Page 1 of 1

NAQT distributions posted

Posted: Mon Mar 09, 2009 2:53 pm
by Important Bird Area
NAQT is pleased to announce that our distribution is now available on naqt.com.

High school distribution

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 3:14 pm
by at your pleasure
I personally think that lit and myth should be split to 4/4 lit and 1/1 myth, which garuantees a reasonable balance between lit and myth. Still, I'm glad to see NAQT make itself more transparent by doing stuff like this.

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 3:32 pm
by Sen. Estes Kefauver (D-TN)
I appreciate this move towards openness in your new role with NAQT, and look forward to more. THank you Jeff.

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 3:32 pm
by Important Bird Area
Anti-Climacus wrote:I personally think that lit and myth should be split to 4/4 lit and 1/1 myth, which garuantees a reasonable balance between lit and myth.
Do you mean balance between packets within a tournament set?

If so, we already do this:

For IS #76:

Round (Myth, Religious Literature, Non-myth Non-religious Literature)

1 (1/1, -, 4/3)
2 (1/1, -, 3/4)
3 (0/1, -, 5/3)
4 (1/0, -, 3/5)
5 (1/1, -, 4/3)
6 (0/1, -, 4/4)
7 (-, 1/0, 4/4)
8 (1/0, 0/1, 3/4)
9 (0/1, -, 5/3)
10 (1/0, 0/1, 3/4)
11 (-, 1/0, 4/4)
12 (-, 0/1, 4/4)
13 (0/1, 1/0, 4/3)
14 (-, 1/0, 3/5)
15 (-, 0/1, 4/4)

This looks slightly suboptimal to me, in that the religion tends toward the end of the tournament and the myth towards the beginning, and in that some rounds have 1/1 myth and some have nothing; but I don't believe it's wildly out of balance.

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 4:03 pm
by TheCzarMan
I don't like the Science Dist., could go elsewhere to Fine Arts or Social Science.

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 4:13 pm
by Matt Weiner
Thanks for doing this.

I noticed that the high school distribution says it's for IS and IS-A sets; is there a reason HSNCT is not mentioned on that page?

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 4:32 pm
by Important Bird Area
Matt Weiner wrote:Thanks for doing this.

I noticed that the high school distribution says it's for IS and IS-A sets; is there a reason HSNCT is not mentioned on that page?
HSNCT uses the same weights as the posted distribution. I've asked R. to add this clarification.

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 7:14 pm
by AlphaQuizBowler
That's a bit too much science and, if myth is in with lit, not enough lit, in my opinion.

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 8:13 pm
by Deviant Insider
Thank you very much for posting this and engaging in this discussion--there is a good discussion going on in the college section.

My biggest complaint about the distribution is the PC_Sport 1.7/1.7. Getting that down to 1/1 would be an improvement, and the extra questions could go into any category.

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 8:15 pm
by Blackboard Monitor Vimes
Shcool wrote:Thank you very much for posting this and engaging in this discussion--there is a good discussion going on in the college section.

My biggest complaint about the distribution is the PC_Sport 1.7/1.7. Getting that down to 1/1 would be an improvement, and the extra questions could go into any category.
The extra questions could go into lit, which is grossly underrepresented compared to history and science given that religion and myth are shoved into that distribution currently.

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 8:28 pm
by cvdwightw
Initially I thought the huge increase in the science weights was due to mathcalc, but it's not. I agree with reducing the science distribution to something like 78/68 (eliminating 4 mathcalc and 12/12 proportional to current weights). The resulting 16/12 could be funneled mostly into literature to give that equal standing with history, with the rest spread across other categories. I understand the rationale for having relatively low numbers of philosophy, social science, and even perhaps arts (low conversion rates), but I don't understand why academic literature has less than 4/4 a packet. Are people really not converting literature questions?

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 8:38 pm
by Mechanical Beasts
cvdwightw wrote:I understand the rationale for having relatively low numbers of philosophy, social science, and even perhaps arts (low conversion rates), but I don't understand why academic literature has less than 4/4 a packet. Are people really not converting literature questions?
This. Moreover, I can even understand not giving world lit 1/1 per packet at this level, but I can't understand giving it 2/2 per tournament. Unless 4/4 of misc lit is shadow world lit, I think that can stand to be increased a bit.

