Allow me to describe the bigger picture.
Many books could be written (and have been written) on the state of train service in the US when Amtrak was established. Suffice it to say, it was a complete disaster, passenger and freight alike.
The so-called "rationalization" programs of the railroads at the time were poorly planned out, and driven by short-term political considerations. But at the time, no other plans made any more sense. The federal government was massively funding the competition (roads & air) while taxing the railroads, and so there was no rational management choice to make; it was bankruptcy, government takeover, government subsidy, or asset-stripping for all of them.
Regarding passenger trains, Amtrak was handed a network of extremely long-distance trains where the shorter local trains had mostly been cut. This doesn't make any business sense. When the railroad network was originally constructed, the short local passenger trains were run first, and the longer-distance trains ran as overlays on the network.
Depending on how you do your accounting, the longer-distance trains may look like the most profitable in that situation -- and that's how they looked to most of the railroad execs at the time -- but they make their profit based on network effects and depend for that profit on the existence of the local branch lines. The all-sleeper trains were the pinnacle of the "high end overlay", and depended on a huge underlying network, including many coach trains per day along the same routes.
It is not a coincidence that Amtrak's most successful lines run between cities with substantial urban rail and "commuter rail" operations. And it is not a coincidence that the most successful "long distance" routes (Auto Train aside) are connected to many "corridor" routes as well as to urban rail and commuter rail routes.
In 1971, most of the "commuter rail" services (local services) had already been cut, nationwide. The Northeast (Boston to DC) retained a fairly large collection, operated by government order; these mostly ended up in Conrail and were transferred to local authorities in 1983, while NYS bought the LIRR outright. The Chicago area started supporting its local routes in 1974 through Metra (and NICTD). The Peninsula Corridor in the SF Bay Area was adopted by Caltrain in 1980. Everything else either went to Amtrak or was cancelled entirely.
In 1971, the urban rail systems had already been decimated and were still shrinking. The complete "legacy list" for the US (including subways, els, and streetcars) is Boston, New York, Newark, Philadelphia, Cleveland, Pittsburgh, Chicago, New Orleans, and San Francisco -- nine cities. And even those systems were shrinking, with New Orleans down to a single streetcar line which was operating only because it was a "historic landmark". As far as I can tell, all the other urban rail systems in the US were dismantled well before 1971, with the exception of the El Paso-Juarez streetcar which lasted until 1973 or 1974.
There was some new urban rail construction in the 1970s, but it seemed to go out of its way to avoid the old intercity railway lines. BART and MARTA specifically failed to connect with all the existing trains, and WMATA made very little effort to connect to the adjacent intercity rail stations despite running right next to the railway lines. This behavior continued into the 1980s, with Baltimore, Buffalo, Portland, Sacramento, San Jose, and Los Angeles all building urban rail lines which you couldn't take to catch an intercity train; of the new lines in the 1980s, only San Diego had connectivity from the start. (Some of these have since been fixed.)
It turns out, as several studies of the replacement of streetcars by buses show, that on the whole people don't really like buses; replacing rail with bus makes ridership drop, a lot, consistently. The bustitutions of the 1920s/1930s/1940/1950s/1960s destroyed the underlying local service which was feeding the longer-distance train services. So Amtrak was handed a system which was missing its underpinnings, except in a few areas. Given this, it is actually surprising that Amtrak did as well as it did during the 1970s.
The areas where the underpinnings were retained or have been rebuilt are the areas where Amtrak is currently most successful. The revival of urban rail and commuter rail services in cities across the country provides the base which allows for longer-distance trains to thrive.
However, it's gonna require a lot more than that before all-sleeper trains can be supported. Notably, there *are* still all-sleeper trains (CityNightLine's Copenhagen-Basel route) running through Germany... but they are amidst an enormous supply of other trains, including a number of CityNightLine sleeper trains with coaches (though all the coaches are rather "Pullman-like"), a huge number of intercity day trains, and extensive urban rail services.
If enough rail improvements happen in the Northeast and the Midwest, we might eventually see an all-sleeper train from Chicago to New York again, but not until after we have a lot of day trains running. Likewise, if there are enough improvements in Virginia and North Carolina, we might eventually see something similar from Boston to North Carolina, but it's a long way off.