Which state has the BEST overall highway/freeway connectivity?

[hotdogPi]

Dirt Roads' post has me questioning my own methodology. For this exercise, if the quickest way (not shortest way) includes all freeways, i.e. Colorado Springs to Grand Junction, would that be considered "connected"? Even Pueblo to Grand Junction, right now according to Google Maps, taking I-25->CO470->I-70 is one minute quicker than taking the more direct US50, and it's the beginning of rush hour.

I think it is fair to say that if the all freeway (or all freeway plus "highways") path is the quickest (or even close timewise), that should qualify as "connected" (even if you have to backtrack a bit).  But perhaps the OP should weigh in with an opinion.

No, it shouldn't. In my Massachusetts example, Worcester to Brockton's fastest path is all freeway, but a hypothetical straight line would be much faster. It's 61 miles driving and 41 miles as the crow flies.

[webny99]

Dirt Roads' post has me questioning my own methodology. For this exercise, if the quickest way (not shortest way) includes all freeways, i.e. Colorado Springs to Grand Junction, would that be considered "connected"? Even Pueblo to Grand Junction, right now according to Google Maps, taking I-25->CO470->I-70 is one minute quicker than taking the more direct US50, and it's the beginning of rush hour.

I think it is fair to say that if the all freeway (or all freeway plus "highways") path is the quickest (or even close timewise), that should qualify as "connected" (even if you have to backtrack a bit).  But perhaps the OP should weigh in with an opinion.

No, it shouldn't. In my Massachusetts example, Worcester to Brockton's fastest path is all freeway, but a hypothetical straight line would be much faster. It's 61 miles driving and 41 miles as the crow flies.

I tend to agree with DirtRoads here. I would say if that if a freeway (or non-stop divided highway) connection is the fastest route (or within a few minutes of being the fastest route, or within 5% for a longer trip), that would count as connected.

There's often shorter routes available on two-lane roads or four-lane arterials, but these are often along pre-interstate routes/corridors that would be difficult to upgrade and/or adversely affected by geography/terrain. As such, I think in most cases it's unreasonable to expect the shortest route to also be all freeway if there is already a reasonable freeway alternate. To use your example, I don't think Worcester<>Brockton not having a direct freeway connection between them means the connectivity is poor. It's just a matter of the freeways being located such that you have to backtrack a bit to use them, but that is inevitable in many cases. A true lack of connectivity would be where you are forced to use two-lane roads or arterials because the city has no freeways at all, or take a significantly longer route to stay on the freeway.

[TheStranger]

So I did this exercise with California...

Where things get interesting:  Route 152 currently has numerous stoplights as an arterial in Los Banos (though a bypass has been proposed for decades).

If that counts as highway, that expands the state's 4-lane highway connectivity massively.

BUT if that segment is viewed as a local boulevard only and not a highway...

1. Los Angeles at around 4 million

2. San Diego at 1.38 million
1 <> 2 via I-5

3. San Jose at 1 million
1 <> 2 via US 101 (directly), I-5/CA 152/US 101 (shorter)
1 <> 3 via either I-5/US 101 (historically all US 101) or I-5/CA 152/US 101

4. San Francisco at 815K
1 <> 4 via US 101 (directly), I-5/I-580/I-80 or I-5/CA 152/US 101 (shorter)
2 <> 4 via I-5/US 101 or I-5/I-580/I-80 or I-5/CA 152/US 101
3 <> 4 via US 101 or I-280 directly

5. Fresno at 544K
1 <> 5 via I-5/CA 99 (historically all US 99) directly
2 <> 5 via I-5/CA 99
3 <> 5 via US 101/CA 152/CA 99
4 <> 5 via I-80/I-580/I-205/I-5 north/CA 120/CA 99 or US 101/CA 152/CA 99

If 152 is considered a highway through Los Banos, then I've so far made it to the top 20 cities in California in population where each city pairing is connected by at least a four-lane highway; but if we are excluding any segment which can be seen as an arterial (with multiple stoplights or stop signs) then the pairing of San Jose and Fresno (3rd largest and 5th largest) qualifies as the largest two-city pairing in this state without a direct no-stoplight route!

