PurpleChest
Joined: Oct 25, 2003
|
Britnoth you miss the point again by coming at it from the wrong angle and seeking only proof for your own current theorums.
The point is not that a diverse police force should be the main priority for funding or enforced in some way, but that one that limits this happening is illegal (under Uk law and EU law) expensive (in compensation for discrimination payments) innefficient (by not accepting/getting the best people for the job) and contrary to the will of the poeple (as supported by all political parties outside of the very lunatic fringe: BNP mainly but some elements of UKIP also).
Perhaps if one day you are forced to quit your job to spend 1 day a week caring for an elderly relative (to take just one random example NOT using race) rather than have your employer see the value of retaining your knowledge and skill and experience by being flexible to be able to offer you 4 days a week of employment either temporarily or permanently, you will wish for diversity to have had an impact for you.
I say it again as clearly as i can: Responding positively to the needs of the individual, if they do not contradict the needs of the organisation, and not using them as a barrier to that persons progression or as a means of harassment of that person in order to maximise the performance of every individual and th organisation.
While you continue to see this solely as 'my taxes give black people jobs unfairly' I'm afraid it is you that comes over as race focused and a bit racist. |
_________________
Barbarus hic ego sum, quia non intelligor illis -Ovid
I am a barbarian here because i am not understood by anyone |
|
pac
Joined: Oct 03, 2005
|
 Posted:
Dec 30, 2008 - 19:22 |
 
|
So, Brit is saying that the box unfairly favours gay coaches? |
|
|
westerner
Joined: Jul 02, 2008
|
| Posted:
Dec 30, 2008 - 19:27 |
|
| TheCetusProject wrote: | | 1) many arguments against the BR component of the scheduler are fundamental philosophic objections. Data are not required to support our position. |
In theory, one can have a position based purely on principle, but when arguing to change an existing practice, I feel it's useful to quantify (or at least not exaggerate) the amount in dispute before assessing what action should be taken.
| TheCetusProject wrote: | | 2) Since TS-difference can be positive and negative, one must be careful before making inferences based on averages. Were I to play 10 games at +100 TS and 10 games at -100 TS I would care little that my average TS difference is 0. Anyone taking averages of TS differences should use the absolute values of the differences to make better inferences. |
Any coach will experience +/- TS matches randomly based purely on what's in the box. The point at issue is perceived bias against good coaches. Average is a better measure for that than variability. (EDIT in case it's not clear: TS differences are currently capped at 15.)
| TheCetusProject wrote: | | Edit: 4) If you wanted a data-based approach to analysing the situation, you would have to take the system that most of us "anti" people are happy with: where BR is an influence only to decide which TS-fair matchups are made. You would try that system out and compare the results with the current system. Then you could try to analyse the relative values of the two systems in a structured way. But since we don't have such a system to compare with, I don't really see what a data-based analysis of just the one system can really do. |
Too complex. You don't need that to analyze complaints of unreasonable TS bias under the current system. Just need to look at a sample of actual matches experienced under the current system and average them out. |
_________________
\x/es
Last edited by westerner on %b %30, %2008 - %19:%Dec; edited 2 times in total |
|
TheCetusProject
Joined: May 25, 2004
|
| Posted:
Dec 30, 2008 - 19:35 |
|
I think the assessment of the problem as being "bias against good coaches" is wrong. I am an average coach, and I will not suffer bias. But if I am getting lots of games playing up TS and lots of games playing down TS I will be unhappy, even if my average TS difference is zero. I challenge the idea that people only dislike playing up TS. I would dislike frequently playing down TS against better coaches.
Similarly, were I a good coach playing at a significant TS disadvantage most of the time, and being unhappy with the experience, and I was occasionally pitted against a much stronger player and got a significant TS advantage for these games, such that my average TS difference fell below a certain level that was judged "reasonable", I wouldn't think "oh, my average is in the reasonable limits, I must be satisfied", I would think that I was just getting lots of matches I didn't really want.
