself, but it's a commonly cited candidate for worst Interstate. Surprisingly, though, most of the responses so far have been about worst choice of Interstate designation, not that the roads themselves are so bad.

[briantroutman]

I must say that I'm a little surprised by all of the hatred for I-97. Not just here, but all of the "my renumbering proposal" kinds of threads.

I still remember when I first saw I-97 on a map back around '94. I was quite young...still kind of a proto-roadgeek...and this was well before I ever heard of m.t.r. or the MUTCD or any of the things I'd discover in the following years. And I was surprised to see I-97 shields on the map, but at least it fit the grid–and the fact that it was the shortest 2-digit Interstate I had ever seen was intriguing. And in part, I think those kinds of oddities–not out-and-out violations, but oddities–are the kinds of things that got me interested in roadgeeking in the first place. Even now, I'm completely fine with I-97. Correct me if I'm wrong, but it's not as if a bunch of other Eastern states are begging AASHTO for an Interstate designation, but they can't get one because 97 is already taken.

Contrast that with 99: A lowlife sludge of a politician from Nowheresville handpicks a vanity number for his pork-barrel graft-way... Gets my vote without question. ...if we're talking about numbering, that is.