Free Allegiance Forums

We can talk about 'statistical accuracy' til the cows come home, fact is; any ranking system put in place is going to be arbitrary. Allegiance only needs something with a modicum of accuracy in regard to how skilled each player is, I think the best indicator would include kills/ejects, the number of points you get, and the number of hours you've played.

Technical Wonder wrote:QUOTE (Technical Wonder @ Dec 11 2007, 09:31 AM) We can talk about 'statistical accuracy' til the cows come home, fact is; any ranking system put in place is going to be arbitrary. Allegiance only needs something with a modicum of accuracy in regard to how skilled each player is, I think the best indicator would include kills/ejects, the number of points you get, and the number of hours you've played.

I support a sliding scale system with permanent AB.

Terralthra wrote:QUOTE (Terralthra @ Dec 10 2007, 11:07 PM) Mandatory auto-balancing is essential to any accurate rating system. Turning it off for SGs is certainly going to be the appropriate thing to do, and those games will also have to not count towards any of the players' rankings, as well. The historical fact that the vast majority of allegiance players are against auto-balance in pick-up games is one data point that points in the direction that the 'majority of the community' doesn't really want an accurate rating system nearly as much as the people in said group proclaim.

This is a fundamental problem, I agree. If such a system were to be implemented it would have to be imposed by authority, not democracy. I think it is less about whether or not people want an accurate ranking system and more about whether people are willing to make the sacrifices necessary to have one. Symantics perhaps, but symantics worth considering when trying to impose a ranking system on the community. Mr. C has the right attitude - he's willing to give up a bit of freedom to achieve balanced gameplay.

I've heard the rumblings of this TrueSkill thing, and people smarter than me seem to think it could be very promising. I hope that those that are working on adapting it to allegiance keep doing so until it comes to completion, and that the powers that be have the courage to impose it. I think it could do a great deal in the long run for improving allegiance gameplay.

Terralthra wrote:QUOTE (Terralthra @ Dec 10 2007, 10:13 PM) I'm confused. You're agreeing with my earlier statement that the majority of the community doesn't support AB, and hence, does not really want accurate ranking. Is this your version of a concession?

You usually are, confused that is. I said that there will not be an accurate system, Trueskill might be a nice fun thing and perhaps better, not accurate.

See this is your typical crap right here:

A) Community doesn't support AB.
Therefore
Community doesn't want accurate rankings.

See who the $#@! are you to draw conclusion like that, and present them as valid points, that's why you use Wookies so often. It's not even a close expression of truth, it's utter BS. At best the community doesn't support a broken AB system, the community is unlikely to support mandatory AB, the community is clearly fine with a voluntary AB system that works- now we just need it to work (the system itself not the rating component). Once that's done having ratings only count with AB games (my suggestion Day 1) is a nice addition, couple that with a rating system that works more often then not (even Helo I suppose in the darkest of winters) will work better.

As for "B)" that's why you tick me off, your god complex.




QUOTE Nice attempt to obscure that you haven't answered any of the points I raised in my posts.[/quote]

Watch out for the Wookie. /rolleyes.gif" style="vertical-align:middle" emoid=":roll:" border="0" alt="rolleyes.gif" />


Anyway, *MY* point is that using AllegAge to average with any rating system will be more accurate then without doing so.


*Blood pressure increasing*

`yr

P.S. Merry hoho /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" />

Deleted? Doah.

jgbaxter wrote:QUOTE (jgbaxter @ Dec 11 2007, 03:38 PM) Anyway, *MY* point is that using AllegAge to average with any rating system will be more accurate then without doing so.

Why would adding a totally arbitrary factor to a statistical system improve it? Why not take into account number of forum posts? It has the same bearing on actual Alleg ability as unique days logged in.

I think it is clear we should simply let Spidey rank everyone from #1 all the way to #Bacon.

Dengaroth wrote:QUOTE (Dengaroth @ Dec 11 2007, 05:32 AM) And it most certainly is. For conclusive proof, see the accumulated data on human history.

Also see any issue of Lobster Ghostboy quarterly.


I am still awaiting my ranking Spidey you nubbin.

jgbaxter wrote:QUOTE (jgbaxter @ Dec 12 2007, 01:38 AM) You usually are, confused that is. I said that there will not be an accurate system, Trueskill might be a nice fun thing and perhaps better, not accurate.

See this is your typical crap right here:

A) Community doesn't support AB.
Therefore
Community doesn't want accurate rankings.

Sorry about that 'logic' crap. I suppose if I switched to your system of 'baseless assertions with no support at all' you'd be much happier? See, you skipped a step which is critical:

A) Mandatory AB is essential to an accurate rating system.
Most of the community would vote against mandatory AB for pick-up games
ERGO
C) Most of the the community does not really want an accurate rating system, as they are unwilling to make the requisite sacrifices.

