| Poll |
| Is this good |
| Ye |
|
10% |
[ 8 ] |
| Na |
|
67% |
[ 54 ] |
| Maybe |
|
22% |
[ 18 ] |
|
| Total Votes : 80 |
|
Frankenstein
Joined: Jan 26, 2005
|
 Posted:
May 24, 2005 - 13:41 |
 
|
| Christer wrote: | | The main difference between [R], [F] and [U] is the ranking system used, and the existance of these divisions give people a choice. |
As much as I appreciate [F] and the tournaments in [U]: the only official cups (aka the real thing) on fumbbl are held in [R] after all. Personally, I hate coach ranking as much as I'd love to participate in those cups.  So far, I could resist the temptation though.
| Christer wrote: | | Am I to blame for the fact that most people choose to play in [R]? |
If you allowed coaches to opt out of coach ranking (as in the good old times), even more people would play in [R]. I'd love to "blame" you for that!   |
|
|
thesquig
Joined: Apr 11, 2004
|
| Posted:
May 24, 2005 - 13:50 |
|
| Farinroderel wrote: |
If you allowed coaches to opt out of coach ranking (as in the good old times), even more people would play in [R] |
We dont agree on much. But I think this is a good idea. However, it would make unranked less popular still. |
_________________
Nuffle Sucks!!!
 |
|
Mully
Joined: Aug 02, 2003
|
| Posted:
May 24, 2005 - 13:54 |
|
| Quote: |
Well, is there any chance, that we can come back to our topic - the ranking formula?
|
Circ - this was covered several months ago. Check the search option on the forums when Christer 1st revised the coach rankings. I went head to head with him discussing issues similar to what you have.
He listened to what I had to say but kept his changes believeing that CR was mush more important than TS. This is is call and his site so other than throwing in my $.02 there's not much more I could say.
I think the emphasis on CR was his attempt to stop coaches from picking on newbies to boost CR, but I dont know if that worked as you see alot of top 50 coaches with 80-90% win percentage which is not very realistic if you are picking even games. |
_________________
Owner of the REAL Larson
Come join the CCC League |
|
mstrchef13
Joined: Dec 11, 2004
|
| Posted:
May 24, 2005 - 13:56 |
|
About CR...
I'm quite certain that something could be done with calculating the standard deviation from the mean of CR or to convert the CR to a percentile score before calculating how valuable the game is to a CR. That way, if you're in the 40th percentile and I'm in the 35th percentile, then given everything else equal my beating you is as likely as someone in the 75th percentile beating someone in the 80th percentile.
About TS...
There have been complaints about how TS overvalues some skills and undervalues others. I don't claim to even begin to understand how it's calculated and I've looked at the formulas. But there are a few things I noticed that I wondered about. The first is that there is only a passing mention of MNG players. My interpretation is that the player is treated as if he isn't on the roster. I think that there should be some adjustment factor for team size. A MNG player on a team with 14 players is much less debilitating than a MNG player on a team with 11 or less players. Or worse, 3 MNG players on a team with only 11 to start with...
The second is that the default AV is 8. But, who is more likely to get squashed, a wood elf lino (AV7 + AG4) or a human lino (AV 8 + AG3)? Because the wood elf is much less likely to get hit, I think that he's less likely to get squashed. But, according to the strength formula, 6-3-3-8 and 6-3-4-7 are valued the same. Because of that, in terms of TS, I think that there is a bias somewhere. Not exactly sure how to put it in words. I just think that elves (particularly woodies and pro elves) aren't rated to be as strong in the TS formula as they really are...
The third is that I'm not exactly certain why some negatraits are direct reduction in player value (WA) while others are a multiplicative reduction (Bonehead). In essence, this means that the Minotaur is just as wild at 100 SPP as he is at 0 SPP, but the Ogre is more stupid at 100 SPP than he is at 0 SPP. At least that's what it looks like to me. Now, don't get me wrong, I think WA sucks (stupid minotaur failed his WA roll 7 of 14 times) but Bonehead sucks as bad. Oh, and Pro should not be treated as a standard skill for Big Guys...
Finally, there is a small note, almost an afterthought, at the bottom of the TS page that says each handicap roll is worth 5 TS points. Now, let's be realistic. Am I more likely to beat you because I got palmed coin or Morley's revenge? Some handicap rolls are very strong. Some just suck stale beer. I think if handicaps are to be included in the TS, then they should be treated individually. "Bad Press" shouldn't get included. "I am the Greatest" should cause a TS reduction of the opponent of the average of the two who got it (since basically one of them becomes MNG). Stuff like that...
I'm tired of the flames. Like Circ said, let's get back to the original topic. I don't know if I'm blowing smoke out my arse, but these are just things that occurred to me that might jumpstart the conversation. |
|
|
Laviak
Joined: Jul 19, 2004
|
| Posted:
May 24, 2005 - 14:33 |
|
I must say ... to me, the formula used to calculate CR is not really an important issue. More important is how accurate the TS formula is (purely my opinion, I'm sure there are plenty who would disagree).
Don't get me wrong, the current TS formula is a WAY better estimator of 'how good' a team is than TR... at least, it is over a certain range. I just think that it could be improved (just don't ask me how to improve it ...)
