Page 3 of 3

Re: Chicago Open 2016 Set Discussion

Posted: Mon Aug 01, 2016 4:47 pm
by felgon123
Amizda Calyx wrote:I'm a little disappointed that several clues were added to the Trypanosoma tossup that I had specifically avoided because they were non-unique -- there seems to have been a conflation between kinetoplastids or trypanosomatids and Trypanosoma, despite important other kinetoplastid genuses existing such as Leishmania, which also has mini/maxicircles and also uses polycistronic transcription followed by trans-splicing.
Yes, I was powerfully annoyed by this question, because I had an early buzz on a non-unique clue you've pointed out and said "kinetoplastids," the most correct answer at that point because that's the most general group to which the clue actually applies. I was ruled wrong because the question has said "this genus"--a wholly unhelpful qualifier because, again, the clue was not unique. I'm calling attention to it because this is not the first time this situation has occurred. In the finals of the 2015 ICT, there was a tossup on octopuses which had a clue that is not unique to octopuses and applies to other cephalopods. I buzzed early and said "cephalopods," and was ruled wrong because the tossup had said "this order." Inserting "this [taxonomic rank]" phrases into tossups with non-unique clues does not somehow make those clues unique; it just punishes people on a technicality when they buzz in and give the best answer for your non-unique clues. This is a lazy and bad writing tendency that biology writers ought to avoid.

Re: Chicago Open 2016 Set Discussion

Posted: Mon Aug 01, 2016 5:53 pm
by adamsil
felgon123 wrote:
Amizda Calyx wrote:I'm a little disappointed that several clues were added to the Trypanosoma tossup that I had specifically avoided because they were non-unique -- there seems to have been a conflation between kinetoplastids or trypanosomatids and Trypanosoma, despite important other kinetoplastid genuses existing such as Leishmania, which also has mini/maxicircles and also uses polycistronic transcription followed by trans-splicing.
Yes, I was powerfully annoyed by this question, because I had an early buzz on a non-unique clue you've pointed out and said "kinetoplastids," the most correct answer at that point because that's the most general group to which the clue actually applies. I was ruled wrong because the question has said "this genus"--a wholly unhelpful qualifier because, again, the clue was not unique. I'm calling attention to it because this is not the first time this situation has occurred. In the finals of the 2015 ICT, there was a tossup on octopuses which had a clue that is not unique to octopuses and applies to other cephalopods. I buzzed early and said "cephalopods," and was ruled wrong because the tossup had said "this order." Inserting "this [taxonomic rank]" phrases into tossups with non-unique clues does not somehow make those clues unique; it just punishes people on a technicality when they buzz in and give the best answer for your non-unique clues. This is a lazy and bad writing tendency that biology writers ought to avoid.
Sorry about this, Tommy and Joelle. Looking back, there are a couple issues with this question--originally, I had the sentence planned as something like "transcription from the VGS promoter in this genus is polycistronic and followed by trans splicing", but I split it into two sentences to make it clear that it's not just from that locus, then also butchered the second clue, because it's clearly supposed to be the VSG promoter. Yikes. For what it's worth, Tommy, I agree with you--tossups on taxa only really work if you're cluing organisms that are specifically known by their scientific names (like Pichia, or Clostridium, or Salmonella, or whatever), and if you'd protested that buzz, it'd probably have gone your way. Very sorry to everyone for all my dumb screw-ups.

For what it's worth now, the tripartite attachment complex clue was thrown into the end of the sentence about mtDNA to specifically point to trypanosomes. I did do enough reverse clue lookup to realize that minicircles and maxicircles are not unique, even to kinetoplastids, but I think the TAC is only in trypanosomes. The sentence in question is

Organisms of this genus have mtDNA packaged into minicircles and maxicircles and anchored by the tripartite attachment complex.

If you can find a source saying that's in Leishmania too, then double mea culpa, but I think that sentence by itself is okay.

Re: Chicago Open 2016 Set Discussion

Posted: Mon Aug 01, 2016 9:26 pm
by Periplus of the Erythraean Sea
I think it may be useful to examine teams which had a similar composition year-to-year:

Andrew Hart/Rob Carson team: 16.29 (2015) vs 16.23 (2016) - Basically the same. The difference here is that in 2015, Andrew and Rob played with Shan and Adam, whereas they played with Tejas and Billy Busse this year; I have no clue how to compare these players, Shan is very good at high-level tournaments and Adam's a good player (especially on science), but Tejas also scales well and Billy's one of the best high-level science specialists there's been. I'm not really sure if Rob and Andrew got worse or better, but given their substantial editing work on collegiate tournaments, I doubt they got that much worse even if they did get worse. In 2016, this team received an outsized low number of History bonuses (26, compared to the high 30s for each Literature and Science); they were substantially better than average in the latter two categories, and slightly above average in the former.

(additional conclusion: Billy still really knows his stuff, despite mainly editing and not being an active collegiate player for two years - 17.10 was the second-highest Science bonus conversion in the field. This suggests that Rob and Andrew, who also have edited a lot but not played a ton, probably haven't really declined drastically since last year either.)

