The Long Island Rail Road has lowered its November on-time performance numbers in response to commuter advocates who said the agency had manipulated statistics to make it appear that train service was better than it was in the weeks after superstorm Sandy.
The LIRR had reported last week in its monthly Train Talk newsletter that 91.9 percent of its trains arrived on time in November -- a month of major service disruptions caused by the Oct. 29 storm.
But at Monday's Metropolitan Transportation Authority railroad committee meeting, LIRR officials announced they would adopt a November on-time performance rate of 88.1 percent.
The difference stems from the definition of a train being late.
Normally, the LIRR considers a train late if it arrives 6 minutes or more after its scheduled time.
After Sandy, the agency advised customers to expect delays of as much as 15 minutes. With the assumption of delays, the LIRR reasoned that a train would be considered late only if it arrived 16 minutes or more after its scheduled time.
The LIRR Commuter's Council challenged the post-Sandy calculations, calling them "almost like a slap in the face" of riders who knew better.
"I saw people handing it [the Train Talk newsletter] around . . . and laughing," Commuter's Council chairman Mark Epstein said. "It didn't reflect the riders' everyday experience."
The statistics account only for trains running during the LIRR's limited schedule in November -- about 3,500 fewer than usual.
"We know how important it is to be accurate," LIRR customer service vice president Joe Calderone said of on-time performance numbers. "We know we had a lot of crowded trains and a lot of canceled and late trains, and we want the statistics to reflect the customer experience."
Damage inside two East River rail tunnels owned and maintained by Amtrak severely limited the LIRR's capacity to run trains into and out of Penn Station. Amtrak completed repairs to the tunnels this past weekend, Calderone said.
Peter Haynes -- president of the LIRR Commuters Campaign, another advocacy group -- said the debate proved LIRR on-time performance numbers were "fake and meaningless."
Using the LIRR's logic, on-time performance during the storm, when no trains were running, "was 100 percent because no trains were late," Haynes said.