Completely random suggestion ... how about hiding the CR completely, but still calculating it, and just showing the 'coach standing'. Then, no one really knows what their CR is, and has nothing to complain about, but they can still say "HAHAHA i'm ranked number 666 of all coaches in FUMBBL!!!". I'm sure there will still be plenty of complaints, but I actually think it would work quite well |
_________________
We Fink Wer Orks
--------
Help save blood bowl, foul an elf today!. |
|
sk8bcn
Joined: Apr 13, 2004
|
| Posted:
May 24, 2005 - 17:08 |
|
rrrraaaaaaaaaaa another nerving thread.
Let me think 20 secs:
-Ranked is a lame because there is CR
-U (without CR) is a lame because nobody plays it.
-ladder (wich propose a system that makes plays best coaches one against the other) should be deleted because nobody plays in it.
-stunty -the non competitve group- sucks because it is non competitive
mmmmmmmm
ok I see.
No really guys who moans about CR should go play in U. As soon as U community playing open will grow it will solve this problem.
After that, about R tournament:
I play in them, for CR uniquely. Honestly I don't find them better as U tourney. Average play level in U is as high as in R tournament, but they also are more creative.
For the topic: I am no fan of the CR calculation proposed here, similar to ladder somehow in the principle. I would rather see more investisment to see ladder revive. |
|
|
LordSnotball
Joined: Nov 05, 2004
|
| Posted:
May 24, 2005 - 23:48 |
|
As far as i'm concerned, if you don't like the system, don't play it. don't go off and tell others they are wrong for playing something different somewhere else. its simple, some ppl like sci-fi, others like fantasy. some, strange tho it may seem, some actually like RL! *gasp*
The range of personality differences can be seen in the range of teams around. different races, different choices, different builds, and after TR 150, its almost impossible to find 2 teams identical...
my point? why should i conform you your opinion and gameplay? why should you conform to mine? why should anyone conform, period? christer has done an amazing job to provide flexibility and options, and at the end of the day, you choose what you play, like it or lump it.
now, rl i used to have a BB gaming group of 4 ppl. where i'm living atm, i doubt there's another BBowler within 2 hrs drive. Compared to isolation and just looking at my BB box in the cupboard and sighing, i actually get to play occassionally... isn't that great? *lame cheers* i'd like to say thanks to christer for allowing the community to grow, and keep it going...
oh, and instead of CR, i like looking at my win rate... rarely around 30%! |
_________________
-Snottie
The Congregation - Always Recruiting
[url=http://igolocal.net/badge.php?user_id=1949] [/url] |
|
sk8bcn
Joined: Apr 13, 2004
|
| Posted:
May 25, 2005 - 00:19 |
|
|
Mithrilpoint
Joined: Mar 16, 2004
|
| Posted:
Jul 21, 2005 - 03:04 |
|
I just had a stray thought on the topic of Coach Ranking. I chose this thread (of the many possible) to post it in.
Making Coach Ranking matter.
One of the main points against CR is that it doesn´t actually show the skill of the coach, that people cherrypick weak opponents and thus it does not live up to its name. This is an idea on how to improve that.
As the emphasis is placed on CR rather than on TS in the ranking formula (Circularlogic among others has some good points about this) wouldn´t it be an idea to further limit the playable opponents by putting a restriction of maybe 10 cr points on a ranked game?
This meaning that you would be able to play and upload games against all coaches, but only matches vs. coaches within 10 CR points of your own would count for CR purposes.
As I see this it would produce both pros and cons.
Pros:
It would not be possible for the 160+ rated coach to play coaches below 150, gaining 0,35 cr per game and thus slowly rising in the CR-race without really showing any skill...ever.
Cherrypicking noobs would be made more difficult for the same reasons.
The games that matter CR-wise would be against people of roughly your own skill. And this i would conceive as good.
Cons:
It would be possible to pick games just outside the 10 cr range, strengthen your team, making it ready for the games that mattered (again, CR-wise).
More restriction on opponents in an open environment is never positive. But bear in mind that you could still upload the result of the match, gain money and skills, just not CR.
-o-
Possibly i have left out several things/consequences of this so let me hear them.
What do you say to this good people?
M |
|
|
Wyrmtainter
Joined: Nov 30, 2004
|
| Posted:
Jul 21, 2005 - 05:02 |
|
EDITED |
Last edited by Wyrmtainter on %b %21, %2005 - %05:%Jul; edited 1 time in total |
|
Wyrmtainter
Joined: Nov 30, 2004
|
| Posted:
Jul 21, 2005 - 05:44 |
|
As for Team Strenght and some skills... Here is a little suggestion:
1) Keep the actual TS formulas as for a Basic TS
2) Implement an "Actual" TS which could be added between () after the Basic TS on a match-up with another team.
3) Tackle-Dodge:
a) If No Dodge on one side, the Oppo Tackle skills = 0 Actual TS (since they are useless for that game)
b) Count the Tackles on 1 side. Divide by 11 (As Max players on pitch. Divide by number of players Avail at beginning of game if under 11). Add together the TS value of Dodge skills of the whole Oppo team. Multiply by the % obtained here. This would be the Actual TS of the Dodge skills for that game
c) Repeat point a) and b) for the other side.
