So it’s that time of year again … No, I am not talking about spring, or finals, or papers, or preparing for the summer, or celebrating bids to Regionals.
Nope, instead it is time to bash the Bama Section on RSD.
Happens every year. No matter how well the top team from the Bama Section performs (see ALL of LSU’s past performances at Regionals compared to their seed going into the tourney), Bama Secs gets no love. Not only that, but the section seems to be the whipping post for all of the wannabe ultimate gods and geniuses out there among the seemingly dwindling legitimate and logical ultimate minds.
For all of the non-respect the section gets, people sure do like to talk about us!
Not that the players/captains/coaches/bloggers of the Bama Section are free of blame here. Sure, we have added fuel to the fire in past years, but I always considered it self-defense. I mean, if no one else out there is going to stand up for the Bama Section’s honor than we certainly should.
Oh well, all of the chatter and Bama Section bashing will mean nothing once Regionals gets underway. Use it as bulletin board material if you like, but just make sure to break seed and beat at least one team that plans on looking past a Bama Section team at Regionals.
Oh, and one last thing …
You probably consider me to be one of those “wannabe ultimate gods and geniuses” that spends my time spitting on the Bama section. I consider myself to be one of the “legitimate and logical ultimate minds.” Is there any way we can resolve this conflict and be friends?
Seriously, if you feel I’m going out of my way to illogically bash and denigrate the Bama section then please tell me where and how I am doing that. Because if that is what is happening then I’m doing something horribly wrong when attempting to seed regionals. I dont care who ends up where, I just want to find a process for seeding by which two distinct reasonable people will arrive at the same seeds when given the same set of information.
allen
ps. I do think that the comments on the Bama section this year are connected to “fanning the flames” in previous years more than this year’s teams.
You wrote when seeding UNT how muddled your logic became when it seems that the 3 teams (UNT, Wash U, and LSU) all have legitimate arguments for that seed. The reason you end up seeding UNT over the other two is because of wins over in regionals teams which is by far the least logical step of your process. Why does it matter that LSU’s wins are over teams not at regionals since those teams were still quality opponents and obviously quality wins?
Also, I don’t agree with the logical step to discount wins against opponents just because the opponent has already been seeded. Such as LSU’s W-L vs Arkansas and Arkansas’s W over Texas (though this doesn’t affect their seeding in your system). LSU has a W-L vs Arkansas. Arkansas finished two spots ahead of Truman St. Your system completly fails to take these results into account, and so fails to accurately seed teams.
After this step LSU goes from 4th to 9th, behind Wash U, & Truman St. due to losses to common opponents at a tournament where we were missing more than half of our team and probably should have gone as an X-team (hindsight is 20-20). Is it not painfully obvious that we did not have anywhere near a full squad (in fact missing 7 of our top 10) at mudbowl where we lost to Ga Tech and Emory (also losing to Bama who we later beat 13-2 with our full squad).
Fair enough. You have identified two parts of my seeding approach (laid out in advance) that you disagree with. I’ll try to expand on those and I would be very happy to entertain alternative seeding methods and refine the one that I am using.
Before going into that, I would like to raise one very simple point — this discussion is now about how the mechanism I chose to do the seedings and not about any respect (or lack of respect) for the Bama sections and LSU.
The first disagreement that you have is with making a distinction between teams at regionals and teams not at regionals when considering head to head results. This is a distinction that I am not 100% happy with at this point, though I do think it is worthwhile to make since it does allow for an additional *objective* mechanism for breaking apparent ties (in my current methodology the only place where it was used was to breakt eh UNT/WashU/LSU dilemma at 4th). If I did not make that distinction and instead went straight to RRI, UNT would still have gotten the 4th seed due to a higher RRI. So in these seedings it did not impact anything.
I am not 100% certain that differentiating between common opponents at regionals and common opponents not at regionals is the right thing to do. From my experience the last couple of years I have found it to be useful in preventing RRI (which I find to be very undesirable due to the relative lack of results) from deciding seeding positions.