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 8:53 pm
by Important Bird Area
everyday847 wrote:
cvdwightw wrote:I understand the rationale for having relatively low numbers of philosophy, social science, and even perhaps arts (low conversion rates), but I don't understand why academic literature has less than 4/4 a packet. Are people really not converting literature questions?
This. Moreover, I can even understand not giving world lit 1/1 per packet at this level, but I can't understand giving it 2/2 per tournament. Unless 4/4 of misc lit is shadow world lit, I think that can stand to be increased a bit.
At least some of it is shadow world lit; but it is also a long-standing trend in our feedback that people are not, in fact, converting literature questions at the same rate as history and science. Continental European and non-Western literature have been particular problem areas (yes, this includes those A-series Stendhal tossups we were talking about a while back).

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 9:09 pm
by AKKOLADE
Thank you for posting the distribution on both the high school and the college levels, NAQT. I really appreciate this move towards openness.

Re: NAQT distributions posted

Posted: Mon Mar 09, 2009 10:58 pm
by Huang
It's disappointing to see Computation/Geography/"Sports"/Current Events/General Knowledge comprise 7.9/7 of a NAQT packet. Out of these 5 categories, "sport"'s distribution of 1.7/1.7 undercuts the "A" in NAQT the most. Why can't geography be reduced to 1/1 and "sports"/current events/general knowledge form 1/1 of a packet? Or am I unaware of certain requirements NAQT must fulfill?

Re: NAQT distributions posted

Posted: Tue Mar 10, 2009 12:19 pm
by Important Bird Area
bt_green_warbler wrote:
Matt Weiner wrote:I noticed that the high school distribution says it's for IS and IS-A sets; is there a reason HSNCT is not mentioned on that page?
HSNCT uses the same weights as the posted distribution. I've asked R. to add this clarification.
R. informs me that while the weights for HSNCT are in fact very similar to the IS distribution, the HSNCT distribution shares certain features with our college distribution (notably the split of philosophy questions to ensure balance among eras).

Therefore we've posted a third page with the complete HSNCT distribution.

Re: NAQT distributions posted

Posted: Tue Mar 10, 2009 12:38 pm
by The Laughing Man
everyday847 wrote:
cvdwightw wrote:I understand the rationale for having relatively low numbers of philosophy, social science, and even perhaps arts (low conversion rates), but I don't understand why academic literature has less than 4/4 a packet. Are people really not converting literature questions?
This. Moreover, I can even understand not giving world lit 1/1 per packet at this level, but I can't understand giving it 2/2 per tournament. Unless 4/4 of misc lit is shadow world lit, I think that can stand to be increased a bit.
Especially at HSNCT, the field is strong enough that there is convertable world lit that there should be at the very least .5/.5 world lit per round. Really, I think at HSNCT American, English, Continental and World could be roughly equal without hurting conversion too much.

Re: NAQT distributions posted

Posted: Tue Mar 10, 2009 3:04 pm
by First Chairman
bt_green_warbler wrote:NAQT is pleased to announce that our distribution is now available on naqt.com.

High school distribution
If it hasn't been said enough (and it should be said more), thank you.

Re: NAQT distributions posted

Posted: Tue Mar 10, 2009 3:14 pm
by Matt Weiner
The Laughing Man wrote: Really, I think at HSNCT American, English, Continental and World could be roughly equal without hurting conversion too much.
That's more than even collegiate distributions do--world lit is 1/1 in ACF and ACF-like distros, but that's out of 5/5 including the extra 1/1 lit for tiebreakers. I'm not sure I would want to write 18/18 world lit for HSNCT or NSC; I think there are some problems with conversion when you get to that much. With that said, yeah, only 2/2 per tournament is far too little.

Re: NAQT distributions posted

Posted: Tue Mar 10, 2009 3:19 pm
by Mechanical Beasts
Matt Weiner wrote:
The Laughing Man wrote: Really, I think at HSNCT American, English, Continental and World could be roughly equal without hurting conversion too much.
That's more than even collegiate distributions do--world lit is 1/1 in ACF and ACF-like distros, but that's out of 5/5 including the extra 1/1 lit for tiebreakers. I'm not sure I would want to write 18/18 world lit for HSNCT or NSC; I think there are some problems with conversion when you get to that much. With that said, yeah, only 2/2 per tournament is far too little.
You forget that State College gets fed up when 62: A Model Kit appears in the leadin to that oh-so-easy Cortazar tossup. (I mean, hasn't every high school team read Around the Day in Eighty Worlds?)

Snark aside: yeah, I agree. Having planned out world lit for HFT with Kyle, 10/10 was around the boundary of what felt comfortable, and even then it felt like there were too many common-link-on-a-country questions.

