The Quizbowl Resource Center

RyuAqua wrote:Also out of curiosity, does anyone know/remember what the Byko formula was, and why ti might go wrong, doing something such as putting a one-tournament winner from New Hampshire on top? I remember hearing it was weighted somehow based on the previous rankings of teams, meaning beating the number 2 team moved you much further up than beating the number 200 team.

It used a system called KRACH, which essentially recursively calculated a ranking.

Edited to fix the link --mgmt.

The real crippling flaw of the Bykowski ranking was really the fact that he included results from every bad tournament he could get his hands on. If it were brought back with only tournaments being run on A-sets or better being included, I would have really enjoy having that to rank teams too, sort of a BCS computer ranking to Fred's AP poll.

Jeremy Gibbs Freesy Does It wrote:The real crippling flaw of the Bykowski ranking was really the fact that he included results from every bad tournament he could get his hands on. If it were brought back with only tournaments being run on A-sets or better being included, I would have really enjoy having that to rank teams too, sort of a BCS computer ranking to Fred's AP poll.

There should totally be a coaches poll.

Then we should have a poll that's the average of all the other polls. That would give us the best rankings.

Well, the Dec. 4th Houston Hoedown at St. John's is now in the record books -- and Cistercian Prep had some nice results.
Our JV "A" team took the # 1 spot at the end of the day in that division. For a group of 4 freshmen who've never played together as an ensemble, I was very impressed with how they meshed, fought back, and ultimately, won the day!

Our Varsity "A" had a nice run through the entire prelim rounds going 6 - 0 along with Seven Lakes.
In single elimination they continued doing nicely. Only the finals presented
a bit of challenge. In a LASA 'A' vs. CPS 'A", LASA came up the better.
LASA played a great, great game and deserved the # 1 slot at the end of the day.

When loooking back over the entire day, the Cistercian varsity 'A' did a nice job
all the way through -- both LASA and CPS finished the day with with 9-1 records,
if memory serves me correctly.

Hopefully, the upcoming national rankings will make my heart go pitter-patter with
some improvement in our standing, but at the end of the day, the game is still Atlanta/Nationals.

PS: Our Varsity "B" team team did very commendable work, too, running the prelims with 5-1 records;
however, the seedings of the gaggle of 5-1 teams didn't give us any leeway, so they exited sooner than we
would have hoped. But, they are a fantastically gifted group of Sophomores (and 1 junior)!

Many thanks to Chris Romero, grandmaster of NAQT-TEXAS and Ted Curry, our host at St. John's School/Houston.

Remember-the-Alamo-Remember-Goliad wrote:Well, the Dec. 4th Houston Hoedown at St. John's is now in the record books -- and Cistercian Prep had some nice results.

Stats for the aforementioned tournament can be found here: http://www.texasquizbowl.org/d/10hohoho ... dings.html

Chandragupta Maurya wrote:

Remember-the-Alamo-Remember-Goliad wrote:Well, the Dec. 4th Houston Hoedown at St. John's is now in the record books -- and Cistercian Prep had some nice results.

Stats for the aforementioned tournament can be found here: http://www.texasquizbowl.org/d/10hohoho ... dings.html

What set was this tournament run on?

As stated (twice!) in the other thread, IS98.

I know someone mentioned a tournament at St. John's earlier. Does anyone know what set it used?

There could be a poll like the preseason one mentioned in another thread that was made up of ballots from people who submit them - that would be interesting

Chandragupta Maurya wrote:There could be a poll like the preseason one mentioned in another thread that was made up of ballots from people who submit them - that would be interesting

As in a midseason poll?

SoLegit12 wrote:

Chandragupta Maurya wrote:There could be a poll like the preseason one mentioned in another thread that was made up of ballots from people who submit them - that would be interesting

As in a midseason poll?

Sort of, but in December, not May

Last year's midseason poll was published at the end of January.

We should probably start the mid-season poll after this week since it seems there won't be many big tournaments on the 18th and definitely none on the 25th.

Arsonists Get All the Girls wrote:It seems there won't be many big tournaments on the 18th...

Don't let Jonah hear you say this, or he'll need to remind you that ADVANTAGE is being represented by FIVE STATES!!!

The mid-season poll will not be happening until after the tournaments of December.

RyuAqua wrote:Dwight: with people such as Coach C and myself looking back to they days of an objective (if flawed, as Byko's was) ranking model for teams, is there any way one could engineer (and run) a new nationwide set of team rankings from the statistical stuff you've been working on? Or would that require a completely different approach, since this model only works for an individual question set?

So as I understand it, Byko's model (and the KRACH model it derives from) and NAQT's model (and the ELO model it derives from) both use only W-L record and compute rankings recursively that way. What this model does is use actual question-conversion statistics to generate a theoretically-more-real model of how teams that don't play each other might fare in a real game.