Now, your second complaint concerns removing Arkansas from consideration once they have been seeded. Again, I am not 100% sure that this is the right thing to do. I find it appealing because, once Arkansas has been seeded, we have asserted that they are better than all teams that have not yet been seeded. Given this assertion, I find the transitive argument of LSU > Arkansas, Arkansas > Washington, therefore LSU > Washington to be suspect since because we have already asserted that the first part (LSU > Arkansas) is false by seeding Arkansas ahead of LSU. That is why I think it is a reasonable thing to do.
But what happens if we do not take that step? In this case, Texas State again receives the 5 seed off of common opponents and RRI (http://upa.org/scores/scores.cgi?div=23&page=3&team=1210&team=1723). When we move to the 6th seed and compare washington to LSU, washington again wins the battle of common opponents 3-2 (http://upa.org/scores/scores.cgi?div=23&page=3&team=1210&team=1240) when Arkansas is included for consideration. When we move to the 7th seed, truman state again is justifiably ahead of LSU based on common opponents (http://upa.org/scores/scores.cgi?div=23&page=3&team=1210&team=1295) by 3-1 if you take the implicit loss to Arkansas into account. Note that truman never played arkansas, so maybe truman would have won that game.
It is painfully obvious now, after you have explicitly told me, that you lost games at MudBowl thanks to taking a small team (sounds vaguely excusish to me to be perfectly honest). But how can I differentiate those losses from your loss to Missouri state. Similarly how can I tell that Texas Arlington wasnt playing with only rookies when you beat them 12-9 in Dallas? Or how can I tell that Washington wasnt playing an all lefty game when they lost 8-13 to Davidson? All that I have in front of me is the score reporter and the results reported for the participating teams and I am making a best effort attempt to seed fairly and consistently based on that data.
I respect that you dislike the seedings and believe that LSU will finish higher than seeded at regionals. At this point, I don’t see a consistent argument for seeding them higher than 9th. I am not saying that such an argument cannot be constructed, just that I dont see it.
If I look at your seedings (or the ones attributed to you on RSD), I have a few questions:
(1) Why seed UNT at 3rd rather than Arkansas given that Arkansas beat UNT head to head?
(2) Why seed LSU ahead of Wash U at sectionals based on the common opponent Arkansas while not mentioning Davidson, Michigan state and Notre dame? Especially when Notre Dame is the highest ranked (by RRI) of the opponents the two teams have in common?
Differentiating common opponents at the tournament from common opponents not at the tournament is a step to reduce the likelihood of a team being seeded below a team that they beat in a head to head match up. I’ll try to explain with the 4th place example here.
In the UNT/LSU/Washington example, UNT>atm>lsu is a possible seeding order explicitly justified by head to head results (UNT over atm at sectionals, atm over lsu at another tournament). If we break the tie at 4th place by placing lsu on top, then we will necessarily have LSU seeded ahead of atm despite the head to head competition. If we instead put UNT (or washington) on top, then we have avoided seeding lsu ahead of atm (a team that they lost to) for the moment. The three way tie is based on common opponents, and by respecting the common opponent at the tournament more than the other common opponents, we potentially avoid seeding a team below a team that they beat.
In some cases it is unavoidable that a team will be seeded below a team that they beat, for example texas > kansas > arkansas > texas. But it is something that I believe should be avoided if possible.
Does that make any sense at all?
In the process of going through this discussion, I realized that I made a mistake when considering the 4th seed:
Candidate teams are: Washington, UNT, LSU.
UNT > LSU : UNT has a (implicit due to sectional finish) win over common opponent atm at regionals. LSU has a win over common opponent Davidson not at regionals. If we do not differntiate between types of common opponents, then fall back to RRI and UNT > LSU.
UNT > Washington : as previously stated
Washington > LSU : as previously stated.
It was a mistake by my part in applying the methodology. This does not invalidate the concern that considering common opponents differently may not be the right thing to do. It does strengthen the case for UNT as the 4 seed however.
I can’t begin to analyze every one of those arguments you have, that would take hours. Apparently you spent some hours writing them unfortunantly for yourself.
I will say your sarcasm makes you sound like an ass. Do you even play on a college team right now? Who the fuck are you? Will I even get a chance to play you at regionals?
I pointed out what I thought were two fallacies in your process, and you half-heartedly agreed. I’m done with seedings. See you there, I suppose.
