The Mad Scientist Top 25 Ranking Debate Topic

metsmax
Posts: 1598 (0)
Hall of Famer
5 2 836 2
Pro
colonel

if you sincerely want to produce something and get constructive input, I suggest that you start a fresh thread in which you set out

1. the approach you propose to take

2. open questions about which you would like more information from experienced players

3. information - if any - you would need from WIS or other users to produce a sample ranking for some division in some world

This thread has seen some evolution of your concept, some good questions and answers and a lot of unhelpful ad hominem. If the goal is really to produce something of interest and use to an audience, something like that might lead to a constructive path.

Personally, I am a skeptic, but I would be interested in seeing the output. My view remains that rankings are better derived - or mostly derived - from game results rather than from inputs to the game engine - but heck, although I think your scheme would not work, maybe it would be interesting.
12/28/2009 11:12 AM
colonels19
Posts: 4399 (0)
Hall of Famer
78 90 3 1 168 2 1 11
Prospect
Quote: Originally Posted By zhawks on 12/28/2009
Quote: Originally Posted By colonels19 on 12/28/2009

Quote: Originally Posted By zhawks on 12/28/2009
It doesn't matter how much or how little ratings play in your system. Ratings do not have a place in tournament seeding so even .1% ratings it too much.
Unless you've been hired as seble's towel boy, you don't know this for certain...you don't program HD.



I don't have to be hired to know whether how talented your team is or isn't shouldn't be a part of whether your team makes the tourney or not, only your resume should. That is just common sense. you said "Ratings do not have a place in tournament seeding so even .1% ratings it too much." How do you know this concretely? You don't work for WIS, so this is nothing more than another bizarre claim you made that you have to and are unlikely to prove.

If you think that somehow how talented your team is ratings wise somehow should matter then there is nothing to debate. In SOS, it works....for an ENTIRE RANKING SYSTEM, its preposterous. That just wouldn't be logical thinking on your part.
12/28/2009 11:12 AM
a_in_the_b
Posts: 1653 (0)
Hall of Famer
1 18 666 3
Pro
I think you are parsing his statement wrong.

It is easily assumed and the most parsimonious and likely interpretation that he believes or is saying that they don't have any (rightful) place in seeding.



Interpreting it the other way just so you can try to mock it is just being bull headed for the sake of it.

12/28/2009 11:15 AM
colonels19
Posts: 4399 (0)
Hall of Famer
78 90 3 1 168 2 1 11
Prospect
Quote: Originally Posted By oldresorter on 12/28/2009
colonels - how would the system handle this?

i am having trouble posting team data on here, but in world 1, in the acc UNC is 6-20 rpi 188 / #3 SOS, in the big east rutgers is 24-2 rpi/rank both #2, #57 SOS yet their team totals when adjusted for one walkon, are near identical, I spent 30 seconds to find this one example, since I am in the acc, I found highly rated poor performing unc immediately, then looked at the ranked teams, my team is one, so rutgers was 2, seriously, the first 2 teams i looked at were rutgers and unc (unc was 8 better when adjusted for the walkon)

If I use the OTR SOS, the hierarchy would be as follows...Win over UNC, Win over Rutgers, Loss to UNC, Loss to Rutgers. If their ratings are nearly identical as you say than the difference between the 2 wins, would be relatively small. If I use my real life SOS...the Rutgers win is a ton better all day.

12/28/2009 11:16 AM
a_in_the_b
Posts: 1653 (0)
Hall of Famer
1 18 666 3
Pro
Quote: Originally Posted By colonels19 on 12/28/2009
Quote: Originally Posted By a_in_the_b on 12/28/2009
If we think that using ratings alone is flawed for SOS< and SOS is a strong component of the system, then why does the rest of the system matter until the major flaw(In our view) is fixed?
Key words are bolded.

You haven't seen what the rankings have rolled out...you can't really say you'd like them or not because you haven't seen what they've done/will do yet. If you don't see OTR SOS as a viable way to judge/gauge overall team strength, then I can't help you...we disagree, big deal....but its a disagreement of OPINION and not FACT. I am not wrong because I want to roll out an OTR SOS.