[sprjus4]

Dirt Roads' post has me questioning my own methodology. For this exercise, if the quickest way (not shortest way) includes all freeways, i.e. Colorado Springs to Grand Junction, would that be considered "connected"? Even Pueblo to Grand Junction, right now according to Google Maps, taking I-25->CO470->I-70 is one minute quicker than taking the more direct US50, and it's the beginning of rush hour.

I think it is fair to say that if the all freeway (or all freeway plus "highways") path is the quickest (or even close timewise), that should qualify as "connected" (even if you have to backtrack a bit).  But perhaps the OP should weigh in with an opinion.

No, it shouldn't. In my Massachusetts example, Worcester to Brockton's fastest path is all freeway, but a hypothetical straight line would be much faster. It's 61 miles driving and 41 miles as the crow flies.

How is it not connected? The fastest route is I-90 to I-95 to I-93 to SH-24. It's all freeway / interstate.

If you disqualified every city pair that didn't have direct connections, but instead had to use two interstate highways and transfer at some junction point, you'd be eliminating a lot of connectivity.

[Roadgeekteen]

Dirt Roads' post has me questioning my own methodology. For this exercise, if the quickest way (not shortest way) includes all freeways, i.e. Colorado Springs to Grand Junction, would that be considered "connected"? Even Pueblo to Grand Junction, right now according to Google Maps, taking I-25->CO470->I-70 is one minute quicker than taking the more direct US50, and it's the beginning of rush hour.

I think it is fair to say that if the all freeway (or all freeway plus "highways") path is the quickest (or even close timewise), that should qualify as "connected" (even if you have to backtrack a bit).  But perhaps the OP should weigh in with an opinion.

No, it shouldn't. In my Massachusetts example, Worcester to Brockton's fastest path is all freeway, but a hypothetical straight line would be much faster. It's 61 miles driving and 41 miles as the crow flies.

How is it not connected? The fastest route is I-90 to I-95 to I-93 to SH-24. It's all freeway / interstate.

If you disqualified every city pair that didn't have direct connections, but instead had to use two interstate highways and transfer at some junction point, you'd be eliminating a lot of connectivity.

I think the point is if an as the crow flies interstate is build directly connecting the 2 it would be shorter.

[thspfc]

Dirt Roads' post has me questioning my own methodology. For this exercise, if the quickest way (not shortest way) includes all freeways, i.e. Colorado Springs to Grand Junction, would that be considered "connected"? Even Pueblo to Grand Junction, right now according to Google Maps, taking I-25->CO470->I-70 is one minute quicker than taking the more direct US50, and it's the beginning of rush hour.

I think it is fair to say that if the all freeway (or all freeway plus "highways") path is the quickest (or even close timewise), that should qualify as "connected" (even if you have to backtrack a bit).  But perhaps the OP should weigh in with an opinion.

No, it shouldn't. In my Massachusetts example, Worcester to Brockton's fastest path is all freeway, but a hypothetical straight line would be much faster. It's 61 miles driving and 41 miles as the crow flies.

I tend to agree with DirtRoads here. I would say if that if a freeway (or non-stop divided highway) connection is the fastest route (or within a few minutes of being the fastest route, or within 5% for a longer trip), that would count as connected.

There's often shorter routes available on two-lane roads or four-lane arterials, but these are often along pre-interstate routes/corridors that would be difficult to upgrade and/or adversely affected by geography/terrain. As such, I think in most cases it's unreasonable to expect the shortest route to also be all freeway if there is already a reasonable freeway alternate. To use your example, I don't think Worcester<>Brockton not having a direct freeway connection between them means the connectivity is poor. It's just a matter of the freeways being located such that you have to backtrack a bit to use them, but that is inevitable in many cases. A true lack of connectivity would be where you are forced to use two-lane roads or arterials because the city has no freeways at all, or take a significantly longer route to stay on the freeway.