Re: NAQT distributions posted

Posted: Tue Mar 10, 2009 4:34 pm
by Blackboard Monitor Vimes
bt_green_warbler wrote:
bt_green_warbler wrote:
Matt Weiner wrote:I noticed that the high school distribution says it's for IS and IS-A sets; is there a reason HSNCT is not mentioned on that page?
HSNCT uses the same weights as the posted distribution. I've asked R. to add this clarification.
R. informs me that while the weights for HSNCT are in fact very similar to the IS distribution, the HSNCT distribution shares certain features with our college distribution (notably the split of philosophy questions to ensure balance among eras).

Therefore we've posted a third page with the complete HSNCT distribution.
Glad to finally be able to see these, but why on earth is there more current events than fine arts, especially at HSNCT?

Re: NAQT distributions posted

Posted: Tue Mar 10, 2009 5:17 pm
by Important Bird Area
MLWGS-Gir wrote:why on earth is there more current events than fine arts, especially at HSNCT?
Some of the contributing factors here:

1. NAQT believes that our IS distribution and HSNCT distribution should be broadly similar, to ensure the fairness of qualification. (Consider here the case of a marginal HSNCT team that is strong in fine arts: if the HSNCT were significantly shifted, they could justifiably claim that they would perform better at HSNCT than a team that qualified ahead of them by answering that extra current events tossup.)

2. Fine arts, like philosophy and social science, sees consistently weak conversion numbers at lower levels of difficulty.

Re: NAQT distributions posted

Posted: Tue Mar 10, 2009 5:22 pm
by Matt Weiner
bt_green_warbler wrote:Consider here the case of a marginal HSNCT team that is strong in fine arts: if the HSNCT were significantly shifted, they could justifiably claim that they would perform better at HSNCT than a team that qualified ahead of them by answering that extra current events tossup that would go by the wayside at the more difficult tounament.
I'll split this off to a new thread if you think it's too far afield a topic, but I disagree with this reasoning. It seems that, in reality, any team that consistently misses out on a qualification spot by a one-question swing in a single game will either qualify anyway at a different tournament*, or be accepted for a wild-card bid. Thus, I do not think the above hypothetical is a convincing argument against bifurcating the IS and HSNCT distributions.

*Since, under your scenario, this team must finish in exactly N+1th place at a tournament qualifying N teams, one loss behind the team in Nth place, and must lose one of their games by 90 or fewer points, and their opponent in that game must have gotten a current events tossup right, otherwise the scenario doesn't make sense. Also, this team must not qualify by any other tournnament, and thus not attend the HSNCT as a result. Is there any example of such a team actually existing?

Re: NAQT distributions posted

Posted: Tue Mar 10, 2009 6:18 pm
by cvdwightw
Matt Weiner wrote:
bt_green_warbler wrote:Consider here the case of a marginal HSNCT team that is strong in fine arts: if the HSNCT were significantly shifted, they could justifiably claim that they would perform better at HSNCT than a team that qualified ahead of them by answering that extra current events tossup that would go by the wayside at the more difficult tounament.
I'll split this off to a new thread if you think it's too far afield a topic, but I disagree with this reasoning. It seems that, in reality, any team that consistently misses out on a qualification spot by a one-question swing in a single game will either qualify anyway at a different tournament*, or be accepted for a wild-card bid. Thus, I do not think the above hypothetical is a convincing argument against bifurcating the IS and HSNCT distributions.

*Since, under your scenario, this team must finish in exactly N+1th place at a tournament qualifying N teams, one loss behind the team in Nth place, and must lose one of their games by 90 or fewer points, and their opponent in that game must have gotten a current events tossup right, otherwise the scenario doesn't make sense. Also, this team must not qualify by any other tournnament, and thus not attend the HSNCT as a result. Is there any example of such a team actually existing?
Matt's hypothetical scenario is not the only condition when this would hold true (consider the case of three teams in a circle of death for less than three top bracket spots; a strong fine arts team would be disadvantaged compared to a strong current events team if the tiebreaker used was points per game, or less strongly, bonus conversion). In the case of a small circuit with few qualifying tournaments and many quality teams, these scenarios could plausibly (though improbably) exist.

While this would be fairly impossible to prove without mounds of data, I would conjecture that by and large, the playoff teams at HSNCT are converting IS current events/pop culture and IS fine arts/social science at a roughly similar rate (top playoff teams may even be converting fine arts at a higher rate), while the teams finishing 3-7 at HSNCT or not making it there in the first place are converting current events at a much higher proportion than fine arts. Since the vast majority of any given regular tournament field will be teams that either don't qualify for HSNCT or aren't in any sort of playoff contention, the conversion rate will likely be low overall. It's entirely possible that this conjecture is false and actually no teams are converting current events at an equal level to fine arts, but this appears to be one case where stratifying the data (HSNCT playoff teams vs HSNCT qualifiers vs non-qualifiers) makes sense.