Schulz
28 – LSU
(1) You pointed out what you consider to be flaws in the system. I acknowledged that those were *choices* and explained why I made them.
(2) I pointed out that even if I abandoned the steps that you objected to, the seedings in this case didnt change.
(3) I asked you specific questions about how you arrived at your seedings and why you considered some factors and not others.
(4) I apologize for the sarcasm, it was not my intention to be an ass, but rather to point out that *I didnt know* the special circumstances surrounding your losses at mudbowl *nor do I know* of any special circumstances regarding any of your victories.
(4) I’m out of eligibility and have been for multiple years at this point. I coached Texas for 5 years. I have been the Texas SC for the last two years. I have been trying to work out an objective and fair system for seeding tournaments (the series specifically) for the last couple of years and attempted to enter into this discussion b/c you objected to the methodology. I was hoping to have a reasoned and logical conversation so I could learn (a) what you disagreed with and (b) any productive solutions.
Thank you for your time. If I see you at regionals, I’ll be one of the people wearing an Orange shirt.
So what’s the hold-up on the actual seedings at this point? Are we just waiting to see who wins the internet argument? I’m ready to have an actual opponent to look forward to rather than listen to a bunch of nameless 30 year-olds who used to be a backup for some 12th seeded regionals team in 2001. Fuck this shit.
RC is in contact with all the captains through email. I would guess they’ll be out sometime next week.
(3) I asked you specific questions about how you arrived at your seedings and why you considered some factors and not others.
I think H-H is the only clear cut objective way to seed teams. Past that, I think common opponents should be behind things like tournament finish and RRI because opponents change strength week to week. Ex. LSU beats Missourri 13-5 and loses to Missourri St. 14-12 later at Mardi Gras. But then Missourri make Regionals where Missourri St. does not so clearly something strange has happened there: either Missourri missing players or St. playing with nonseries eligible players or injuries to either team. The common opponents thing is just a tool you use so that you can make an algorithm so a computer could seed regionals. I don’t see the need for computers to be a final ranking method. Also, later in season should mean more than early season an example of why sectionals should mean more than Jan/Feb finishes. College basketball uses this logic, along with college football (though in unintentionally I suppose).
doesnt RRI take into account all opponents, which would incorporate common opponents.? if RRI is considered then you are considering common opponents. considering common opponents and RRI would be putting double weight on common opponents.
also isnt RRI a computer algorithm? i would assume that RRI has alot of weight in the final seedings of most series tournaments. since it is upa created, and the event is upa sanctioned. looks like the computers win again…
Chap (whoever the fuck you are), RC is in total control of seedings (but they also have to be approved by National Director or something), but you’re mostly incorrect about the use of RRI in seedings. Ex. Bama sectionals 2009 … I’m pretty sure you were aware that we were seeded 5th with the highest RRI.
Yeah, I’m basically saying common opponents shouldn’t be given near the weight in seedings that everyone is trying to give it.
THat video was completely uncalled for!
A few quick things
1) Allen, I dont think anyone was accusing you specifically of bashing the bama section. If they were, I think they were out of line. That was mostly other people while you were just involved in the discussion in between on RSD. You have put yourself on the firing block by showing up here. I would say sorry, but I am pretty sure you enjoy it.
2) To take one of your examples, when LSU played UT-A at Denton on Sunday I am pretty sure they had at least 2 players who have been out of eligibiity for several years on their team. They probably didn’t do this at sectionals and other non-home tournaments. In this case it is not LSU making an excuse for something they did. It is out of their control. This is one of countless examples in any given year of why objective seedings based on results will not give the best result even when based on common opponents. This process necessitates human interaction and subjectivity if we want to truly rank the strength of teams even reflectively. Unfortunately this means that there will be a ton of people lobbying and arguing. You are trying to avoid this with an objective system. I think it is unavoidable.
Ouch, nothing like pissing off one of the only observers at Regionals before it even starts. I’d make sure you feet stay onsides if I were you. Nice work!
brackets should be coming out soon right?
Seedings are in. WooHoo.
http://upa.org/scores/tourn.cgi?div=18&id=5921