It has to be a case-by-case basis. What if the all-freeway route is only faster because the direct route is clogged with traffic because it's underbuilt? In the case of Colorado, I doubt that US-50 is ever backed up and it certainly does not need to be a freeway, so having to drive CO Springs to Grand Junction via Denver in order to stay on freeways should not be counted against the state's highway network.

[JayhawkCO]

Well on that note while I'm coming up with edge cases, how about Greeley, CO? Is it "served" by I-25 when its suburbs are close to it and the highway goes through Weld County which is Greeley's MSA? Or does the fact that US34 isn't limited access from I-25 mean it's not on the highway network?

[Dirt Roads]

Another twist I wasn't thinking about, coming from the PPP I-495 HOT Lanes thread:

<snipped> I-95 north of Baltimore is officially an MdTA toll facility...the entire distance from the Baltimore city limit to the Delaware line.  It gives the illusion of being a freeway but in reality is not...<more snipped>

I didn't consider toll roads as not qualifying as "freeways" (even though they aren't "free").  But I do consider situations where ticket tolls cause such bad traffic issues that the connection between a toll road and a "highway" no longer functions in the sense of connectivity.  But I suspect that the majority of the folks here at AARoads don't consider toll roads as "freeways".

In that case, much of the southeastern part of West Virginia becomes "disconnected" from the rest of the state.  While I'm at it, let me go ahead and disqualify Corridor G (US-119) over the Southridge development issue.

• St. Albans #12
• Vienna #13
• Bluefield #14
• Princeton #23
• Keyser #27
• Grafton #28
• Lewisburg #33
• Kingwood #37
• Williamson #41
• Philippi #42
• Madison #44
• Moorefield #48 (located on Corridor H, but still not yet connected to rest of the world)
• Wellsburg #50

Beckley remains connected to the rest of the state via Corridor L (US-19).  Princeton and Bluefield remain connected to the Eastern Panhandle via I-77 and I-81, but loses connectivity elsewhere.  Similarly, Lewisburg remains connected to the Eastern Panhandle via I-64 and I-81.  But other than Beckley and Oak Hill, the rest of the southern part of the state got just wiped off this map.  Perhaps that is what Rothman was wondering about?

Makes me wonder about West Virginia.

[webny99]

I would say if that if a freeway (or non-stop divided highway) connection is the fastest route (or within a few minutes of being the fastest route, or within 5% for a longer trip), that would count as connected.

There's often shorter routes available on two-lane roads or four-lane arterials, but these are often along pre-interstate routes/corridors that would be difficult to upgrade and/or adversely affected by geography/terrain. As such, I think in most cases it's unreasonable to expect the shortest route to also be all freeway if there is already a reasonable freeway alternate. To use your example, I don't think Worcester<>Brockton not having a direct freeway connection between them means the connectivity is poor. It's just a matter of the freeways being located such that you have to backtrack a bit to use them, but that is inevitable in many cases. A true lack of connectivity would be where you are forced to use two-lane roads or arterials because the city has no freeways at all, or take a significantly longer route to stay on the freeway.

It has to be a case-by-case basis. What if the all-freeway route is only faster because the direct route is clogged with traffic because it's underbuilt? In the case of Colorado, I doubt that US-50 is ever backed up and it certainly does not need to be a freeway, so having to drive CO Springs to Grand Junction via Denver in order to stay on freeways should not be counted against the state's highway network.

I agree to a certain extent, but the first question is whether there's actually a single corridor that goes directly between those two points to begin with. For Worcester to Brockton, there definitely isn't. Using click and drag to try to find the route with the shortest possible distance involves a route with dozens of turns and at least five different numbered routes, and still only reduces the mileage from to 61 miles to 52 miles, while almost doubling up the freeway route in travel time. Trying to upgrade the connectivity strikes me as a total lost cause considering the multitude of different roads involved, the limited mileage savings, and the massive scale of improvements that would be needed to get the trip time into the same ballpark as the existing freeway route.

To me that doesn't mean connectivity is poor, it's just a reminder that MA - like much of New England - was densely populated and had an established road network long before the interstate system, so they're working with a much more complex framework than other parts of the country.



