The Quizbowl Resource Center

NGTech wrote:

theMoMA wrote:Also, all teams in a round robin play what amounts to the same schedule, so by trying to find the "most difficult schedule" you're "solving" a problem that doesn't even exist in your own format.

While it may seem to be trying to solve a problem that doesn't exist the problem does in fact exist. The round robin format is assumed to be an even format based on the fact that every team faces every other team. The problem is that teams are made of people not computers, as such the strength of a team changes from round to round. The only ways to counteract this is to let each team face each other team multiple times, or use a mathematical method to adjust the field.

Conveniently, there were leftover packets, which would've allowed you to do this, wherein teams that had competed at a similar level during the first round-robin would play an additional game, providing more information to rank the teams. And had the best team still not been certain, you may even have had an extra packet, for them to play a third time to decide who was the winner!

Also, worth noting, your individual "scoring system" sucks. Taking, for instance, the game against empty chairs:
Team A plays the Empty Chairs. Adam answers 5 tossups, Alex answers 5 tossups, Andrew answers 4 tossups, and Abigail answers 3 tossups. The Empty Chairs do not buzz.
Team B plays the Empty Chairs. Ben, playing solo, answers 2 tossups, and then takes a nap (or decides not to buzz anymore on an incorrect belief this will help his tiebreaker score). The Empty Chairs do not buzz.
According to your individual scoring system, Adam gets .294 as does Alex, while Ben gets 1.000.

This is not a hypothetical. Take these two games that actually happened in the same round, on the same questions:
GT 390, NGU 70. GT: Gregory 6, Keith 0, Robert 0, Bradley 8; NGU: Travis 2, DoRinda 0, Matthew 2, Jason 1; Total answered 19.
Berry 180, Snead 90. Berry: Sarah 0, Greg 3, Dylan 6, Anthony 0; Snead: Mark 5; Total answered 14.

All else equal, answering more questions is better, right? Answering more questions against other players who are answering more questions is even better, right? However, this isn't borne out. Gregory (GT) scores 6/19 = .316, while Mark scores 5/14 = .357. Bradley scores 8/19 = .4211, while Dylan scores 6/14 = .4286.
Your intentions may have been grand and noble, but their result was counter-productive and bad. If you'd mentioned your ideas ahead of time, dozens of people with years of experience could have explained to you why they would produce counter-productive and bad results.

jonpin wrote: Conveniently, there were leftover packets, which would've allowed you to do this, wherein teams that had competed at a similar level during the first round-robin would play an additional game, providing more information to rank the teams. And had the best team still not been certain, you may even have had an extra packet, for them to play a third time to decide who was the winner!

As a physicist I always prefer more data, but the schedule was not the area I was in charge of. I was also selfishly wanting to leave as I was fighting off bad sinus problems and some kind of stomach bug. In my opinion a triple round robin would be the best system as it would greatly reduce any tie chances, but that would be a long tournament.

jonpin wrote: Also, worth noting, your individual "scoring system" sucks. Taking, for instance, the game against empty chairs:
Team A plays the Empty Chairs. Adam answers 5 tossups, Alex answers 5 tossups, Andrew answers 4 tossups, and Abigail answers 3 tossups. The Empty Chairs do not buzz.
Team B plays the Empty Chairs. Ben, playing solo, answers 2 tossups, and then takes a nap (or decides not to buzz anymore on an incorrect belief this will help his tiebreaker score). The Empty Chairs do not buzz.
According to your individual scoring system, Adam gets .294 as does Alex, while Ben gets 1.000.

This is not a hypothetical. Take these two games that actually happened in the same round, on the same questions:
GT 390, NGU 70. GT: Gregory 6, Keith 0, Robert 0, Bradley 8; NGU: Travis 2, DoRinda 0, Matthew 2, Jason 1; Total answered 19.
Berry 180, Snead 90. Berry: Sarah 0, Greg 3, Dylan 6, Anthony 0; Snead: Mark 5; Total answered 14.

All else equal, answering more questions is better, right? Answering more questions against other players who are answering more questions is even better, right? However, this isn't borne out. Gregory (GT) scores 6/19 = .316, while Mark scores 5/14 = .357. Bradley scores 8/19 = .4211, while Dylan scores 6/14 = .4286.

I had not noticed that problem. Thanks for pointing it out. After looking at the ranking it appears this error caused 3 places where a person with better PPG was ranked below a lower one. In retrospect I should have realized a term to correct for the total points in the round should have been used.

jonpin wrote: Your intentions may have been grand and noble, but their result was counter-productive and bad. If you'd mentioned your ideas ahead of time, dozens of people with years of experience could have explained to you why they would produce counter-productive and bad results.

While the design of my stat has been drawing condemnation here on the forums it was used in our last tournament where I received quite a few complements on designing a system that attempts to even the field.

NGTech wrote:While the design of my stat has been drawing condemnation here on the forums it was used in our last tournament where I received quite a few complements on designing a system that attempts to even the field.

What tournament was this? It seems to me that nobody who was in your current field enjoyed this statistic at all based on what they have posted here.

Jeremy Gibbs Freesy Does It wrote:

NGTech wrote:While the design of my stat has been drawing condemnation here on the forums it was used in our last tournament where I received quite a few complements on designing a system that attempts to even the field.