SAve, for me, using the rankins as the sole arbiter of SOS is an competely fatal flaw. Again, if overall ratings are not sufficient to determine the quality of a team by themselves, they are also not adequate to determine the quality of its opponents by itself. If its an inadequate rater of a team, its an inadequate rater of a team, whether talking about a team or the teams it plays.



12/28/2009 11:17 AM
colonels19
Posts: 4399 (0)
Hall of Famer
78 90 3 1 168 2 1 11
Prospect
Quote: Originally Posted By a_in_the_b on 12/28/2009I find it amazing that he apparently thinks its somehow bad form of me to post exactly where the original post is so people can read it themselves, and unfair to him.
Come on man, you're not stupid. You posted the only part of my post that surprisingly backs your "personal attack" argument against me, while conveniently leaving out the most important part...what the OTHERWISE was referring to...come on man, I'm a college graduate for crying outside. You know what you did and to try to spin it like you did nothing wrong is pathetic.
12/28/2009 11:19 AM
colonels19
Posts: 4399 (0)
Hall of Famer
78 90 3 1 168 2 1 11
Prospect
Quote: Originally Posted By a_in_the_b on 12/28/2009
Hell with it. Someone can't give you anything that requires even a MODICUM of interpretation or analytical skills to understand> You aren't willing to extend even a MODICUM of effort to see how this applies to you, so I hardly see why I or anyone else should apply a modicum of effort to try to explain. Its obvious that the implication is that the aggregate number is an inadequate measure to tell the story. Who plays, who doesn't, where they play, what their numbers are. ... This is why I want to make the adjustment to the overalls to account for which guys are playing the minutes and which aren't....I'm with you on that.



12/28/2009 11:20 AM
colonels19
Posts: 4399 (0)
Hall of Famer
78 90 3 1 168 2 1 11
Prospect
Quote: Originally Posted By zhawks on 12/28/2009But if your system can't handle extreme examples then it is flawed, it has to be able to account for stuff like this
My system will NEVER see a team of one 910 and nine 10s....so I guess I'm going to have to be content with being flawed then................
12/28/2009 11:22 AM
a_in_the_b
Posts: 1653 (0)
Hall of Famer
1 18 666 3
Pro
Quote: Originally Posted By colonels19 on 12/28/2009
Quote: Originally Posted By a_in_the_b on 12/28/2009
I find it amazing that he apparently thinks its somehow bad form of me to post exactly where the original post is so people can read it themselves, and unfair to him.
Come on man, you're not stupid. You posted the only part of my post that surprisingly backs your "personal attack" argument against me, while conveniently leaving out the most important part...what the OTHERWISE was referring to...come on man, I'm a college graduate for crying outside. You know what you did and to try to spin it like you did nothing wrong is pathetic


If I was trying to do that, I wouldn't have provided the means for people to check on it, I woudl have just it out, obviously.

12/28/2009 11:22 AM
a_in_the_b
Posts: 1653 (0)
Hall of Famer
1 18 666 3
Pro
Quote: Originally Posted By colonels19 on 12/28/2009
Quote: Originally Posted By zhawks on 12/28/2009
But if your system can't handle extreme examples then it is flawed, it has to be able to account for stuff like this.
My system will NEVER see a team of one 910 and nine 10s....so I guess I'm going to have to be content with being flawed then...............




You aren't stupid, you know damn well that isn't what the esxample meant, and its rather disengenuous to pretend that it does.

Straw man all the way.

12/28/2009 11:23 AM
isack24
Posts: 1566 (0)
Hall of Famer
81 19
Prospect
Quote: Originally Posted By metsmax on 12/28/2009
colonel

if you sincerely want to produce something and get constructive input, I suggest that you start a fresh thread in which you set out

1. the approach you propose to take

2. open questions about which you would like more information from experienced players

3. information - if any - you would need from WIS or other users to produce a sample ranking for some division in some world

This thread has seen some evolution of your concept, some good questions and answers and a lot of unhelpful ad hominem. If the goal is really to produce something of interest and use to an audience, something like that might lead to a constructive path.

Personally, I am a skeptic, but I would be interested in seeing the output. My view remains that rankings are better derived - or mostly derived - from game results rather than from inputs to the game engine - but heck, although I think your scheme would not work, maybe it would be interesting.