Interestingly, Pueblo to Grand Junction does have a more-or-less direct non-freeway routing (US 50) which is currently identical time-wise to I-25>I-70. I would consider that connected; since there is a freeway option that won't take much extra time, traffic is not reliant on US 50 to get between those cities. Also, US 50 is clearly impacted by the terrain, which is part of the reason it takes longer and would be a major factor in any potential upgrades.

For an example to your point where I do think there is a debate to be had, how about Charlotte to Raleigh, NC? The fastest route is I-85 to I-40, but there is a more direct route using NC 49 and US 64 which saves about 20 miles, but adds 20 minutes travel time because it's mostly non-freeway. That to me a case where it's not immediately clear whether that's acceptable connectivity, and there's a lot more room for debate about whether the extra distance is acceptable or if NC should be addressing that corridor.

[sprjus4]

For an example to your point where I do think there is a debate to be had, how about Charlotte to Raleigh, NC? The fastest route is I-85 to I-40, but there is a more direct route using NC 49 and US 64 which saves about 20 miles, but adds 20 minutes travel time because it's mostly non-freeway. That to me a case where it's not immediately clear whether that's acceptable connectivity, and there's a lot more room for debate about whether the extra distance is acceptable or if NC should be addressing that corridor.

This is certainly an interesting case... I've always viewed I-85/I-40 as the Raleigh to Charlotte corridor, and NCDOT has improved this corridor significantly over the last 30 years, with probably around 80-90 miles of 8 lane widening, a 70 mph 6 lane bypass around Greensboro, and ongoing 6 lane widening on I-40 which will complete a minimum 6 lane corridor between the two cities.

Obviously, these improvements are geared at improving I-85, but if you look at its role in connectivity within not only North Carolina, but further south, I-85 provides a major link connecting Raleigh (via I-40), Durham, Greensboro, Winston-Salem (via I-40 and I-285), High Point, Charlotte, Spartanburg / Greenville, SC, and Atlanta. Raleigh would certainly be part of this, with that connection via I-40 which is busy enough to receive ongoing 6 lane widening. Some of this I-40 traffic may be heading west on I-40 past Greensboro, but a lot is likely bound to I-85 South.

It's not the most direct corridor, but probably the most common routing. I view the slightly shorter, but arterial alignment as an alternative. Any improvements on that routing should consist of 4 lane widening, not necessarily limited access. I just don't see the traffic volumes to warrant full freeway, and it's not like Charlotte to Raleigh does not have a freeway to begin with.

[sprjus4]

I didn't consider toll roads as not qualifying as "freeways" (even though they aren't "free").  But I do consider situations where ticket tolls cause such bad traffic issues that the connection between a toll road and a "highway" no longer functions in the sense of connectivity.  But I suspect that the majority of the folks here at AARoads don't consider toll roads as "freeways".

The term "freeway"  does not necessarily refer to charging tolls or not.

The FHWA defines a "freeway"  as the following:

The Highway Capacity Manual (HCM) (7) defines a freeway as a divided highway with full control of access and two or more lanes for the exclusive use of traffic in each direction. Freeways provide uninterrupted flow (Note: "Uninterrupted"  is used to describe the type of facility, not the quality of the traffic flow at any given time. A freeway experiencing extreme congestion, for example, is still an uninterrupted-flow facility because the causes of congestion are internal.) Opposing directions of flow are continuously separated by a raised barrier, an at-grade median, or a continuous raised median (Figure 1-4). Operating conditions on a freeway primarily result from interactions among vehicles and drivers in the traffic stream and among vehicles, drivers, and the geometric characteristics of the freeway.

