For years, Howard County officials, planners, parents and school system representatives have bemoaned the inaccuracies of student enrollment projections.
They've made improving them part of their political platforms, and called the lack of number credibility something that cannot be "tolerated" (Superintendent John R. O'Rourke) and "unacceptable" (County Councilman Christopher J. Merdon).
But now - after a recent in-house assessment that shows the process is not necessarily any better after improvement efforts, definitely not any worse, and overall not nearly as bad as once thought - number crunchers are hoping the criticism will be quashed, or at least quieted.
"Because they're projections, there are going to be error rates," said David C. Drown, who took over the predictions task in 2001.
But the error rate on the whole, he said, is extremely small, involving at the county level miscounts that ranged from 47 filled seats in the best of the past eight years to 419 filled seats in the worst - a tiny fraction of the more than 40,000 total students.
It is at the individual school level where trouble remains.
And that translates to a host of problems, such as not having enough seats for students, planning additions to buildings that do not need them, overstaffing or understaffing schools and missing chances to shut down housing development in areas with more growth than the school system can handle.
The consensus, though, is that Drown is doing a better job than those who came before him.
"David has taken on much more responsibility and worked real hard at it and made improvements," said Jim Eacker.
He is vice chairman of a committee that analyzes the Adequate Public Facilities Ordinance (APFO), which looks to the projections to determine whether development should continue in an area.
"It ain't easy," Eacker said. "I also think he's telling it like it really is much better than his predecessor."
Drown's message, though, is not always liked.
The numbers are critical to growing Howard County. Drown said the system may not be able to improve those figures based on the number of variables - births, construction delays, relocations - that the system deals with.
"With any statistical approximation, the bigger your sample is, the more accurate you will get," Drown said.
But the smaller the population, such as that of a single school, the more room for miscalculation.
Even so, school system officials have determined that the figures serve their needs.
"The numbers can be used for the things that we need them to be used for," such as capital planning, said Deputy Superintendent Sydney Cousin.
Drown said the system provides a decent regional focus, one that can be used to allocate resources, though last-minute shifting may be required if the process overprojects one school and underprojects another, a frequent occurrence. Merdon and Eacker agree.
But critics still say that is not good enough for the Howard school system, particularly those who regularly endure redistricting of their children and claim new schools are being built in the wrong places.
And most stakeholders agree that the numbers are nowhere near good enough for the county to beneficially invoke APFO, halting development, which is supposed to keep the schools from constant crowding.
"If they're not working for APFO, I don't know how they can be working for the school system," Eacker said. "We have the same end objective of having the facilities in place to serve kids when kids get there."
"If they're short 100 seats in a given school, they have to do makeshift arrangements to take care of those kids," he said.
"Is that the end of the world? Probably not, but if you ask the parents of those 100 kids, they'll tell you they're real unhappy."
Two years ago, an outside firm assessed the school system's projection model and recommended changes.
The suggestions included upgrading software that dated to 1986, working more collaboratively with other county agencies - such as the Department of Planning and Zoning - and incorporating housing and birth data in the process.
Those modifications were made, along with a move to align the school's geographical information system platforms to the county's, making cooperative analysis of such things as student migration patterns more feasible.
And a partnership with the county's Health Department has resulted in a key procedure that determines where newborns, previously untracked, live.
But just what those changes mean to overall accuracy will be told only through time.
"We've done a lot of things over the last two years I believe will result in improving the accuracy of the projections," Drown said. "I'm not 100 percent sure they're going to improve ... but we are determined to be better than we are. The process will be ongoing."
Planned adjustments consist of finding a better method to interpret where those newborn babies will be before they start kindergarten, analyzing the effects of full-day kindergarten on the county and improving housing construction and potential birth estimates, which some say requires a crystal ball.
"Everyone should be realistic about the limitations of projections as they relate to managed growth," said school board member Courtney Watson.
"But we should never stop improving the process."
Drown said he expects success to show itself by next year and for people to keep the big picture in mind when evaluating the process.
"People remember the big misses," he said. "It's like the weatherman. Nine times out of 10 the guy's right, but people remember when he's wrong."