What tournament was this? It seems to me that nobody who was in your current field enjoyed this statistic at all based on what they have posted here.

The tournament he is referring to is probably this one: http://naqt.com/stats/tournament-teams. ... nt_id=3549

Alright, a word to the wise then. Don't assume that a high school field where literally a third of the teams are coached by noted idiot Robin Richards and the rest are of varying degrees of not being active on the national circuit is at all going to be a good way to determine what college quizbowlers want to do. A lot of high school coaches who aren't really active love really stupid ideas about quizbowl, and this seems to be no exception.

DarkMatter wrote:

Jeremy Gibbs Freesy Does It wrote:

NGTech wrote:While the design of my stat has been drawing condemnation here on the forums it was used in our last tournament where I received quite a few complements on designing a system that attempts to even the field.

What tournament was this? It seems to me that nobody who was in your current field enjoyed this statistic at all based on what they have posted here.

The tournament he is referring to is probably this one: http://naqt.com/stats/tournament-teams. ... nt_id=3549

That high school tournament couldn't have used this stat as a tiebreaker (there were no tiebreakers because it was perfectly-sorted round-robin from 8-0 to 0-8).

bt_green_warbler wrote:

DarkMatter wrote:

Jeremy Gibbs Freesy Does It wrote:

NGTech wrote:While the design of my stat has been drawing condemnation here on the forums it was used in our last tournament where I received quite a few complements on designing a system that attempts to even the field.

What tournament was this? It seems to me that nobody who was in your current field enjoyed this statistic at all based on what they have posted here.

The tournament he is referring to is probably this one: http://naqt.com/stats/tournament-teams. ... nt_id=3549

That high school tournament couldn't have used this stat as a tiebreaker (there were no tiebreakers because it was perfectly-sorted round-robin from 8-0 to 0-8).

Good point. The teams may have known about the tiebreaker in advance though, so they may have had an opinion concerning the tiebreaker even though it wasn't actually applied to the tournament. There seems to be no evidence that they have held another tournament this year, I checked both the NAQT results page and the both the high school tournaments section and the archives.

DarkMatter wrote: The teams may have known about the tiebreaker in advance though

This would surprise me somewhat.

Jeremy Gibbs Freesy Does It wrote:Alright, a word to the wise then. Don't assume that a high school field where literally a third of the teams are coached by noted idiot Robin Richards and the rest are of varying degrees of not being active on the national circuit is at all going to be a good way to determine what college quizbowlers want to do. A lot of high school coaches who aren't really active love really stupid ideas about quizbowl, and this seems to be no exception.

High school teams and coaches were not the only ones to compliment the stat, we received some from a few coaches and competitors at this most recent tournament.

I also count complaints from 4 attendees (half the field, including first place, second place, and statistically third place). Of the four remaining, one "team" was a high school playing solo. If you're trying to defend something which literally half the field protested against, you've got a problem here.

NGTech wrote: High school teams and coaches were not the only ones to compliment the stat, we received some from a few coaches and competitors at this most recent tournament.

Of the teams that actually complimented this stat, I'm curious as to how they performed in the tournament.

What is NAQT's stance on this statistic? Just curious.

This is really a response to a PM I got today (and some similar remarks already present in this thread), but I think it's worth saying publicly. I also hope it's not backseat modding, but if it is it's worth a warning/tempban. No matter who you are-- the president of a question provider, a coach, a TD, or just a quizbowl player-- if you come into a forum full of quizbowlers endorsing or defending (or whatever it's called when you're "against it being used before the tournament started") what everybody else thinks is a bad idea, your post is not exempt from criticism. No one cares how much of a say you have in your club's decision-making. When, for example, someone comes on insisting that math comp absolutely must be a part of the distribution, no one checks to see whether he's a set editor or whatever before telling him why he's wrong. When you post something on a forum, it becomes fair game for anyone to respond to as they think is appropriate.

DarkMatter wrote:What is NAQT's stance on this statistic? Just curious.

I'll direct you to NAQT's tiebreaker policy, which among other things states:

If ties must be broken (as is the case in rebracketing or determining the championship, but not necessarily for reporting overall results for lower-finishing teams), they must be played off, at the buzzer.

List of wrestling-based comic books wrote:This is really a response to a PM I got today (and some similar remarks already present in this thread), but I think it's worth saying publicly. I also hope it's not backseat modding, but if it is it's worth a warning/tempban. No matter who you are-- the president of a question provider, a coach, a TD, or just a quizbowl player-- if you come into a forum full of quizbowlers endorsing or defending (or whatever it's called when you're "against it being used before the tournament started") what everybody else thinks is a bad idea, your post is not exempt from criticism. No one cares how much of a say you have in your club's decision-making. When, for example, someone comes on insisting that math comp absolutely must be a part of the distribution, no one checks to see whether he's a set editor or whatever before telling him why he's wrong. When you post something on a forum, it becomes fair game for anyone to respond to as they think is appropriate.

Based on your links, I wonder if the person who sent your PM is the same one who sent one to me today.

If anyone has issues with PMs that have been sent, please notify me.