Additionally, there is an excerpt about toll roads:

A tollway or toll road is similar to a freeway, except that tolls are collected at designated points along the facility, either electronically, manually, or some combination. Although the collection of tolls may involve interruptions of traffic flow (Figure 1-5), these facilities should generally be treated as "freeways", particularly with respect to strategies and technologies for management and operations. Special attention should be given to the unique characteristics, lane management opportunities, and constraints associated with toll collection facilities. Accordingly, the term "freeway" as used in this Handbook refers to any limited access facility, including the interstate system, expressways, toll roads, and connecting bridges and tunnels.

https://ops.fhwa.dot.gov/freewaymgmt/publications/frwy_mgmt_handbook/chapter1_02.htm

[Dirt Roads]

For an example to your point where I do think there is a debate to be had, how about Charlotte to Raleigh, NC? The fastest route is I-85 to I-40, but there is a more direct route using NC 49 and US 64 which saves about 20 miles, but adds 20 minutes travel time because it's mostly non-freeway. That to me a case where it's not immediately clear whether that's acceptable connectivity, and there's a lot more room for debate about whether the extra distance is acceptable or if NC should be addressing that corridor.

This is certainly an interesting case... I've always viewed I-85/I-40 as the Raleigh to Charlotte corridor, and NCDOT has improved this corridor significantly over the last 30 years, with probably around 80-90 miles of 8 lane widening, a 70 mph 6 lane bypass around Greensboro, and ongoing 6 lane widening on I-40 which will complete a minimum 6 lane corridor between the two cities

It's not the most direct corridor, but probably the most common routing. I view the slightly shorter, but arterial alignment as an alternative. Any improvements on that routing should consist of 4 lane widening, not necessarily limited access. I just don't see the traffic volumes to warrant full freeway, and it's not like Charlotte to Raleigh does not have a freeway to begin with.

Don't forget that NCDOT has been incrementally improving the NC-49//US-64 Corridor between Charlotte and Raleigh for decades, greatly improving the trip time by at least one hour over the last 25 years alone.  I'm particularly fond of the Farmer Bypass (as if Farmer ever really needed a bypass), which was completed in 1959 and has a distinct feel of a "Super Two" with a few interchanges and even a local road underpass, but doesn't quite make the cut. 

[sprjus4]

^ That is true, and the US-64 Bypass around Asheboro that was completed recently also helps to improve this routing.

[hotdogPi]

So if there's a not quite direct freeway route that's the fastest, having a direct road between the two cities that's a surface road somehow reduces connectivity?

[hbelkins]

Another twist I wasn't thinking about, coming from the PPP I-495 HOT Lanes thread:

<snipped> I-95 north of Baltimore is officially an MdTA toll facility...the entire distance from the Baltimore city limit to the Delaware line.  It gives the illusion of being a freeway but in reality is not...<more snipped>

I didn't consider toll roads as not qualifying as "freeways" (even though they aren't "free").  But I do consider situations where ticket tolls cause such bad traffic issues that the connection between a toll road and a "highway" no longer functions in the sense of connectivity.  But I suspect that the majority of the folks here at AARoads don't consider toll roads as "freeways".

In that case, much of the southeastern part of West Virginia becomes "disconnected" from the rest of the state.  While I'm at it, let me go ahead and disqualify Corridor G (US-119) over the Southridge development issue.

• St. Albans #12
• Vienna #13
• Bluefield #14
• Princeton #23
• Keyser #27
• Grafton #28
• Lewisburg #33
• Kingwood #37
• Williamson #41
• Philippi #42
• Madison #44
• Moorefield #48 (located on Corridor H, but still not yet connected to rest of the world)
• Wellsburg #50

Beckley remains connected to the rest of the state via Corridor L (US-19).  Princeton and Bluefield remain connected to the Eastern Panhandle via I-77 and I-81, but loses connectivity elsewhere.  Similarly, Lewisburg remains connected to the Eastern Panhandle via I-64 and I-81.  But other than Beckley and Oak Hill, the rest of the southern part of the state got just wiped off this map.  Perhaps that is what Rothman was wondering about?

Makes me wonder about West Virginia.

If you're gong to disqualify Corridor G based on Southridge, you have to disqualify Corridor L based on Summersville.

[Dirt Roads]

If you're gong to disqualify Corridor G based on Southridge, you have to disqualify Corridor L based on Summersville.

Do you mean because of the shopping center traffic or because of the speed trap? 

