Byko's Ratings

Captain Sinico · Post by **Captain Sinico** » Sun Oct 14, 2007 4:43 pm

It seems to me that the easiest way to handle this for ranking purposes is to just label whichever team from a given school finishes highest (regardless of its in-tournament designation) as the A team. That is a posteriori the best team the school fielded that day, which, I think, is what we're really trying to get at. So, for ranking purposes, it doesn't matter very much what labels a coach chooses in the aggregate.

Later,
MaS

Tower Monarch · Post by **Tower Monarch** » Sun Oct 14, 2007 6:32 pm

A situation very similar to this last scenario occured yesterday at TJ. I'll refer anyone interested to read Swissboy's response to my disappointment with their roster in the TJIAT forum, but I hope if this tournament is added to the rankings, I definitely support Immaculate Deception's plan to account for it.

The Atom Strikes! · Post by **The Atom Strikes!** » Sun Oct 14, 2007 6:42 pm

I think that in cases where team selectors' picks are uncertain, Simone's idea would be best.

btressler · Post by **btressler** » Tue Oct 16, 2007 7:38 pm

I think the easiest policy here is for Dave to accept the tournament as posted, and for you to plead your case privately if you think it should show up in his system differently.

I actually have already done this for JIAT, telling Dave that the "F" team will now be playing as team "B" until further notice.

It is very possible that the 'old' team B might beat the 'new' team B in the future, but I won't beg Dave to retroactively change the letters in that case.

Due to Kellenberg's and Seton Hall's travels and the plethora of teams going this week to Princeton that were at JIAT, I'd say the mid-Atlantic should be somewhat connected very soon.

The Atom Strikes! · Post by **The Atom Strikes!** » Tue Oct 16, 2007 8:14 pm

But are TJIAT's playoff results out yet? It would be kind of unrepresentative to put in the tournament without them.

Nine-Tenths Ideas · Post by **Nine-Tenths Ideas** » Tue Oct 16, 2007 8:21 pm

Just a question- how often are these updated?
Isaac

Byko · Post by **Byko** » Thu Oct 18, 2007 8:39 am

NeverHitTina wrote:Just a question- how often are these updated?
Isaac

Basically, it happens when I have time. Right now, I'm still finishing up writing/editing one final tournament set that needs to get out by this weekend. So I'm hoping to add some more data sometime next week.

Sen. Estes Kefauver (D-TN) · Fri Oct 19, 2007 1:34 am

Another question -
What does this mean for regions that feature a large number of junior varsity tournaments? Often times the JV teams are simply the B or C team of a school, but they compete in separate divisions with similar teams. How are these results treated? What happens when players move back and forth on these teams? For instance, my 1st year I played a couple JV tournaments with Christine P. since we were both new, and soon after we moved up to the varsity team. However, our varsity didn't play anything all year until we moved up for a number of reasons. So, while it's irrelevant to this year, would our JV team have counted as NKC A until we left it? I know of other teams that are very similar, with moving players, or else with young players who are kept on JV until, like, their junior year or something like that.

Ben Dillon · Post by **Ben Dillon** » Fri Oct 19, 2007 2:48 pm

The possibility of players moving between teams, whether through promotion, demotion, attrition, or nefarious design, makes me think that maybe B teams shouldn't even be included in the rankings. For purposes of the rankings, David could take whichever team finishes the highest and designate their result as the one to be ranked.

Obviously, this solution would penalize B teams that are stable, like mine, and B teams that excel from year to year, like TJ's. But I'm not sure any solution really "solves" the problem anyway.

Byko · Post by **Byko** » Fri Oct 19, 2007 3:23 pm

To be honest, I'm not sure what the best way to handle JV tournaments is. There are only teams marked as JV based on the first couple of tournaments this season because of how clearly they were delineated. But I really don't have a good answer and am open to recommendations.

Quantum Mushroom Billiard Hat · Fri Oct 19, 2007 3:39 pm

Perhaps you could assume that JV teams are the next highest teams the school can field, in order. For example, with Something A, Something B, Something C, Something JV A, Something JV B, you can change JV A and B to D and E. I'm not sure how well this would work, but it is a possibility.

STPickrell · Post by **STPickrell** » Fri Oct 19, 2007 4:12 pm

I'd be disinclined to include B teams due to the fact that their makeup is way more inconsistent.

A team members are usually more active and consistent, whereas at many schools the B team is whoever wants to go to that tournament on that weekend.