I was going to suggest the same thing. I agree wholeheartedly with everything.
12/28/2009 11:25 AM
colonels19
Posts: 4399 (0)
Hall of Famer
78 90 3 1 168 2 1 11
Prospect
Quote: Originally Posted By dalter on 12/28/2009
It's not part of the HD selection system though. I don't work for WIS, but I have asked this question directly in the past.

And you think WIS told you the truth? lol Its all about damage control on their end and how you asked the question matters. Did you say, "Ratings don't matter in the NT selection process, do they?" or did you say, "Do ratings have ANY involvement in the NT selection process, yes or no?" because those are 2 entirely different questions.

WIS has lied to me before about things I've questioned that I later found out about, so their word is anything but gold. They're going to want to tell their masses, what the majority like to hear, period. Truth takes a backseat to damage control...I've seen it first hand here.
12/28/2009 11:26 AM
dalter
Posts: 71 (0)
All Star
25
Prospect
Quote: Originally Posted By colonels19 on 12/28/2009

Quote: Originally Posted By oldresorter on 12/28/2009

colonels - how would the system handle this?

i am having trouble posting team data on here, but in world 1, in the acc UNC is 6-20 rpi 188 / #3 SOS, in the big east rutgers is 24-2 rpi/rank both #2, #57 SOS yet their team totals when adjusted for one walkon, are near identical, I spent 30 seconds to find this one example, since I am in the acc, I found highly rated poor performing unc immediately, then looked at the ranked teams, my team is one, so rutgers was 2, seriously, the first 2 teams i looked at were rutgers and unc (unc was 8 better when adjusted for the walkon)

If I use the OTR SOS, the hierarchy would be as follows...Win over UNC, Win over Rutgers, Loss to UNC, Loss to Rutgers. If their ratings are nearly identical as you say than the difference between the 2 wins, would be relatively small. If I use my real life SOS...the Rutgers win is a ton better all day.

OK, but the system that many of us have issues w. is the one that bases SOS on overall ratings, which in OR's example would have a win vs. a 6-20, 188 rpi team worth slightly more as a win vs. a 24-2, 2 rpi team.

I just can't get past that. How do you reconcile that?
12/28/2009 11:26 AM
zhawks
Posts: 881 (0)
Hall of Famer
208 66 38
Pro
Quote: Originally Posted By colonels19 on 12/28/2009
Quote: Originally Posted By zhawks on 12/28/2009

Quote: Originally Posted By colonels19 on 12/28/2009

Quote: Originally Posted By zhawks on 12/28/2009
It doesn't matter how much or how little ratings play in your system. Ratings do not have a place in tournament seeding so even .1% ratings it too much.
Unless you've been hired as seble's towel boy, you don't know this for certain...you don't program HD.



I don't have to be hired to know whether how talented your team is or isn't shouldn't be a part of whether your team makes the tourney or not, only your resume should. That is just common sense. you said "Ratings do not have a place in tournament seeding so even .1% ratings it too much." How do you know this concretely? You don't work for WIS, so this is nothing more than another bizarre claim you made that you have to and are unlikely to prove.

If you think that somehow how talented your team is ratings wise somehow should matter then there is nothing to debate. In SOS, it works....for an ENTIRE RANKING SYSTEM, its preposterous. That just wouldn't be logical thinking on your part.
How talented your team might be should not be part of any consideration for whether you make the NT or not, that is why we play the games, otherwise let's just give the title to someone after recruiting.
12/28/2009 11:27 AM
zhawks
Posts: 881 (0)
Hall of Famer
208 66 38
Pro
Quote: Originally Posted By a_in_the_b on 12/28/2009
I think you are parsing his statement wrong.

It is easily assumed and the most parsimonious and likely interpretation that he believes or is saying that they don't have any (rightful) place in seeding.



Interpreting it the other way just so you can try to mock it is just being bull headed for the sake of it.



Yup. And we know it isn't used currently and rightfully so.
12/28/2009 11:30 AM
◂ Prev 1...64|65|66|67|68...75 Next ▸
The Mad Scientist Top 25 Ranking Debate Topic
△Top△Bottom

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2026 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.