Mending the Montauk Branch Delay Mess, Part 2

These days, in our look into the Montauk Branch delay mess, most of our focus falls on the hideous delays that tend to build up on Sundays.  In the past couple months the LIRR has had quite a few Sundays they would probably like to forget.  There is usually one or two things that go wrong and then the entire day is ruined.

Compounding delays is just one of the hazards of single-tracked territory.  And, unfortunately, there's very little that the LIRR can do about it.  Double tracking from Sayville to Montauk will never, ever, ever, ever happen.  It's way to far for way to few people.

So if there's no real way to stop delays from compounding once they get rolling there's only two things the LIRR can do:

1. Stop the original delay from happening in the first place.
2. Build in measures to lessen delays once they get started.

The second one of those two is a tricky bit.  For the most part, the only way to stop delays once they get going is to have a buffer built into the schedule.  This concept, called "schedule padding" is a double edged sword.

Schedule padding works like this: lets say it takes 30 minutes for a train to get from point A to point B.  Now, logic would dictate that you would schedule the train to stop at point A at 4:00pm and point B at 4:30pm.  But when you pad the schedules you build yourself in a buffer so if something goes wrong the whole place doesn't crumble down on top of you.

So if the railroad wanted to pad the schedule for the train in the example above it would schedule 45 minutes for the train to get from point A to point B.  That way, if the train is delayed for as many as 15 minutes, the train can still arrive at point B on time.

Now like it or not, the practice of schedule padding is widely used among the railroads of the United States.  Amtrak pads the h-e-double hockey sticks out of their long distance trains (and yet they still suffer such awful OTP's, but I digress).  Even the LIRR pads their schedules.  They don't pad them all that much, but they do.  A one or two minute cushion built into the schedule added to their six minute grace period for arriving "on time" gives them some wiggle room if something goes wrong.

And the LIRR pads schedules on the Montauk branch too.  Most particularly, schedules get padded a lot when a train has to meet another one in single track territory.  Sometimes buffers as long as 10-15 minutes are built into schedules.

Unfortunately, however, the schedule padding tightrope can be a dangerous line to walk.  You can't pad the schedules too much.  We don't want it to take 4 1/2 hours to get to the Hamptons.  There might not be any delays and you'll get there "on time," but who on earth wants to sit on a train for a 270 scheduled minutes?

So it looks like schedule padding isn't going to fix our problems.  So that leaves us with option 1: stop the original delay from happening in the first place.

"No kidding. Sherlock, but how do we do that?" you might say.  Unfortunately, there is no real way to figure that out.  It's really a matter of luck.  Hopefully, the scheduling gurus will keep tinkering with the schedules until we can finally get to the point where no train is extraordinarily delayed.

Take the Friday afternoon migration of thousands of Long Islanders and city dwellers to the Hamptons.  If I had this blog this time last year, chances are I would be talking about how atrocious the Friday afternoon delays were.  In previous years the Friday evening trains were often the most egregiously delayed.

But, for the most part, this year the Friday trains have been very, very well behaved.  Their improvement over last year is nothing short of phenomenal.

"Okay, you got me going, how did they do it?" Well, they accomplished option: they stopped the original delay from happening in the first place.

In years past, the culprit--that first delay that got the ball rolling--was frequently one train in particular.  The Cannonball.  More often than not, the Cannonball would be so delayed that it would ruin the hope of any other train possibly arriving anywhere close to "on time" for the rest of the night.

You see, the pre-2013 Cannonball was always scheduled to arrive Jamaica at 4:22pm and leave at 4:25pm.  That means they left three minutes--only three bloody minutes-- at Jamaica to get hundreds and hundreds of people off of Penn and Brooklyn connections and cram them into 12 C3's.  I can bet that it was darn near impossible to get all those people off the connections, across the platforms, onto the C3's, seated, and all passengers and overflow thereof firmly inserted into the car's vestibule within 180 quick seconds.

And there it was.  They weren't 20 minutes into the journey and the train was already ten minutes down...and then it only gets worse form there.  Those 10 minutes eventually compounded into 20 minutes, then 40, then 80, then 160, and then the rest of the evening's down the toilet.

But all that has pretty much gone away.  I must say, I was a bit skeptical at first when the news came out in April that the Cannonball's western terminus would be moved to Penn Station and the Jamaica stop would be dropped.  But, as it turns out, the move was nothing short of brilliant.  Without having that pesky Jamaica stop delaying things, the train can now get all the way to the Hamptons without suffering any serious initial terminal delay.

All in all, it looks like the changes made to the Cannonball plus some well placed schedule padding have all but cured the Friday evening delay mess.  Now it's not perfect, by any stretch.  Perfection is not to be expected of a single-tracked MBS rail line with all that many people.  But the delays are nothing like last year's.  They're reasonable--10-20 minutes.

At this point in the year, the worst delay to the Cannonball has been only 20 minutes (and that was on the very first day).  That's right, 20 minutes was the longest time the Cannonball was delayed so far this year.  And the train has even arrived "on time" twice!  The average delay to the Cannonball from Memorail day weekend through two weeks ago was only 9.8 minutes.  That's pretty impressive for a train that would frequently score ten times that amount of delay minutes this time last year.

And that outstanding OTP performance has made all of the trains that follow the Cannonball perform much better as well.

So it looks like the Friday evening rush has been "cured" (or as good as it is going to get).

But how do we now fix the Sunday mess?  Unfortunately, it is difficult to pinpoint one specific troublemaker.  Over the next week I will take a more through look through the OTP records for the Sunday Montauk trains and I will report back soon.

Stay tuned!

For more information on this topic, see Mending the Montauk Branch Delay Mess Part 1 and Part 3.