Minor league stats take leap forward
by Will Lingo
The issue you're holding in your hand may have the most accurate end-of-season minor league statistics ever.
A bold declaration to be sure, but we've been to the source and have seen the way stats were collected this season. And we were impressed.
For those of you who haven't followed the minor league stat soap opera day by day, we had our third official statistician in three years this season. After Howe Sportsdata (now Sportsticker) had the contract for years, Minor League Baseball awarded it to the Sports Network for the 2004 season.
Major League Baseball then decided to take over responsibility for minor league stats in the latest renewal of the Professional Baseball Agreement, the contract between the major and minor leagues. After doing so, it hired MLB Advanced Media, the spinoff company that runs MLB.com among its many other projects, to collect the minor league stats.
Minor league teams, already feeling uneasy after a season of statistical upheaval following years of stability, became even more skeptical after hearing how MLBAM wanted to gather the stats.
Even minor league teams, which aren't always on the cutting edge of technology, recognized this as a step back. They have scored games by computer for years, transmitting the resulting statistics to the company that compiled them--whichever company that was.
Now they could score games however they liked--still by computer, or by hand--but MLBAM would gather the statistics by getting the play by play called in by phone every half-inning.
Step Back Is A Step Forward
On its face, the method doesn't make much sense. You're making the process of gathering the stats much more labor-intensive, and much less efficient.
Until you actually see the process in action at their offices in New York. MLB Advanced Media hired a roomful of people to work their phone bank and do the data entry as teams called in their stats, and gave these people hours of training to make sure they did it right.
And that's why they're gathering stats this way in the first place: to make sure it's done right. When any organization that gathered stats by computer did it in the past, it simply accepted a team's data feed each night and went back and made corrections as needed.
MLBAM wanted to gather the stats half-inning by half-inning so it could catch mistakes when they happened, and so it could have its own employees watching over the process.
MLBAM technicians worked five or six games a night, staggered as much as possible, and followed the audio of each game in addition to taking phone calls from the ballparks. Their job was to enter the stats accurately and seek out potential errors and inconsistencies as they cropped up.
So if a team incorrectly called in a run as unearned or credited a batter with an RBI when he shouldn't really get one, the MLBAM technician would talk to the team about it and try to fix it. If there was a disagreement about a scoring decision, it got passed on to a supervisor, who resolved it.
Many more mistakes were likely avoided before they could ever be recorded, and many beyond that were fixed not within days, but within minutes. So while this system could evolve into teams doing pitch-by-pitch stat gathering over time, for the near term the phone method is here to stay.
Passion For Stats
That's not to say this season hasn't been without its growing pains. We haven't been able to get all the statistics for our Website that we've wanted, and there have been the inevitable details of working together that can only be smoothed out with time.
But what's most important is that we have a minor league statistician that cares passionately about getting the numbers right. We could see it in their eyes and hear it in their voices, and in the end we could see it in their work, which is where it matters most.
We heard several stories, which unfortunately we can't repeat, about MLBAM going to the mat to tell teams there was one correct way to gather and report statistics, and that's the way it was going to be done.
Keep in mind that we're still not sure who the Midwest League batting champion of 2004 was. The official statistics say that it was Peoria's Brendan Ryan, who batted .322. But some argue that it's Cedar Rapids' Howie Kendrick, who batted .367 but didn't have enough plate appearances to qualify. We're frankly not sure because we have no way to know exactly how many plate appearances he had last season.
MLBAM is working on that, and we're trying to help, by building a comprehensive database of all minor league stats from at least the most recent seasons. That will happen down the road, but we can at least be confident that this year's numbers are right.
And in the long run, minor league baseball has a stat provider that wants the numbers to be right and wants gathering and disseminating them to be a process that becomes better and more efficient every year. That's more important than anything else--and the best news we've heard in years.