So, in brief. First, if there is no dodge on a team, the opposition Tackles arent counted for this game, since they will not be useful anyway. If there is an actual dodge-tackle match-up, then the Dodge skills TS value would be dropped for this game ONLY.
Sure Hands - Strip Ball:
The same formula than Dodge-Tackle could be used for this skill, exception made of point a), since Sure Hands is useful for picking the ball, w/wo presence of strip ball on the other side. It would only reduce the value of Strip Ball by the presence of Sure Hands on the Oppo.
4) Razor Sharp Claw/ Dirty Player:
I consider DP as RSC since 9 times in 10, a Foul is made with so much assists that the DP is used on the Injury anyway.
a) add the TS Value of RSC and DP of the whole team together.
b) Calculate the Average Armor Value of the Oppo. Divide 8 by this value. (8 being an average Armor)
c) Multiply the Added TS Value of RSC and DP by the % obtained in b). This is the Actual TS value of RSC/DP for the game.
In brief, an higher Armor would reduce the value of the oppo RSC/DP, while a lower Armor would increase it.
5) Claw
a) Calculate the Average Armor of the Oppo Team. Round down. Find out the TS-diff between this AV and if it was 2 lower.
b) Divide the number of Claws on the team by 11 (or number of avail players if less than 11).
c) Multiply the result of a) by the % obtained in b). This is a TS Reduction on the Oppo side.
Since Claw just lowers the actual armor value of the oppo without any other effect, it is normal that it lowers that armor TS accordingly.
Mighty Blow:
a) Follow the rules of #4, divided by 2
b) Follow the rules of #5, divided by 2
I shuffled this idea in my head a while, here i word it out for some brainstorming. |
|
|
Laviak
Joined: Jul 19, 2004
|
| Posted:
Jul 21, 2005 - 06:44 |
|
Interesting thoughts Wyrmtainter.
Something else to consider is the effect of handicaps. Currently, they are worth 5 TS each, however, the real value is so variable - e.g. Virus - unfortunately, you don't know until after the match starts, so this could really be applied only to the change in CR calculated after the match, not whether or not the match is valid to be played (i.e. within 40TS).
With the dodge/tackle idea, I'll have to have a think about the formula.
Some things to consider:
- a single player with tackle can be a pretty big help, it can be used to blitz every turn - probably the first tackle is worth more against a team with several dodgers than the rest of them. Even more so if that player has ST4, mighty blow, or similar.
- On the other hand, as soon as there are players without tackle, a single dodger can be used to mark that player, keeping him/her relatively safe from the tacklers.
With the rules for armour/injury modifying skills, there are some interesting formulae used in determining the "total effective number of players", based on things like niggles and armour value. This is used to modify the total team value. I think that the injury/player modifiers should work in the same way (though I haven't come up with an actual formula for it  ). One thing to remember is that if you have a single claw, it will probably be used for a lot of blitzes (note: this kinda depends on dodge/tackle and other skills too). This can be pretty important against teams like wood elves that tend to dodge their players away.
I'd also like to say that I think MithrilPoint's idea has a bit of merit. It might help to prevent the more experienced coaches from picking on newer ones. |
_________________
We Fink Wer Orks
--------
Help save blood bowl, foul an elf today!. |
|
sk8bcn
Joined: Apr 13, 2004
|
| Posted:
Jul 21, 2005 - 08:46 |
|
to wyrm: the ideas are good but the formula becomes relative. It can then be calculated ONLY vs another team and not by his own. That's a problem IMO. |
_________________
Join NL Raises from the Ashes |
|
sk8bcn
Joined: Apr 13, 2004
|
| Posted:
Jul 21, 2005 - 08:56 |
|
to Mith now:
main problem of the solution is the following case:
Imagine I play Circular for exemple in a fair game. I am far lower than 10 CR below him. Imagine I beat him. So I get no reward for that? |
_________________
Join NL Raises from the Ashes |
|
Wyrmtainter
Joined: Nov 30, 2004
|
| Posted:
Jul 21, 2005 - 11:17 |
|
Thats exactly the point, to be relative.
In fact in this idea the Team Strenght as we know it wouldnt change, therefore on a team page by example, it wouldnt change anything. The ideas I gave is to ADD a new TS value (I called it Actual, where Basic is the TS as we know it), which is relative to the match-up of skills, that would only show and be used in that specific match-up, for the purpose of CR changes after the game as well as for the 40TS cap.
By example, a Zon team that is at 145TS, goes on the gamefinder. ITs TS doesnt change, yet the relative ADDED TS (i suggested between () just after the TS), might be 130 against some dwarves. In the same fashion, a 160TS Chaos Team with 4 RSC might be worth 170 if matched against let say, a team of flings.
That new value would only be used for the purpose of CR changes (repeating myself here)
As for Laviak input, it is true that the Tackle skill can be very useful even if only 1 or 2...as long as the players with it are maneuvrable. Same goes with Strip Ball. |
|
|
|
|
| |
Gnomes are trash
Search