Proportion of non-irritating unbalanced matches is more important than average TS difference. |
|
|
Britnoth
Joined: Aug 02, 2003
|
| Posted:
Dec 30, 2008 - 19:39 |
|
| PurpleChest wrote: | Being black, gay, female, dissabled etc etc will slow and limit your progression in the organisation. Statistics prove it unarguably,
=======
I am as like to be championing the rights of male pattern baldness, or youth, or people with tattoos or class background as that of ethnic minorities.
======
Perhaps if one day you are forced to quit your job to spend 1 day a week caring for an elderly relative (to take just one random example NOT using race)
While you continue to see this solely as 'my taxes give black people jobs unfairly' I'm afraid it is you that comes over as race focused and a bit racist. |
So, which argument are you trying to use here?
First its ethnic minorities are unfairly treated and need help.
Then its people with minor differences in physical appearance.
Then it is wanting to work 4 days a week you are championing.
Sorry, how does wanting to work less hours or part time fit in with 'diversity' ???
Surely anyone can be in that position whatever race or whatever ???
Sounds like you cant make a rational argument for whatever you do mate. You bring up race then say thats out of date, you bring up facetious stuff like baldness and then drop it for something which is more related to flexible working than anyones preconceived ideas about another race/gender/age or whatever.
Except fall back on the old 'if you dont support diversity youre a racist' one I guess eh. Well I think ill stop there.
| Quote: | | Proportion of non-irritating unbalanced matches is more important than average TS difference. |
Spot on. |
|
|
PeteW
Joined: Aug 05, 2005
|
| Posted:
Dec 30, 2008 - 19:40 |
|
| Britnoth wrote: | | You get paid to inform people that baldness is okay? What rights does being bald provide me? Can I be protected from constant abuse like 'cueball' and 'lex luthor' ? |
/me points at Brit and yells "Slaphead!"
 |
_________________
"Jesus loves me this I know, 'cos my Bible tells me so." MrMojo - where did you go? |
|
westerner
Joined: Jul 02, 2008
|
| Posted:
Dec 30, 2008 - 19:49 |
|
| TheCetusProject wrote: | | But if I am getting lots of games playing up TS and lots of games playing down TS I will be unhappy, even if my average TS difference is zero. |
Well, that is a different problem, not connected to the inclusion of BR in the scheduler. Random TS differences are a tradeoff for quick nonpicked matches. The more coaches/teams the scheduler has to work with in a given round, the better a job it can do in this regard. |
_________________
\x/es |
|
TheCetusProject
Joined: May 25, 2004
|
| Posted:
Dec 30, 2008 - 19:50 |
|
I also think that those supporting TS boosts for weaker coaches have not experienced how bad it is to play with a 33TS advantage and still get massacred. |
|
|
PurpleChest
Joined: Oct 25, 2003
|
Britnoth i really think you are deliberately trying not to see the point, possibly for comic effect. I have continually used examples that differ to help you see the scope of the concept you are failing to grasp, you contiinue to respond to the individual and ignore the totality.
So i will repeat without example for you:
Responding positively to the needs of the individual, if they do not contradict the needs of the organisation, and not using them as a barrier to that persons progression or as a means of harassment of that person in order to maximise the performance of every individual and the organisation. |
_________________
Barbarus hic ego sum, quia non intelligor illis -Ovid
I am a barbarian here because i am not understood by anyone |
|
TheCetusProject
Joined: May 25, 2004
|
| Posted:
Dec 30, 2008 - 19:52 |
|
| westerner wrote: | | TheCetusProject wrote: | | But if I am getting lots of games playing up TS and lots of games playing down TS I will be unhappy, even if my average TS difference is zero. |
Well, that is a different problem, not connected to the inclusion of BR in the scheduler. Random TS differences are a tradeoff for quick nonpicked matches. The more coaches/teams the scheduler has to work with in a given round, the better a job it can do in this regard. |
Of course it is not totally tied in with having BR in the scheduler, but clearly having the scheduler behave in a way that is often happier to give TS-unbalanced games than it is to give TS-balanced ones, cannot do anything but make the problem worse. If I were frequently getting significantly TS-unbalanced matches with random opponents, so that they average out at 0 difference, I would be unhappy. If I were getting the same unbalanced matches but based on BR, I would be slightly happier, but not anywhere near as happy as I would be to get non-significantly unbalanced TS matches against random opponents... even if the average TS difference would be non-zero! |
|
|
westerner
Joined: Jul 02, 2008
|
| Posted:
Dec 30, 2008 - 20:12 |
|
| TheCetusProject wrote: | | Of course it is not totally tied in with having BR in the scheduler, but clearly having the scheduler behave in a way that is often happier to give TS-unbalanced games than it is to give TS-balanced ones, cannot do anything but make the problem worse. If I were frequently getting significantly TS-unbalanced matches with random opponents, so that they average out at 0 difference, I would be unhappy. If I were getting the same unbalanced matches but based on BR, I would be slightly happier, but not anywhere near as happy as I would be to get non-significantly unbalanced TS matches against random opponents... even if the average TS difference would be non-zero! |
It's a bit tricky to follow your train of thought there, but let's keep the BR-factor and general TS variability separate. Do I read you correctly that you see general TS variability as the larger issue?