Will Alston/Joey Goldman/Richard Yu team: 16.13 (2015) vs 16.50 (2016) - This team's players each improved throughout the 2016 year; the team also added a stronger fourth player, who scored more points per game than last year's fourth player despite ending up in the top bracket. Idiosyncratic bonus numbers don't seem to have affected our numbers a ton this year. (we were pretty even in strength across the large categories)

Charles Hang/Seth Ebner/Will Mason team: 10.70 (2015) vs 13.06 (2016) - This team added one player and improved by 2.3 points per bonus; its players also all definitely improved during the year to some extent (I suspect Charles improved substantially, given his performances at tournaments like ICT and MYSTERIUM). I will note, though, that if you look at the conversion statistics, this team heard an outsize number of History bonuses, on which it averaged 14.46 points per bonus, as compared to Literature ones - a good bit higher than the field average on history and higher than its average conversion in any of the other large categories. They did also hear an equal number of science bonuses as history ones, which they did do worse than the field on, but this team also had a lower PPB than the field as a whole.

Athena Kern/Foster Hughes team: 9.52 (2015) vs 11.01 (2016) - This team traded Jason Zhou and Alston Boyd (who, in 2015, were a B-team freshman and a high schooler - nonetheless, they were the team's highest scorers) for Daniel Hothem and Ophir Lifshitz, veterans who were nonetheless outscored by Athena (who clearly improved significantly, based on comparing her CO PPG numbers as well as her in-season performances).

I don't know if these comparisons really mean anything, but they're out there.

Re: Chicago Open 2016 Set Discussion

Posted: Fri Aug 05, 2016 10:16 am
by DumbJaques
I'm dreadfully late to the party on this (the price of moving the day after CO), but I wanted to weigh in on a few points:

First, thanks to the editors (John, Mike, Matt, Mike, Aaron, and others) for their hard work on this set. Let's not forget, as we traverse statistical sorcery and the far reaches of MathTossupHellscape, that as recently as a few months ago this tournament was in danger of literally not happening. I, and I think all of us, are deeply appreciative that it did.

Second and related: Yeah, this tournament was too hard. There are statistical reasons to think so (though I don't think they're quite the ones identified), but I'm actually speaking much more of things you can't get from (admittedly super cool!) conversion stats.
For instance, when you toss up "The Great Mississippi Flood," and basically nobody answers it until the end, you have in fact basically written a one line tossup that says "say a superlative word (or 1928-1), then the big river in the southern US, then the word 'flood', which I have already given you." Of course, some of us manage to even fuck that up, but in general this isn't "hard" in a statistical sense. But in every experienced, felt way, it's brutally hard and then aggravating to boot. Some questions like this are inevitable in any hard set, but it seemed like there were a lot of them at this year's CO.

But I want to redirect the focus of this away from the editors a bit. In particular, the team was very clear that submissions would matter more this year than they necessarily have in the past - John was not, as he said, chief editing, which is generally the best oversight for difficulty issues. Mike Cheyne literally warned us not to write ass-hard things, because we would get the tournament we submitted.

Well, we did. I mean, we almost certainly got something far LESS brutal than the tournament we submitted, because A) that's almost always true and B) some of you fucking monsters showed me your submissions and Jesus Christ, what's wrong with you?
Look, we can't just whine at the editors about this - we need to stop writing packets differently than we write our own tournaments. I'll admit to being guilty of this in the past (though not as guilty as CERTAIN TEAMS around the 2012-2014 period, in which submissions were likely to be about as useful as the packet from South Harmon Institute of Technology B). And sometimes it's going to happen - schedules are full, you're up against a financial deadline, etc. etc.
But sets like CO, which only get main site packets and require huge amounts of work, really need our A game. And we need to stop acting like our A game is to breakout that Fuck You tossup we've had in our pockets all year. Like, dude. There are SO MANY opportunities for that! There's vanity sets, there's the IRC, there's even Tricon (somehow, still, against the laws of man and nature). And I'm not even saying don't have a handful of crazy exciting questions in your CO packet - just treat it like you would a round from a tournament you produced: Mix in 5-10% crazy with 90-95% not crazy. Hell, if we kicked it up to 15 or even 20 (NOTE: Not advisable), we'd still be in much better shape than we typically are.

And yeah, that can be hard with group submissions, but if everyone keeps that ratio the packet should be fine. To be REALLY CRAZY, you could even collaborate with your teammates, allowing you to offer such invaluable feedback as, "hey, maybe we shouldn't toss up the third most important character from Villiers de l'Isle-Adam's Axel?", or "wow, I didn't realize there was so much social history about the Chita Republic!"

In summation: This set was too hard, so let's internalize that rather than project it onto the editors, and make some adjustments.

Re: Chicago Open 2016 Set Discussion

Posted: Fri Aug 05, 2016 10:50 am
by Cheynem
While I appreciate Chris' "defense," I tried to be fairly proactive this time around and certainly didn't let in any questions just because I had no other options. At least regarding tossups, I thought all of the ones in the set were appropriate--some clearly weren't, some were. For the most part, I actually thought the submissions were pretty good in terms of history, although, yes, as Chris points out, there were a few packets that just decided to tossup whatever and that's not cool.