That is not a non-sequitur, it is valid logic. It has two statements, and the logical conclusion follows naturally from those two statements. If you have an alternate conclusion you can draw from facts A and B, perhaps you could share it with us. Since you didn't do that the first three times I stated my logic, I can only assume that you have no alternate explanation, and are calling it 'crap' in a desperate, futile attempt to paint it as something other than a rational conclusion.

jgbaxter wrote:QUOTE (jgbaxter @ Dec 12 2007, 01:38 AM) See who the $#@! are you to draw conclusion like that, and present them as valid points,

Someone capable of rational and logical thought. I can understand your confusion.

jgbaxter wrote:QUOTE (jgbaxter @ Dec 12 2007, 01:38 AM) that's why you use Wookies so often. It's not even a close expression of truth, it's utter BS. At best the community doesn't support a broken AB system, the community is unlikely to support mandatory AB, the community is clearly fine with a voluntary AB system that works- now we just need it to work (the system itself not the rating component). Once that's done having ratings only count with AB games (my suggestion Day 1) is a nice addition, couple that with a rating system that works more often then not (even Helo I suppose in the darkest of winters) will work better.

In R4, the AB system which people screamed was 'broken' was fixed. How many games have you seen with AB on since then? I have seen precisely 0. You can spout unfounded assertions all day long, but the simple truth is, we have voluntary AB right now. No one uses it. Do you have any evidence that that would magically change?

jgbaxter wrote:QUOTE (jgbaxter @ Dec 12 2007, 01:38 AM) As for "B)" that's why you tick me off, your god complex.
Watch out for the Wookie. /rolleyes.gif" style="vertical-align:middle" emoid=":roll:" border="0" alt="rolleyes.gif" />

I don't have a god complex, I have a 'simple well-reasoned argument', which you have utterly failed to answer, rebut, or even contest.

jgbaxter wrote:QUOTE (jgbaxter @ Dec 12 2007, 01:38 AM) Anyway, *MY* point is that using AllegAge to average with any rating system will be more accurate then without doing so.

For the third time: How? How does this averaging improve 'any' rating system? How can you even begin to make such a claim, when you clearly have not even the slightest bit of mathematical knowledge of the systems in question? How can you continue to make such a claim in the face of both an obvious and not infrequent situation in which your 'idea' would produce much less accurate results, and another situation which already exists in which two of the acknowledged worst players in Allegiance have more time in-game than most well-known vets?

This is why I don't get too stressed out about your 'wookiees and ewoks' bull@#(!. You obviously know you don't have the critical thinking abilities to actually rationally debate this sort of issue with me. So instead, you make utterly groundless assertions, while throwing out these stupid well-poisoning lines about ewoks and wookies and god complex, hoping that those observing the thread silently will side with you.

Terralthra wrote:QUOTE (Terralthra @ Dec 10 2007, 06:13 PM) If enough people vote that 1+1 = 3, does that become true? This is not a matter of me knowing better than them, this is the math being self-evident. HELO is not a zero-sum system, a reversible system, a net-correcting system, or a Bayesian-compliant system. It is completely inaccurate. Mucking about with arbitrary constraints (Aarm gets a new account, plays for 1 week, wins 50+ games in a row, loses none, but he's a (1), that's accuracy!) is not going to fix the problems. You seem to be casting this as a democratic issue, when it is not.

Mathametical rigour is not something to which democracy applies. You say "the answers to such things are never absolute," but in math, the answer to "is this system accurate?" can be absolute. Math contains absolute answers to problems, in the vast majority of cases. A system which is inaccurate does not become accurate by popular consent. That's not a "rhetorical trick of someone losing a vote," there is no vote. The population of Allegiance can most certainly decide to use a system, and that's perfectly within their rights, but that doesn't make it an accurate system.

I'm sorry I diverted the hijacking of this thread /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" />

I was not questioning the fact that a properly researched and tested system would be to the benefit of the community. I applaud the efforts made to that end. The check of a vote or at least the consensus of those familiar with the issue is required to check such a system against reality. That is the extent of my specific argument.

I was reacting based on reading too deeply into your words: 'yet another example of a population with a significant number of people who make decisions squarely against their own interests'. Best interest is a subjective thing; and as a libertarian I would say it is a personal thing. 'Their own best interests' is a phrase i have heard used by those on the wrong side of certain democratic processes to explain why others don't agree with them. In my opinion, it is in my best interest to have an accurate ranking system for allegiance, for taxes to go down, and for government to shrink not grow. Others may feel that rankings need to go away entirely, for taxes to be raised as needed to provide services, and for the government to meet the needs of all people. There are no absolute answers to any of these issues as stated.

But, like I said, I trust those who have been involved in this process to keep fighting the good fight. If you wish to discuss this digression further, perhaps we should move it to another forum?