Skip to Content

NY to SF, Transit Times over History

Taking the fastest means of transit available to the public at the time, how long did it historically take to get from New York to San Francisco? Here's what I found:

Year Time (Hours) Log Scale Note
1810 4320 3.635 Astoria Overseas Expedition (Sea route around Cape Horn in Patagonia.)
1849 4320 3.635 Duxbury (Same)
1853 2136 3.330 Flying Cloud (Same route, faster ship.)
1876 83 1.919 Transcontinental Railroad
1934 26 1.415 Boeing 247 (Prop Plane. 10 Passenger Capacity)
1937 17 1.230 DC-3 (Prop Plane. 32 Passenger Capacity)
1959 6 0.778 American Airlines introduces first commercial jet route.
2017 6 0.778 Today's in-air times are similar to those of 1959.

Here's what that looks like on a linear, and on a log (base 10) scale:

Two things stand out. The first is how much of a difference the transcontinental railroad made; a trip that used to take 3 to 6 months by ship could now be made in 3.5 days by rail, a reduction of more than 96%. To scale today, that would be like reducing the travel time from New York to San Francisco to under 15 minutes.

The second is the well-documented fact that transit times haven't gotten any faster since 1959. That isn't to say that planes haven't gotten faster; the SR-71 Blackbird could make the flight in under 1.5 hours. But there are no commercial options that fast. (As others have pointed out, airfare is much cheaper today than it was in 1959; so while speed hasn't improved, costs have.)

Using the log scale chart, we can fit a "Moore's Law" line to the data:

The slope of this line is -0.0169; very, very roughly, this means that for the period under consideration (1810 to 2017), travel times have gotten 10x faster every 59 years. If this trend were to continue, then 59 years from now, we "should" be able to travel from New York to San Francisco in 36 minutes instead of 6 hours.

If you like looking at tech progress through this lens, I recommend J Storrs' Hall's Where is my Flying Car?" Hall makes the argument that the underlying cause of technological stagnation in the physical world is the lack of progress in "usable energy available to our civilization", as measured by kilowatt consumption per capita per year:

Storr calls this the "Henry Adams Curve". Playing this out: It takes about 7.2 megawatts to keep a Boeing 737 flying at a constant altitude. Assuming about 85% of the fuel is spent cruising (vs takeoff, landing, & descent), we're talking about 40 megawatt hours to fly from NY to SF, which is about 225 kilowatt hours per passenger on a fully loaded flight. If we had continued along the Henry Adams curve since 1970 and spent a constant fraction on transcon travel, we'd have a "budget" of 6.7 megawatt hours per passenger for this trip rather than 225 kilowatt hours. In rough terms, that's nearly enough to fly private with 1 other person (based on numbers for the Bombardier Challenger 650).