You do know that we used to have profitable railways in the U.S. when passenger rail here was competitive. But, since high-speed rail doesn't really exist in this country, one must look abroad for precedents, and those, of course, are profitable. So, despite all the economic-development benefits that come from this particular mode and that justify capital expenditures in this infrastructure, instead of airport gates or highway lanes, operating subsidies are not only not necessary; several private profit-maximizing companies are interested in investing in the services, including California High-Speed Rail. In the case of DesertXpress (high-speed rail between Las Vegas and San Bernardino County), private developers are building and financing the capital-intensive infrastructure, itself.
I'm glad you cited the Sprinter because it is an excellent example of most transit being caught in a vicious circle in the U.S. while, in Europe and Asia, virtuous circles are the norm. Levels of service need to increase in order to make transit competitive, and, at the moment, Sprinter doesn't connect major cities or destinations in a timely and convenient fashion. Oceanside, the westerly terminus, has Metrolink, Coaster, and Amtrak services, and Escondido, the easterly terminus, will have California High-Speed Rail. So, let's talk about the ridership of the line after build-out of the more comprehensive system. Similar criticisms are levied by the anti-rail crowd (Reason, Cato, etc.) against the Gold Line trams of L.A. Metro, but they are making these judgments when the corridor is not yet complete. Look at the ridership after the line reaches Claremont, Montclair, and Ontario International Airport. Moreover, transportation investments are a chicken-and-egg proposition by which land uses around transit stations and airports intensify over time, so maturity does matter. Transportation infrastructure, in general, is an asset that realizes its value over the long term.
As for modal share, the virtuous circle is the key. The gains Toronto, and similar cities, might seek are potentially realistic if these places do enough to create the positive-feedback loop that creates exponential, geometric expansion in transit usage.