Friday, October 30, 2009

More on judge assignment

I just realized that comments were being posted (I thought I was getting them sent to my email, but obviously I wasn't). And I love the phrase strikes and dice.

As an habitue of one of those smoke-filled tabrooms, I can say a couple of things. First, protesting that we don't favor our own teams would sound disingenuous, but in reality, we deliberately make sure that there are representatives from more than one school, to prevent that from happening. Of course, anyone who has sat with me in tab knows that my personal motto is, "Make this damned thing end," so the only manipulation I'd be doing, if I could, is closing out the tournament in, if possible, doubles. But seriously, I have often heard accusations made against certain people manipulating tab; as far as I know, none of those people are any longer active in debate, and no one has accused any of my cronies tab team colleagues of anything like that, so at least the northeast is clean.

Secondly, the issue of putting the better judges into the tougher rounds is not really just us making it up. Obviously we can use prior knowledge: we certainly know who all the coaches are, but by default we will highly rank coaches we don't know personally on the basis of the simple idea that, if they're a coach, they know what they're doing until they prove otherwise. (And make no mistake, many coaches quickly and categorically do prove otherwise, but that is up to them, not us.) We also use the experience of each tournament's previous rankings. I keep a spreadsheet of every tournament that I've done with community rankings of every judge in the pools. Community rankings from tournament to tournament tend to remain stable, and I'm able to consult this list when I have no knowledge of my own. Just as a point of information, we've started adding to this list the names of parent judges we have trained ourselves at MHLs, so we know who we can trust at the newer level.

So, realistically, a good tab room ought to be making reasonable and honorable assignments when they are simply not going at random. But this does not subtract from the appeal of strikes and dice. We talked in the podcast about trying it at Ridge, but instead we're going to test MJP (since it's new to me from an operational standpoint, and I'd like to try a little baptism of fire). Maybe we should try S&D at Lakeland, which should be humongous. The big question is, what is the number of strikes that makes sense for a field of, say, 100?


  1. How big of a pool of judges do you have for a field of 100? 50 judges? 25? I think my answer depends largely on the size of the judging pool, more than the field. Something like a strike range of 20-30% sounds reasonable - because that allows teams to strike a high number of judges. Assuming you had a 1 judge for every two competitors and pool of 50 judges - that would mean something like 15 strikes.

    That's a high number, but it shouldn't be inoperable. Assume the worst case and that Madison Memorial and Hen Hud struck 15 different judges and we're debating in round six; there should be about 10 different judges that could fit into that round.

  2. We usually have a ratio of 1 to 3.