Byko Rankings 2007-08

[evilmonkey]

Hey Byko

The results for the Indiana NAQT state Qualifier results are up, along with various leagues across the state:http://inqblots.org/

[tomburr]

The results for the NAQT state tournament in Florida are up:

http://www.naqt.com/stats/results.jsp

[mujason]

I just posted the Tri-County Conference (MO) results; I think they're all new to the ratings except for Osage.

[Byko]

Thanks for the info--I'll be trying to catch up over the next couple of weeks. I've also gotten some results by e-mail. I've been out of town all last week and essentially didn't have internet access (hotel internet was about as good as dial-up), so I'll be playing catchup for a little while. Thanks for your patience.

[ntan]

i'm curious if HSNCT games received more weight in the rankings than did regular season games. For instance, La Jolla vaulted ahead of us after hsnct, which i found interesting, given the fact that we havent lost to them this year in the 3-4 times we played and our slightly superior record at hsnct. i suspect that their win against Stuyvesant pulled them up, but is a single win adequate grounds for increasing a team's ranking so greatly? This obviously is in no way downplaying La Jolla (they are a pretty good team), but I find there are several teams in the California region superior to them (Edison comes to mind), but this superiority isnt reflected in the rankings. That said, I acknowledge the difficulty of creating a numerically based ranking that is bulletproof when applied to teams that solely compete outside the circuits on the eastern seaboard.

[Byko]

At this point, no individual game receives any more weight than any other game. I've gotten the impression that most people prefer it that way, though I have had a couple people say playoff games should matter more. So I'm interested in the thought. But no, that wasn't an aspect of it.

One thing worth noting is that I don't have playoff results included since I don't have them, which could certainly change things a bit.

However, I would say that it's definitely affected by the fact that La Jolla's two best wins, despite being a 6-4 team, were better than Arcadia's best win. Strength of schedule is a big component, and there definitely were many matchups in which teams may have had the same record at that time in the tournament, but over the course of the year, their performace was quite different. Unfortunately, that definitely plays a part in this.

I will say, however, that all is not final--while the results were listed for Stuyvesant A, I don't know how similar this team was to Stuyvesant's actual A team. If it is in fact closer to Stuyvesant B, that would definitely bring La Jolla's rating down since that would be a much weaker opponent (#325 nationally).

[mknupp]

Will you be adding the NAQT HSNCT small school playoffs to the standings?

[Byko]

Will you be adding the NAQT HSNCT small school playoffs to the standings?

Yes...once NAQT posts the scores.

[Howard]

It seems that approximately the bottom 1/4 of the schools are now missing. Is this intentional?

[Byko]

It seems that approximately the bottom 1/4 of the schools are now missing. Is this intentional?

No. The last time the web server actually got as far as making it through a recalculation without losing connection to the database server was about two days ago, and it lost connection about 3/4 of the way through writing the final numbers to the database. And that was two days ago, and it hasn't been able to perform a successful recalculation since then, and I have it set to try hourly.

So, it's a server issue. I have all the individual game data, but the way this is going, I may be accelerating a move to a different server a little faster than planned. I was looking to move it during the offseason, but I wonder if I might need to try to do so next week (with NSC preparation and my job, it won't be happening this week).

[catsasslippers]

Stuy was 3/4ths Stuy B plus Aidan. I'd say that should be considered Stuy B.

[Byko]

Stuy was 3/4ths Stuy B plus Aidan. I'd say that should be considered Stuy B.

Thanks for the info. I've made the change, though I'm not sure the server will ever recalculate the ratings, which would really be a shame.

[harpersferry]

Auburn A (Rockford IL) somehow has 8 games recorded on saturday NSC, and one of them as far as I know (Robinson) never existed.

[Byko]

Auburn A (Rockford IL) somehow has 8 games recorded on saturday NSC, and one of them as far as I know (Robinson) never existed.

One result was entered for the wrong Auburn. Thanks for catching it, and my apologies for the error.

[Down and out in Quintana Roo]

Are the rankings going to be updated with the NAC results as well? I guess that would end the year so we can have a true final rating.

[AndyShootsAndyScores]

Are the rankings going to be updated with the NAC results as well? I guess that would actually be pointless and would skew the ratings by adding matches that weren't quiz bowl.

Fixed.