I'm not trying to defend West Virginia here, but it seems that a "no shopping centers" rule is going to disqualify a lot of otherwise perfectly good "highways" (in both of these cases, truly ultra-high quality partially-limited access highways).  The main issue that I have with Southridge is that the so-called Southridge Boulevard is a glorified entrance to the parking lot of a large big box complex.  There's also several other shopping center entrances (on both sides of Corridor G) that function as parking lot entrances.  Indeed, the roads leading to shopping centers along Corridor L in Summersville have some of those same qualities.

[webny99]

So if there's a not quite direct freeway route that's the fastest, having a direct road between the two cities that's a surface road somehow reduces connectivity?

No, and I don't think anyone is arguing that. However, the value/importance of the freeway increases when there's not a direct surface road, since it's needed for all trips, not just long-distance ones.

[webny99]

I think New York would be average. Overall long-distance connectivity is good, and all the major population centers are well served. Outside of the North Country, it's rare to have to use two-lane roads in the middle of any 2+ hour trip.

I forgot about another major exception, which is related to, but distinct from, the connectivity issues to/from Vermont.

From the Mohawk Valley and points west, getting to the I-87 corridor north of Albany requires either two-lane NY 30A/NY 29, two-lane NY 30/NY 67 (which is a slog through Amsterdam) or going well out of the way to stay on I-90/I-87.  From west of Utica, NY 12 or NY 365 to NY 8 is also an option that's roughly equal time-wise and saves significant mileage, but it's almost two hours on somewhat mountainous backroads that are scenic during the day but not suitable for long-distance night driving.

[Rothman]

I think New York would be average. Overall long-distance connectivity is good, and all the major population centers are well served. Outside of the North Country, it's rare to have to use two-lane roads in the middle of any 2+ hour trip.

I forgot about another major exception, which is related to, but distinct from, the connectivity issues to/from Vermont.

From the Mohawk Valley and points west, getting to the I-87 corridor north of Albany requires either two-lane NY 30A/NY 29, two-lane NY 30/NY 67 (which is a slog through Amsterdam) or going well out of the way to stay on I-90/I-87.  From west of Utica, NY 12 or NY 365 to NY 8 is also an option that's roughly equal time-wise and saves significant mileage, but it's almost two hours on somewhat mountainous backroads that are scenic during the day but not suitable for long-distance night driving.

That's connecting nothing to nothing.

[Dirt Roads]

I think New York would be average. Overall long-distance connectivity is good, and all the major population centers are well served. Outside of the North Country, it's rare to have to use two-lane roads in the middle of any 2+ hour trip.

I forgot about another major exception, which is related to, but distinct from, the connectivity issues to/from Vermont.

From the Mohawk Valley and points west, getting to the I-87 corridor north of Albany requires either two-lane NY 30A/NY 29, two-lane NY 30/NY 67 (which is a slog through Amsterdam) or going well out of the way to stay on I-90/I-87.  From west of Utica, NY 12 or NY 365 to NY 8 is also an option that's roughly equal time-wise and saves significant mileage, but it's almost two hours on somewhat mountainous backroads that are scenic during the day but not suitable for long-distance night driving.

That's connecting nothing to nothing.

It's much less of a priority in today's world of Internet connections, but many states place higher priority on the "connectivity" between county seats (or regional commercial centers) and their respective state capitols.  States like Ohio and Georgia have a spoke-and-hub arrangement including non-Interstate freeways, "highways" and [former] "highways" reaching out from their capital cities.  Indiana has a similar hub-and-spoke layout, except that all but one are Interstates.  North Carolina and Kentucky works similarly, except that spines and diagonal branches collect the non-Interstate freeways and "highways". 

This may not the be case in New York.  It certainly isn't in huge states like California and Texas, but quite frankly I know a bunch of government employees in both of those states that had to travel regularly to their department headquarters in their respective capitals.

[vdeane]

I think New York would be average. Overall long-distance connectivity is good, and all the major population centers are well served. Outside of the North Country, it's rare to have to use two-lane roads in the middle of any 2+ hour trip.

