Georgia could get a black eye this week when 500 high school students from around the country come to Atlanta to compete in the National High School Mock Trial Championship.
The competition, being held between May 7 and May 9, is being hosted by the Georgia Bar Association’s Young Lawyers Division. But controversy is brewing because of an unwillingness to accommodate the request from an Orthodox Jewish school from Brookline, Ma. to not compete on the Sabbath.
So far, the Mock Trial organization has refused to work with the needs of the Massachusetts state champion, the Maimonides School, by moving only two of the 150 trials from Saturday to Friday. The students of Maimonides, an Orthodox Jewish School, observe the Sabbath on Satruday.
A.J. Robinson, president of Central Atlanta Progress, sent me an email about the controversy that he had received from Jewish leaders.
Media packets are going out with the following headline: FREEDOM OF RELIGION ON TRIAL IN ATLANTA: Jewish students are fighting for their day in court.
This is not the first time that the Mock Trial association has faced a similar situation, according to information in Robinson’s email.
In 2005, when the championship was being hosted in North Carolina, the New Jersey state champion asked for a similar accommodation. The Torah Academy of Bergen County did get its request, but the National High Schol Mock Trial Championship was furious and resolved to never accommodate Saturday Sabbath observers.
On May 5, 2005, before the first day of the North Caroline event, the organization adopted that resolution.
Now the Attorney General of Georgia is concerned about the current decision of the organization, and the attorney general’s office has asked that the Georgia Bar and the championship accommodate the Maimonides students.
The Anti Defamation League is expected to get involved. Plus, the students are being represented by two prominent Civil Rights attorneys from Washington D.C. — Nathan and Alyza Lewin.
For more information on this case, you can link to an article about the controversy from the Cutting Edge News; or you can watch a youtube piece.
It would seem that in the international home of Civil Rights and the birthplace of Martin Luther King Jr., we could be more sensitive to the traditional religious restrictions of those who come to town.













Wednesday, May 20, 2009
Letter: Mock trial ranking system fundamentally flawed
By Jeff Kosowsky
Regarding "Letter writer, not mock trial organizers, showed no grace" (May 19): unfortunately, Larry Bakko, the decision support system coordinator for the National High School Mock Trial Championship (NHSMTC), continues to blow a combination of statistical fallacies, lies, and canards to impugn the performance of the Sabbath-observing high school team from Massachusetts rather than graciously conceding a successful accommodation welcomed by the vast majority of students, coaches and observers.
As a Harvard Ph.D. in Applied Mathematics, I am shocked by Mr. Bakko's stunning lack of knowledge of elementary statistics. In the following analysis, it is important to note that Massachusetts was one round ahead of the other teams due to the Sabbath accommodation.
Lets begin with round one. Mr. Bakko claims "They [Massachusetts] faced a limited risk, since many teams were still in flight to Atlanta and were not available to create the same pool of competitors." As any first year statistics student knows, sampling from a random subset (those who arrived early) is equivalent to sampling from the entire pool since there is no reason to assume that those teams with earlier flights were any better or worse on average than teams with later flights. Hence, the first round assignment was statistically fair.
Now for round two. Since Massachusetts lost round one, ideally it would have been matched against another team that lost round one. Instead, Massachusetts faced a randomly selected team from the entire pool including 20 teams that went on to win the first round. In fact, the team they played went on to win their first two rounds and ended up ranking higher. Disadvantage Massachusetts.
Turning to round three. Since by now Massachusetts had lost two rounds, it was appropriate to match with another team that lost the first two rounds. Instead, since Massachusetts was one round ahead, it was matched against a team that had lost only one round. Since half of these first round losing teams go on to win the second round, Massachusetts actually risked being matched with a stronger team than it deserved. Disadvantage Massachusetts.
Finally, round four. Since the Massachusetts record was now 1-2, the closest match was either a weak 1-1 team or a strong 0-2 team. Instead, Mr. Bakko chose a relatively strong 1-0 team who won all 3 ballots on their first round. Disadvantage Massachusetts.
The fact that this fourth round team was the Georgia-B team or that it didn't play all the rounds is a canard. The host state runner-up is only invited to participate (as in this case) when there exists an odd number of teams in a round and has no intrinsic "right" to compete in the tournament. In fact, had Massachusetts not participated, the Georgia-B team would not have not competed at all.
Finally, given that Georgia had won the national championships for the past two years, it is disingenuous to claim that the runner-up state team is necessarily weaker than other participants.
The above analysis clearly proves that contrary to Mr. Bakko's assertions, the first round was statistically fair while the last three rounds if anything statistically disadvantaged Massachusetts.
Now what about Mr. Bakko's claim that one of Massachusetts' wins was against a school that ended up ranking #34 or that Maine, which also ranked #20, played a tougher schedule? Here Mr. Bakko enters into the fallacy of retrospective analysis in a system that prospectively matches teams based on the results of prior rounds.
Indeed, Mr. Bakko now treads on very dangerous grounds since a retrospective analysis of the complete 2009 Power Matching system reveals intrinsic flaws in the rankings that are far more severe than the purported "anomalies" attributed to Sabbath accommodation. For example:
- Louisiana ranked #7 yet lost to #18 and then beat #29, #37, and #22. Indeed, two of the three teams it beat won only one round.
Based on these match-ups, it can be argued that Louisiana merits no higher ranking than #19.
- Similarly, #14 ranked Arizona never beat a team better than #34 (out of 40), yet Mr. Bakko complains that Massachusetts which at #20 beat the higher ranked #33 should be asterisked for facing a weak schedule.
- Finally, #5 ranked Nevada beat #35 and #15, lost to #12 and then beat #21, yet was ranked higher than #12 Kentucky, who beat them head-to-head while only losing to near top-ranked #4. In fact, a total of 7 teams were ranked at least 5 places *lower* than a team they beat head-to-head.
The truth is that it is mathematically impossible to create a matching system that after a mere 4 rounds results in a statistically significant complete ordinal ranking of all teams. Claiming the contrary would be equivalent to saying that one could schedule a football season of just 4 games and not just determine the Super Bowl winner but also delineate the precise ranking of all other 31 teams.
Unfortunately, the NHSMTC Power Matching system suffers from even more significant flaws in that the rankings are based purely on raw win-loss, ballot, and point totals. There is no retrospective normalization of these raw scores to reflect the vastly different strength of opponents, the relative generosity of different judging panels in awarding points, the randomness of assignments within a round (particularly the first round) or even the logically conflicting results of direct head-to-head match-ups.
As a result of this analysis, I modify my previous claim to say that the NHSMTC organizers lack not only grace but also statistical aptitude.
Jeff Kosowsky
Newton, MA
The writer is the father of a Mock Trial participant from Massachusetts
- spam
- offensive
- disagree
- off topic
Like