Re: Chicago Open 2016 Set Discussion

Posted: Fri Aug 05, 2016 11:04 am
by The King's Flight to the Scots
DumbJaques wrote:
Well, we did. I mean, we almost certainly got something far LESS brutal than the tournament we submitted, because A) that's almost always true and B) some of you fucking monsters showed me your submissions and Jesus Christ, what's wrong with you?
Look, we can't just whine at the editors about this - we need to stop writing packets differently than we write our own tournaments. I'll admit to being guilty of this in the past (though not as guilty as CERTAIN TEAMS around the 2012-2014 period, in which submissions were likely to be about as useful as the packet from South Harmon Institute of Technology B). And sometimes it's going to happen - schedules are full, you're up against a financial deadline, etc. etc.
Yeah I'll cop to being guilty of this for this tournament (and may even be one of the unnamed teams from the 2012-2014 period! IT IS A MYSTERY). I actually think that the feeling of oppressive difficulty at CO had much more to do with the gratuitous leadin-stacking than anything in the answers. But we submitted a lot of wacko stuff, and the editors did their job by cutting four or five deranged questions, so the responsibility falls on us here, too.

Part of the issue here is that there's no real incentive for packet writers to do a good job. I don't think "best packet discounts" would really work for this, but maybe some amount of transparency on submitted questions would help. If the editors were to keep track of which packets needed minimal editing and release that info, like Matt did for Auroni's packet, say. The obvious problem here is that you don't want to shame new or inexperienced or casual players by exposing how much of their packet got gutted. If there's a tactful way to reveal those stats only for the teams with enough writing experience to be held accountable, I think it would encourage submitting teams to put in the effort.

Re: Chicago Open 2016 Set Discussion

Posted: Fri Aug 05, 2016 11:24 am
by The King's Flight to the Scots
One question I'd ask about submitting packets: do more experienced editors on a submitting team have a responsibility to supervise or edit the submissions of less-experienced teammates? I think that really would lessen the burden on set editors, but it might be undue for the teams sending submissions in. I'm interested to hear what other teams generally do.

Re: Chicago Open 2016 Set Discussion

Posted: Fri Aug 05, 2016 12:12 pm
by Mike Bentley
DumbJaques wrote:For instance, when you toss up "The Great Mississippi Flood," and basically nobody answers it until the end, you have in fact basically written a one line tossup that says "say a superlative word (or 1928-1), then the big river in the southern US, then the word 'flood', which I have already given you." Of course, some of us manage to even fuck that up, but in general this isn't "hard" in a statistical sense. But in every experienced, felt way, it's brutally hard and then aggravating to boot. Some questions like this are inevitable in any hard set, but it seemed like there were a lot of them at this year's CO.
Sorry to hear this question was apparently too difficult for the field. I've come across this several times in the past couple of years and as best I can tell it's an extremely important event in African American history in the South, and at least a somewhat important event in the career of Herbert Hoover.

But with a lot of these more out there answers it can be hard tell what people know.

Re: Chicago Open 2016 Set Discussion

Posted: Fri Aug 05, 2016 1:45 pm
by Habitat_Against_Humanity
Mike Bentley wrote:
DumbJaques wrote:For instance, when you toss up "The Great Mississippi Flood," and basically nobody answers it until the end, you have in fact basically written a one line tossup that says "say a superlative word (or 1928-1), then the big river in the southern US, then the word 'flood', which I have already given you." Of course, some of us manage to even fuck that up, but in general this isn't "hard" in a statistical sense. But in every experienced, felt way, it's brutally hard and then aggravating to boot. Some questions like this are inevitable in any hard set, but it seemed like there were a lot of them at this year's CO.
Sorry to hear this question was apparently too difficult for the field. I've come across this several times in the past couple of years and as best I can tell it's an extremely important event in African American history in the South, and at least a somewhat important event in the career of Herbert Hoover.

But with a lot of these more out there answers it can be hard tell what people know.

...and there's even a Randy Newman song about it! For what little my opinion is worth in this discussion, I would think that this tossup is on a topic both "real" and "important".

Re: Chicago Open 2016 Set Discussion

Posted: Fri Aug 05, 2016 3:37 pm
by Cheynem
I should probably have recommended to Mike that the question be on "the flooding of the Mississippi River" or something more general, with basically the same clues. I think it's tough to remember the specific year. That said, I also agree it's an important topic and Mike picked good clues for it.

Re: Chicago Open 2016 Set Discussion

Posted: Sat Aug 13, 2016 5:57 pm
by Red Panda Cub
I am very late to the game here, but I just wanted to thank John for picking super excellent hard parts for the philosophy bonuses. Things like "always already", "state of exception" and "virtue epistemology" show a commendable effort in getting the lay of the land on a subject and picking the sort of thing that rewards understanding and knowledge of an often neglected sort. Thumbs up.