On the other hand, Byko, if you want to include B and lower teams, I can send you the Valley District results which include 34 teams from 9 schools.

emactruman · Post by **emactruman** » Tue Oct 23, 2007 11:13 am

Byko,

If your looking for more scores of a random league in Michigan, try

http://www.megaconferencequizbowl.com

or

fulton.edzone.net/high/quizbowl/quizbowlmain.html

also if you contact Ann Saunders from UM-flints T.V. 28 she can send you scores from their tournament.

Ben Dillon · Post by **Ben Dillon** » Tue Oct 23, 2007 5:47 pm

How feasible/desirable would it be to simply lump together any non-A team results as B? Usually there's enough of a falloff after an A team that other teams are interchangeable, and their players are more likely to change during the year anyway.

When the E team beats the D team by a score of 20-10 in a tournament, I don't see this as being significant enough to count, so they may as well have been counted as one team. In fact, any game between two teams of the same school probably shouldn't count for national ratings anyway. Couldn't a team influence the ratings by holding its own intramural tournament then?

Post by **DumbJaques** » Tue Oct 23, 2007 6:36 pm

How feasible/desirable would it be to simply lump together any non-A team results as B? Usually there's enough of a falloff after an A team that other teams are interchangeable, and their players are more likely to change during the year anyway.

Not very, because this is just plain wrong. A lot of teams have B, C, and D teams that are competent to quite varying degrees (the only thing they have in common really is the ability to beat up on so-called "A" teams from programs that lack experience or ability). The regular placement of TJ B, Dorman B, Gov B, RM B, and various others depending on the year is more than enough evidence that there isn't necessarily "enough falloff," and what's more those teams regularly field C teams that are better than most A teams.

In fact, any game between two teams of the same school probably shouldn't count for national ratings anyway. Couldn't a team influence the ratings by holding its own intramural tournament then?

Well, Dave is the one who enters the games so I would think something like that woudl be apparent. But seriously, if a team hosts an intramural tournament for the express purpose of increasing its status in Bykowski's ratings, then said team deserves to be barred from quizbowl for all eternity. Seriously, this would never happen.

Ben Dillon · Post by **Ben Dillon** » Tue Oct 23, 2007 7:06 pm

Okay then, what if coaches designated before a tournament whether the tournament would be rated for their teams? I could see a coach saying that their A and B teams were to be counted as "official" and their C team would be "unofficial". This would allow for B and lower teams that are more stable.

I'm just trying to generate any ideas here, not necessarily good ideas. I also see the problems with these solutions :)

Dachspmg · Post by **Dachspmg** » Tue Oct 23, 2007 9:39 pm

Ben Dillon wrote:Okay then, what if coaches designated before a tournament whether the tournament would be rated for their teams?

That might work if every school had a coach. :p

Post by **theMoMA** » Tue Oct 23, 2007 10:27 pm

So why does it really matter what players constitute each team? Just treat each team as an entity independent of players. It seems like this is another confounding of backward-looking and forward-looking stats. All you should really care about is which teams did what. Who they were composed of, how many points they won by, etc. shouldn't matter because rankings all about results.

STPickrell · Post by **STPickrell** » Wed Oct 24, 2007 6:32 am

Dachspmg wrote:
Ben Dillon wrote:Okay then, what if coaches designated before a tournament whether the tournament would be rated for their teams?
That might work if every school had a coach. :p

I don't know if I ever actually met Mrs. Donohue (the TJ sponsor) or not.

Nine-Tenths Ideas · Post by **Nine-Tenths Ideas** » Wed Oct 24, 2007 8:58 am

Byko-
Mr. Keegan of Blake recently held a 1 team v. 1 team tournament with 6 rounds and full results, including point totals verified by both team captains. He said he'd be happy to send you the results but needs you email.

Byko · Post by **Byko** » Wed Oct 24, 2007 9:24 am

NeverHitTina wrote:Byko-
Mr. Keegan of Blake recently held a 1 team v. 1 team tournament with 6 rounds and full results, including point totals verified by both team captains. He said he'd be happy to send you the results but needs you email.

Thanks--I'll be sure to drop him an e-mail sometime this week.

Nine-Tenths Ideas · Post by **Nine-Tenths Ideas** » Wed Oct 31, 2007 5:01 pm

Byko- Results email has been sent.

Byko · Post by **Byko** » Thu Nov 01, 2007 7:09 am

NeverHitTina wrote:Byko- Results email has been sent.

Got it--thanks.