[Down and out in Quintana Roo]

So then they should take off all the matches that are of that format from the current rankings? That would eliminate another 7 wins by Charter, i guess, and 4 by us, among others.

[quizbowllee]

So then they should take off all the matches that are of that format from the current rankings? That would eliminate another 7 wins by Charter, i guess, and 4 by us, among others.

I think (I could be wrong) that Andy has the NAC confused with the PAC.

[dtaylor4]

I think (I could be wrong) that Andy has the NAC confused with the PAC.

No, Andy's right on the money.

[sabine01]

You might want to look at this on Byko's ratings site:

What types of quiz bowl games are not included?

Any of the following:

* Games involving three or more teams competing against each other simultaneously (e.g., Panasonic Academic Challenge, It's Academic)*
* Games that are forfeited
* Games against all-star or chimeric teams
* Games against college teams
* Games in a single-subject tournament (e.g., Science Bowl, trash tournaments)
* Byes

* While it is true that these games could be set up as a series of pairwise comparisons, there are many circumstances in which winning is all that matters, so the second place team may not actually be better than the third or fourth place team. Thus, it is easier just to ignore these results.

Much as many of us are not fans of Chip, his tournaments don't fall into the guidelines stated by Byko on the ratings site. If there's a weighting, given it is Chip-tacular or Byko has stated a different practice somewhere, feel free to direct my attention to it.

[Matt Weiner]

Well yeah, they're included now, but ideally they shouldn't be. The ratings should just reflect games played by the regular lineup of a team on real questions, since those are the only games that mean anything.

[sabine01]

Well yeah, they're included now, but ideally they shouldn't be. The ratings should just reflect games played by the regular lineup of a team on real questions, since those are the only games that mean anything.

Not arguing that at all; Just stating current practice.

[evilmonkey]

Are the rankings going to be updated with the NAC results as well? I guess that would actually be pointless and would skew the ratings by adding matches that weren't quiz bowl.

Fixed.

Well, you've got a problem then. You've already added several dozen tournaments, leagues, etc. that use QU or the equivalent - why then shouldn't the NAC be added in.

OTOH, if you decide that QU tournaments shouldn't count, where do you stop? Because Illinois questions are math-heavy, do you discount those? Since NAQT has more trash than would be preferred in an Academic competion, do you discount those? More concerning, what about other tournaments whose questions fill in the shades of grey between Good and Bad quizbowl?

EDIT: DISCLAIMER: NOT A DEFENSE OF NAC. DON'T TAKE IT THAT WAY.

[AndyShootsAndyScores]

Had I known that there were already results from QU tournaments in the rankings, I would not have made my post. NAC should definitely be included then.

I have no big problem with any other formats. I wouldn't argue against putting PAC results in there if it were possible. Whether a format is math-heavy or is just plain weird as far as structure goes doesn't concern me. QU has terrible questions and had terrible structure, that's my beef with it.

[Down and out in Quintana Roo]

So then they should take off all the matches that are of that format from the current rankings? That would eliminate another 7 wins by Charter, i guess, and 4 by us, among others.

I think (I could be wrong) that Andy has the NAC confused with the PAC.

Oh, i never commented on this, but i do prefer to be called Andrew. Thanks.

Anyway, the ratings WERE updated with the NAC results and Charter stayed in 5th place overall. Of interest (of course you can see the ratings here) the Manheim win over Charter was the 5th least likely result of the year, apparently, with only a 0.020 likelihood of happening.

And hoorah... the CR C team did end up solidifying our last place in the nation status, albeit the margin got a little smaller. Well, needless to say for them, there's no place to go but up next year!

[evilmonkey]

I think (I could be wrong) that Andy has the NAC confused with the PAC.

Oh, i never commented on this, but i do prefer to be called Andrew. Thanks.

Hehe... the student who you had originally responded to is Lee's student named Andy...

EDIT: Fixed Quote

[Ondes Martenot]

And Booker T Washington is now ranked 24th nation and received 0 votes in the post nationals poll

[Whiter Hydra]

And TJ A is ranked 4th despite getting the most number of first-place votes.

[AndyShootsAndyScores]

I think (I could be wrong) that Andy has the NAC confused with the PAC.

No, Andy's right on the money.

I actually knew what I was talking about this time .