3 ms·

Speed would be nice, but speed is not the problem. If the trains just ran on time, 50mph would be just fine. The problem with Amtrak is frequent multiple-hour d

Speed would be nice, but speed is not the problem. If the trains just ran on time, 50mph would be just fine. The problem with Amtrak is frequent multiple-hour delays that stack up. The schedule is totally unpredictable.

saagarjha 3y ago
High-speed rail typically excels when it can get you somewhere in a few hours because you wouldn't want to hop on a plane for that. At 50 mph, that gets you…basically nowhere. If you're doing 200 then you can cross most states in that time.
mtalantikite 3y ago
For me it's also cost. I have family in the Boston area and often traveling from NYC on Amtrak is at least as expensive -- and usually more expensive -- than flying. But I'd also love if we could go faster than 50mph. TGV in France, which launched 41 years ago, travel between 167mph and 198mph [1]. [1] https://simple.wikipedia.org/wiki/TGV https://simple.wikipedia.org/wiki/TGV
jrockway 3y ago
"On time" is a metric that is focused on heavily, but to me it's meaningless. The LIRR is "on time" via schedule padding; a 50 mile trip to Ronkonkoma takes 80 minutes on the fastest express train. With line speeds largely 80mph throughout the route, commuters are robbed of an hour every single day. But hey, at least the "on time performance" metric is at 99%. Easy to do when your average speed is 37mph on 80mph track. The LIRR Today has a great article, that I cannot for the life of me find, where the author used train speed data between stations to figure out what speed "most trains" accomplished (so accounting for curves, station stops, etc.) and redid the schedules according to that data. Without padding, everyone that uses the LIRR would save hours a week in commuting. Incidentally, the LIRR has been moving away from higher train speeds and better connections to reduce travel time in favor of arriving at terminals within 6 minutes of the scheduled time. Schedule padding, replacing 80mph switches with 60mph switches (for the Elmont work), and removing all scheduled connections at Jamaica. I think it's crazy and I'm glad I don't commute to the city from Long Island. The connections at Jamaica disappearing is the most sad to me; that station has a really unique setup where 3 trains arrive at once, and you can transfer through the middle train to get to the other two destinations. It used to work like clockwork, but obviously if the middle train is late, the OTP for 3 trains decreases. Since that's the metric they care about, and not "can I get from any city terminal to any destination easily", that's what gets optimized out. I don't think it's good. It's nice if the trains run on time, but I'd rather be 20 minutes late once a week than spend an extra 5 hours on the train every week. But, not what the agency values.

saagarjha 3y ago

High-speed rail typically excels when it can get you somewhere in a few hours because you wouldn't want to hop on a plane for that. At 50 mph, that gets you…basically nowhere. If you're doing 200 then you can cross most states in that time.

mtalantikite 3y ago

For me it's also cost. I have family in the Boston area and often traveling from NYC on Amtrak is at least as expensive -- and usually more expensive -- than flying. But I'd also love if we could go faster than 50mph. TGV in France, which launched 41 years ago, travel between 167mph and 198mph [1]. [1] https://simple.wikipedia.org/wiki/TGV https://simple.wikipedia.org/wiki/TGV

jrockway 3y ago

"On time" is a metric that is focused on heavily, but to me it's meaningless. The LIRR is "on time" via schedule padding; a 50 mile trip to Ronkonkoma takes 80 minutes on the fastest express train. With line speeds largely 80mph throughout the route, commuters are robbed of an hour every single day. But hey, at least the "on time performance" metric is at 99%. Easy to do when your average speed is 37mph on 80mph track. The LIRR Today has a great article, that I cannot for the life of me find, where the author used train speed data between stations to figure out what speed "most trains" accomplished (so accounting for curves, station stops, etc.) and redid the schedules according to that data. Without padding, everyone that uses the LIRR would save hours a week in commuting. Incidentally, the LIRR has been moving away from higher train speeds and better connections to reduce travel time in favor of arriving at terminals within 6 minutes of the scheduled time. Schedule padding, replacing 80mph switches with 60mph switches (for the Elmont work), and removing all scheduled connections at Jamaica. I think it's crazy and I'm glad I don't commute to the city from Long Island. The connections at Jamaica disappearing is the most sad to me; that station has a really unique setup where 3 trains arrive at once, and you can transfer through the middle train to get to the other two destinations. It used to work like clockwork, but obviously if the middle train is late, the OTP for 3 trains decreases. Since that's the metric they care about, and not "can I get from any city terminal to any destination easily", that's what gets optimized out. I don't think it's good. It's nice if the trains run on time, but I'd rather be 20 minutes late once a week than spend an extra 5 hours on the train every week. But, not what the agency values.