We can use a modified KRACH rating. Let's start with every team having a value of 1000, because that's a nice number. If we input the "expected long run" W-L matrix into the KRACH ratings, we should get new ratings. (If we want, we can then run the KRACH ratings for each individual tournament on that packet set, using actual W-L data.) Use these ratings as the initial conditions to the KRACH rating of the next packet set, with "expected long run" W-L matrix on a new set (let any teams with no data have a value of 1000).

We can do the exact same thing for the ELO ratings - start every team with a value of 1000 (or 1200 as in NAQT's model), input the "expected long run" W-L matrix into the ELO ratings, compute new ELO ratings (and then modify with actual tournament W-L data, if preferred), input another tournament's "expected long run" W-L matrix (and actual W-L data, if preferred) into the ELO ratings. Actually, with ELO, we can also use the "Monte Carlo simulation" win percentages as well, which is likely going to give more accurate results at the cost of much more computation. The other thing that the ELO ratings allow us to do is assign different K-values (a modifier that says how much a team's ranking increases/decreases) to different tournaments - for instance, "regular" sets could have K-values of 32, while "novice" sets could have K-values of 16.

While I'm at it, one thing I'd love to see is NAQT using the "expected long run" model (or some variant of the "Monte Carlo" model) on the HSNCT prelim statistics to seed teams for the HSNCT playoffs, rather than their current use of PP20H that does not take schedule strength into account. I don't know exactly how feasible this is, though.

So, it's not an update to the rankings, but I just made a post about how I try to evaluate performances on other tournaments. You can find it here.

Just dropping in to let the world of quizbowl know that Ladue is the real thing and should be around the top 20, and lots of other teams from Missouri are suddenly quite strong and worth worrying about if you find your team opposing them in a game. This is quite an exciting development in our traditionally weak state.

Barring the discovery of recently-unpublished statistics, I now have all known data from IS-96 processed (with the usual caveats - powers, 10s, negs, bonus points per bonus controlled all kept, nothing wacky like worksheets). Attached find the (hopefully final) IS-96 "expected long run" stats. I also fixed the error in the Monte Carlo simulation score output, but the Excel file is too big to upload.

Discuss, debate, whatever it is you people do with these rankings. I just make 'em.

HSQB Rankings wrote:90. Centennial (GA)

I'm guessing this is going to change after Centennial's performance at Chattahoochee's BATE mirror today.

AlphaQuizBowler wrote:

HSQB Rankings wrote:90. Centennial (GA)

I'm guessing this is going to change after Centennial's performance at Chattahoochee's BATE mirror today.

Given the early stats, I certainly cannot disagree.

Centennial actually beat Dorman 560-110 in the final round at Chattahoochee today. They have been seriously underestimated so far this year.

EDIT: added actual score of Centennial-Dorman final.

Mmmmmmmmmmmm rankings.

Previously, you had mentioned that, starting with these rankings, you would place greater weight on head-to-head results instead of relying almost exclusively on ppb as you did previously. Did you do this? If so, would you mind explaining your methodology further? When you're feeling more motivated, at least.

Haven't had time, though I doubt anything significant would be changed.

If I'm reading it right, it looks like you said Northmont was only one player strong at ADVANTAGE. If so, that is not accurate. They definitely had a full squad (I don't know if they were missing anyone). It was a different Ohio team, "Hamilton", that was one guy.

Dominator wrote:If I'm reading it right, it looks like you said Northmont was only one player strong at ADVANTAGE. If so, that is not accurate. They definitely had a full squad (I don't know if they were missing anyone). It was a different Ohio team, "Hamilton", that was one guy.

Yeah, that was Joe Nutter from Walnut Hills, if I recall correctly.

Dominator wrote:If I'm reading it right, it looks like you said Northmont was only one player strong at ADVANTAGE. If so, that is not accurate. They definitely had a full squad (I don't know if they were missing anyone). It was a different Ohio team, "Hamilton", that was one guy.

It might also be a typo; I thought I overheard something or other about Northmont being short one player on Saturday. Then again, I don't really know.

Here's the flaw I see with not taking into account actual games: Dorman moved up a spot in these rankings, from 6 to 5, after losing to both TJ B and Centennial. Obviously I think Dorman's still a top 10 team, but that just seems ridiculous. Also, I would have put Centennial ahead of us, if not also ahead of Dorman, after Centennial's performance at CATT.

