Disclaimer: This post will be long.
My point is not that CSX doesn't have the bulk of intermodal traffic, because they do, in and out of New England, but it is rather that NS isn't competitive. That is simply untrue. However, several factors in the past few years have changed the traffic advantage in CSX's favor, namely their opening of the double stack route. NS will follow suit, mostly because they understand what they need to be successful. But at the same time, we have seen a shift in traffic over the past few years from NS to CSX. This has namely been with JB Hunt playing the game and going either railroad. JB Hunt of course is so large that they can do this on a system basis, as in reality their shift to using CSX is to utilize other capacity that has outpaced NS as a whole. However, what you are seeing in New England is unprecedented, as you have a large shift from NS to CSX. This is largely attributed to service issues back in Feb of 2015... where the line had to be embargo'd a few times while they cleared snow. Fact of the matter is, in shippers mind that still is recent memory and JB Hunt has routed some stuff via CSX to Chicago, which is the NS' primary lane into New England, due to it.
In time you'll likely see that come back and I think you have with trains once again getting longer. Of course, they still aren't like what it had been a few years ago with a full intermodal train and a mixed auto/intermodal in Ayer each day. Intermodal as a whole year-over-year is down, with NS seeing drops of around 10%, but it has been forecasted to see a strong second half of 2016 with intermodal being only one of the sectors to continue growth. That is because the primary target of intermodal isn't competitors rail lines, but highway to rail conversions. That will continue to be the focus for some time to come as fuel prices start to increase, and shippers get more innovate folks who are willing to try intermodal. Also remember, a lot of people jumped onto intermodal 15-20 years ago and had bad experiences... boxes getting lost or stranded for days. The physical intermodal movement was lightyears ahead of the technology, and only now are you seeing track and trace starting to catch up with the ability to track individual equipment (boxes now have RFID, AEI tags like rail cars and also some have home-grown tracking technology with GPS in them).
I think it's unfair to say NS made a marginal investment to get into New England, I think they had to put up a lot. At least $200M to get the PAS line, plus the investments that came afterwards in expansions, equipment and other resources. That is only for a small controlling portion on a line that is very important to them. Let's also count the D&H south line purchase in that and put their intermodal dream more in the $400M range for New York and New England. Mechanicville certainly serves parts of Western Mass not just the capital district. In reality, a daycab can go no further than 150 miles from the terminal but regional drivers can go up to 300 (and they will). You have to remember that CSX uses Syracuse as their primary upstate New York intermodal terminal, which is also located on their direct line to Chicago. That pales in comparison to the footprint NS has in their similar fortress in Harrisburg only a couple hundred miles away, which has at least 3 facilities all in the area. Allentown, Taylor, and Mechanicville all cover the remaining territory up into New York and Western Mass.
That is to say, just because you don't see 20 intermodal jobs a day out of the Capital area across to Buffalo doesn't mean the traffic isn't moving over NS, it simply says that the traffic enters through a different NS terminal by truck (Up to 150 miles practically) and goes via the Harrisburg/Pittsburg line. I can tell you first hand that the Taylor yard in Scranton is in a huge blind spot for CSX, which must bring boxes in from Kearny or Syracuse to compete in that market (barely). At the end of the day, they have no lines that close and they can't turn boxes as fast as Hunt or Hub on that terminal which is close to the likes of P&G at Tunk and Walmart at Tannersville, ect (Pittston, Carbondale and others in the area have large RDCs too). For intermodal dray carriers to make their buck, they have to keep drays per driver high, and that is done in short moves. Intermodal drivers can spend half their day waiting on paperwork, drop & hooks, finding empties, and a rash of other things. Due to this, they'd prefer to stay as close to their terminal as possible (within 50 miles is ideal).
The line into New England on PAS, as cumbersome as it may be, is only as strong as the NS system as a whole. When boxes hit Chicago, they need to be mounted and ready for a driver. Also, remember a lot of boxes are crosstowned to other ramps instead of steel wheeled. The steel wheel lanes fine themselves populated by the likes of UPS, the giant in the room. With CSX having the lock down on UPS traffic in Worcester, it certainly gives them a service advantage. You never know, you could see UPS change sides or hedge in the future with improved service on PAS, but that's about the only service that is dependent on hours not days (I think UPS has a ridiculous requirement for on time service by the RRs but pay the most and has top priority). That's important to remember too, the railroad's customers are the dray carriers, not the end user. However, they like to support the success of one another, but often times you see the railroad being the one who mis-handles and damages boxes, or losses paperwork ect (I know some terminal managers who would outright claim otherwise but I disagree).
As I mentioned back a few paragraphs, technology finally caught up. Ingates are faster and terminal dwell time is way down. Now drivers can use smartphone apps to report issues or find paperwork. Before, a driver might sit at a customer for 3 or 4 hours while waiting for office people to find a bill of lading. Now you can access that easily and get a digital copy right to the customer. As a driver, that makes you more product, especially when the paperwork isn't on the box like it should be. Shippers are also getting better at loading containers, which have to be optimized at a slightly smaller weight than traditional van truck. Because the tare weight of the box is heavier than a van, loads should be less than about 43,500lbs. Once again however, this took a while for shippers to catch on, and as such, intermodal carriers spent a lot of time re-working loads, which can get lost in a secured lot or relay yard for days while everyone sorts out the details. You'd be surprised how much of this can get lost. The other flip of the coin for everyone involved is who takes on the burden of owning chassis. Dray carriers didn't want it at first, and eventually you had railroads step up and own chassis fleets. This is starting to change now with dray carriers managing their own chassis which gives them greater control over the repairs and effective use of the equipment, also they don't have to pay per day for chassis to railroads.
This whole long rambly post is I guess to say that there is more than meets the eye to intermodal, it isn't just the railroads driving it, but also the draycarriers and the shippers. In terms of surface transportation, intermodal is fair new (intermodal only 30 years old, trucks about 100, railroads 200 and boats 1000+) so you still have some people in charge of supply chains that think in terms of truck or rail car. But intermodal, and more specifically domestic (I've been talking about domestic the whole time, international is a whole different model) has a long period of growth ahead of it. Look for continued advancements in technology, consolidation in the trucking sector (ala XPO, bring all the modes together or LTL, Van and Intermodal) and capacity expansion on the railroads. Once again, the intermodal guys can only grow as fast as the rail capacity, and right now CSX is beating them out. It takes about 60 loads per intermodal train (each way mind you) to make money, so consider that when you see the trains coming in. All those single stacked boxes present a capacity loss of 1 box, a real challenge when you have the tunnel a few miles long. Even still, they are still making money and they will fill out the train with density once they open up the capacity, they just need Fink & the boys to play ball (or get out of the way).