If so, one approach you can take is to play more teams. That increases your chances of getting a balanced-TS match if that is most important to you. However, since you are currently playing a single B team (as am I  ) then it follows that we might experience more general TS variability. |
_________________
\x/es |
|
EvolveToAnarchism
Joined: Aug 02, 2003
|
| Posted:
Dec 30, 2008 - 20:20 |
|
|
TheCetusProject
Joined: May 25, 2004
|
| Posted:
Dec 30, 2008 - 20:34 |
|
I'm not basing this on my experiences, I am just arguing that the average TS difference is on its own not a particularly useful measurement of whether your matches are at acceptable TS differences most of the time. As an average coach I expect, under the current system, that I will play a fair number of games at acceptable TS, a fair number of games at unacceptable TS upwards against worse players, and a fair number of games at unacceptable TS downwards against better players. I would end up with an average TS difference of about zero.
In a system I favour, that tries to match people who are close in TS and then close in BR, and failing that tries to match you with someone close in TS but distant in BR, I would also expect to end up with an average TS of about zero. So examining average TS difference would not allow me to make any meaningful comparison between the two systems. Obviously I want to make meaningful comparisons, so I need something more subtle than TS difference average.
The thing of interest to me is the proportion of matches where the TS difference is too high. The problem with the blackbox scheduler is that it doesn't think a close match with a close BR is any better than a big TS difference match with a big BR difference. If I am trying to have a data-based analysis of the blackbox scheduler to try to determine whether the proportion of big TS difference games is too high, I can't use TS-difference average as a measurement because TS difference has a sign. 10 games at +15 TS and 10 games at -15 TS is unacceptable, even though the average is 0. If one wants to use an average, the absolute difference is the one to use. |
|
|
westerner
Joined: Jul 02, 2008
|
| Posted:
Dec 30, 2008 - 21:02 |
|
There are many factors that currently go into assigning matchups: BR, BBR, the TS formula, valuation of handicaps, and what's in the box when you activate. My purpose in this thread is to analyze the effect of BR, since that is what many are complaining about.
Average TS difference is a useful measure to test claims of BR-related bias against a particular strong coach. Can we agree on this?
Scenario: A CR180 coach with a very high BR believes that the scheduler discriminates against him because of frequent -TS matchups.
Test: Look at the average TS difference over a reasonable sample of games to see if there is a consistent negative TS trend. If the average TS difference is small, it is reasonable to conclude there isn't much BR-related bias.
I agree this test does not measure overall TS variability or frequency of undesireable matchups but those are not really BR related and that is not what I am trying to analyze. |
_________________
\x/es |
|
Eddy
Joined: Aug 04, 2004
|
| Posted:
Dec 30, 2008 - 21:47 |
|
| westerner wrote: | | There are many factors that currently go into assigning matchups: BR, BBR, the TS formula, valuation of handicaps, and what's in the box when you activate. My purpose in this thread is to analyze the effect of BR, since that is what many are complaining about. |
Yes, but since we have a different opinion of what the premises of [B] should be.
So it does not matter if the data currently shows that there is no "problem", since it is coincidence. It'd be good for bad reasons, instead of good for good reasons. |
_________________
'The generation of random numbers is too important to be left to chance.'
Robert R. Coveyou |
|
|
|
| |
anyone know how to c...
Search