State College A lost to LASA A 385-310 at Harvard Fall
LASA A lost to Hunter A 415-250 at Harvard's Delta Burke
Hunter A lost to Dorman A 350-325 at Harvard Fall
Dorman A lost to Thomas Jefferson B 375-290 at GSAC
Thomas Jefferson B lost to Detroit Catholic Central B 380-200 at Fright
Detroit Catholic Central B lost to Novi B 275-240 at the University of Michigan Autumn Classic
Novi B lost to Ann Arbor Huron 210-170 at the University of Michigan Autumn Classic
Ann Arbor Huron lost to Thurston A 190-155 at the University of Michigan Autumn Classic
Thurston A forfeited to Chelsea at the University of Michigan Autumn Classic
Chelsea forfeited to Jackson at the University of Michigan Autumn Classic

I would like to congratulate Jackson High School on being the best high school quiz bowl team in America. I anticipate that both the NSC and the HSNCT will be canceled this year, as no team stands a chance of making up the 640 point per game difference between the Jackson squad and the second best team in quiz bowl, State College.

styxman wrote:

Dominator wrote:If I'm reading it right, it looks like you said Northmont was only one player strong at ADVANTAGE. If so, that is not accurate. They definitely had a full squad (I don't know if they were missing anyone). It was a different Ohio team, "Hamilton", that was one guy.

It might also be a typo; I thought I overheard something or other about Northmont being short one player on Saturday. Then again, I don't really know.

From what I know of the Northmont team, they were full strength at ADVANTAGE, and were missing one player at the Louisville tournament, playing Sam Blizzard (an 8th grader, albeit an incredibly talented 8th grader) in lieu of Michael Czupryn, their 2nd best player.

Inkana7 wrote:

styxman wrote:

Dominator wrote:If I'm reading it right, it looks like you said Northmont was only one player strong at ADVANTAGE. If so, that is not accurate. They definitely had a full squad (I don't know if they were missing anyone). It was a different Ohio team, "Hamilton", that was one guy.

It might also be a typo; I thought I overheard something or other about Northmont being short one player on Saturday. Then again, I don't really know.

From what I know of the Northmont team, they were full strength at ADVANTAGE, and were missing one player at the Louisville tournament, playing Sam Blizzard (an 8th grader, albeit an incredibly talented 8th grader) in lieu of Michael Czupryn, their 2nd best player.

Jarret is correct. As is Nikhil.

AlphaQuizBowler wrote:Here's the flaw I see with not taking into account actual games: Dorman moved up a spot in these rankings, from 6 to 5, after losing to both TJ B and Centennial. Obviously I think Dorman's still a top 10 team, but that just seems ridiculous. Also, I would have put Centennial ahead of us, if not also ahead of Dorman, after Centennial's performance at CATT.

Dorman dropped from 4 to 5.

A 6 where there was supposed to be a 4? Proof that PPB can't adequately rank teams! Vindication at last!

A PRESENT?! FOR MEEEEEEEEE?!

Attached is my collection of stats that I've been using so far; I haven't touched them in about a month. I hope this is helpful to other people, and possibly to myself if people note mistakes.

Edit: Actually, I think that's an old file. Hold on.

Editx2: Not an old file. Reposting.

cvdwightw wrote:.. Gimranov-formatted stats are harder in their own way - they don't require a manual search, but rather lots of clicking back and forth between games. In short, while it probably takes me 30-60 seconds per team to mine data from NAQT's website, it will take me a lot longer to do it with other stats formats.

This is now fixed in QBSQL-- results on gimranov.com are now more detailed and better-organized. Other installations of QBSQL will change when their administrators update to the most recent version.

I've got a new set of rankings up.

I'm playing around with some polls over on the blog. Give me feedback on if you think it's interesting/worthwhile/whatever.

St. Anselm's is way underrated again, Aidan in a really good player. Are you still not looking at head-to-head?

I'm ridiculously impressed by your efforts, Fred

Wurzel-Flummery wrote:St. Anselm's is way underrated again, Aidan in a really good player. Are you still not looking at head-to-head?

Guess I'm just underrating half of the teams in the top 50.

I've already made my statements about the importance of head-to-head; unfortunately, at this point in the season and the year, I don't have time to go back through every single tournament and track each team's face-offs with each of the 200 or so teams I have under consideration for the rankings.

TheKingInYellow wrote:I'm ridiculously impressed by your efforts, Fred

Thanks.

Somewhat noteworthy: the site for this had 660 hits yesterday, which is the most ever.

Wurzel-Flummery wrote:St. Anselm's is way underrated again, Aidan in a really good player.

As Fred alluded to, I bet you will be shocked when you realize just how much better lots of other regions of the country are this year outside of the mid-Atlantic. Maybe his ranking of St. Anselm's is a little off, but given the results that have been coming in this year it's certainly not an unreasonable ranking. I suspect a lot of people are going to succumb to the trap of assuming that a team that plays really well when you see them is automatically obviously top 15, 10, whatever, when really, there are tons of teams in other regions doing the same thing for their local observers.

Has State College won a high school quizbowl tournament this year?

Nick wrote:Has State College won a high school quizbowl tournament this year?

No, the only HS tournament State College A has gone to was HFT, which they lost to LASA.

I'd love to see some more responses on these polls.

I have an idea that may be explored once the PACE format is announced.

Mr. Morlan, not that it's important, but I thought I should let you know that at our mirror of the Prison Bowl, Alpharetta was yet again without William, their best player. Just wanted to point this out in case you don't know of this when you put up a new set of rankings.

commentary please

what