I forgot about another major exception, which is related to, but distinct from, the connectivity issues to/from Vermont.

From the Mohawk Valley and points west, getting to the I-87 corridor north of Albany requires either two-lane NY 30A/NY 29, two-lane NY 30/NY 67 (which is a slog through Amsterdam) or going well out of the way to stay on I-90/I-87.  From west of Utica, NY 12 or NY 365 to NY 8 is also an option that's roughly equal time-wise and saves significant mileage, but it's almost two hours on somewhat mountainous backroads that are scenic during the day but not suitable for long-distance night driving.

That's connecting nothing to nothing.

There might not be a distinct medium/large metro to the north (well, Montréal, but that's not the best route to get there from anywhere other than Utica, and maybe not even Utica), but I'd hardly say nothing.  In fact, Google doesn't suggest going I-90->I-87 for ANY suburbs in the Capital District north of the Mohawk River bar Halfmoon and Waterford.  Clifton Park it suggests going through Schenectady to NY 146; for everywhere else in Saratoga County it suggests NY 67.  There's also Glens Falls/Lake George and the Adirondacks as mentioned.  Plus Burlington, VT.  Considering that all of upstate NY bar the Capital District and the Southern Tier are affected by this, it is notable.

[hotdogPi]

I wouldn't say that Ohio has a spoke-and-hub system. It has routes in every direction (including multiple diagonals) that hit smaller areas, such as US 30, US 22 and US 42 (both bypass Columbus), US 35, US 250, and for state route freeways, the jumble in the northeast corner. It also has three major cities, meaning there isn't a single hub.

[webny99]

I think New York would be average. Overall long-distance connectivity is good, and all the major population centers are well served. Outside of the North Country, it's rare to have to use two-lane roads in the middle of any 2+ hour trip.

I forgot about another major exception, which is related to, but distinct from, the connectivity issues to/from Vermont.

From the Mohawk Valley and points west, getting to the I-87 corridor north of Albany requires either two-lane NY 30A/NY 29, two-lane NY 30/NY 67 (which is a slog through Amsterdam) or going well out of the way to stay on I-90/I-87.  From west of Utica, NY 12 or NY 365 to NY 8 is also an option that's roughly equal time-wise and saves significant mileage, but it's almost two hours on somewhat mountainous backroads that are scenic during the day but not suitable for long-distance night driving.

That's connecting nothing to nothing.

Rochester, Syracuse, and Buffalo to the eastern Adirondacks?? That's not nothing at all, that's basically the Upstate NY knockoff version of Denver to the Rockies. Concur with all of what vdeane said. It's a connection I've made myself close to a dozen times.

[Rothman]

I think New York would be average. Overall long-distance connectivity is good, and all the major population centers are well served. Outside of the North Country, it's rare to have to use two-lane roads in the middle of any 2+ hour trip.

I forgot about another major exception, which is related to, but distinct from, the connectivity issues to/from Vermont.

From the Mohawk Valley and points west, getting to the I-87 corridor north of Albany requires either two-lane NY 30A/NY 29, two-lane NY 30/NY 67 (which is a slog through Amsterdam) or going well out of the way to stay on I-90/I-87.  From west of Utica, NY 12 or NY 365 to NY 8 is also an option that's roughly equal time-wise and saves significant mileage, but it's almost two hours on somewhat mountainous backroads that are scenic during the day but not suitable for long-distance night driving.

That's connecting nothing to nothing.

Rochester, Syracuse, and Buffalo to the eastern Adirondacks?? That's not nothing at all, that's basically the Upstate NY knockoff version of Denver to the Rockies. Concur with all of what vdeane said. It's a connection I've made myself close to a dozen times.

Rochester, Syracuse and Bufallo to...nothing.

We were talking about urban connectivity rather than just getting people to their cabins.

[vdeane]

Saratoga is nothing?  I'll try to remember that the next time the Northway is a parking lot all the way down to the Thruway.

Fun fact: the two worse weeks for Northway traffic in the entire year and the ones leading up to Labor Day and Travers.