Irgy
Joined: Feb 21, 2007
|
 Posted:
Jan 05, 2009 - 07:24 |
 
|
A review and a set of opinions for what to use for [B] matching. I make no apologies for the fact that it's currently entirely my own opinions. I will however be perfectly happy to incorporate any other good arguments people give into what's written below. It will still be my opinion, but my opinion is surprisingly flexible.
I thought about making this a blog post, but I really do want other people's opinions on what I've written and I'll get more of those in the forums.
Format:
I discuss a bunch of options, comparing them to the one above (they're in a sensible order for doing this).
* means I'm explaining what something is in the first place
+ means something I think is an improvement over the method above
- means something I think is made worse compared to the method above.
TR:
I'm going to use this as a starting point for the comparison, so not much to say other than: It's not as bad as you might think.
TR is already what the game uses for limiting winnings, it's what a lot of leagues use to keep teams in roughly the same range, and people already have to manage their teams for TR so it's not an extra thing for everyone to worry about. It won't always give fair matches, and it's terrible for certain races, but it's mostly fair in the long run, and has a simplicity and purity about it.
One thing I will say though is that [B] really lacks a reward for getting your TR high. In [R] and [L] and probably most other places, getting an uninjured, high TR team is the main goal in many ways. There's arguments for this being a good or a bad thing, and it's another issue in its own right. I mention it mainly because it impacts on some of the issues below.
TV:
As per LRB 5.0
+ This doesn't include the cash you have sitting around. For [B], I think this is a clear improvement, especially as you need to have spent the money before the game anyway.
+ MNG players don't count towards your total, as they won't play anyway. Most people prefer this.
- In this case though, what does mng mean really? In leagues and tournaments it means you have one less player for the next game, making it harder to win. Here it will just mean you get a softer opponent. Your team will usually be less optimal for the loss of a good positional or a key player, but in some cases it will be improved too. The intention of mng has been warped here.
+ The formula for player improvements based on a number of skills and what was rolled rather than on spps is fairer. The spp management that TR forces is still there for managing winnings.
- Giving extra TV for getting +St or doubles makes those good rolls less valuable. In the extreme, if these were costed as high as their true worth, then in the context of [B] they'd be completely unexciting to get, and people might even reject them. However I don't think this is a big problem with the current formula.
My own TV-TS hybrid suggestion:
This is unashamedly my personal opinion on what the best choice is. I'm going to give the reasons why I think so, and you can make of them what you will.
I describe it here relative to TV. I'm also only comparing it to TV here, so half the arguments for and against it are in the TS section below.
* Use the most up-to-date and fancy formula available for calculating the base cost of the players, but then just use the TV method (+2 per skill, more for good skill rolls)
+ I'm all for balancing out the rosters, and adjusting for what is what is over/under priced. I still want to let players live with their own choices
* Include the 5/6 multiplier for each niggle
+ This is roughly the fairest way of accounting for niggles.
- This makes niggles arguably no longer a penalty, again breaking the intent behind rules for getting injuries. I still think the extra variability that comes with niggled players will keep with the spirit of niggling injuries though.
TS:
So I'm taking a kind of idealised TS here, which is a measure of a team's power that should give on average an equal chance of winning to any two teams of the same TS - subject to averaging out the paper-scissors-stone style differences in racial matchups. Some have even suggested the scheduler account for those too, at least for the clsasic tackle vs dodge.
There's a lot of ground in between that and the current TS, but I'm treating that all together for simplicity.
+ Games are more truely balanced with a good TS system.
- The more accurate TS is, the less effect good team management decisions have on things. With an ideal TS, team management is completely irrelevant and you can just build whatever team you like safe in the knowledge that the TS formula will give you easier opponents to offset your bad decisions.
- The more complicated TS is, the more the skill of team management moves away from being about maintaining an effective team with the resources given and towards gaming the TS system. People shouldn't have to understand the TS formula to play well, and people shouldn't get an advantage by finding loopholes in it.
Personally I think it's very important that team management is not factored out of [B], because team management is absolutely what [B] is all about. The one diffrence between [B] and [R] is that people are building teams on equal terms. With already no reward for reaching a high TR, taking out the reward for having a well optimised team makes [B] a completely pointless excercise.
(anything) + a BR factor:
+ Games move towards a 50-50 chance of either side winning, giving a greater proportion of close and therefore interesting games.
- Big TS (or TV or whatever) difference games aren't really much fun. The way I look at it is this: Winning when you're way ahead in TS anyway is lame. Losing because you were way behind in TS is demoralising. Losing when you were way ahead in TS is embaressing, demoralising and frustrating. So the only people having fun must be the skilled coaches winning half the time despite a huge TS disadvantage. But are they having fun? No - in fact they're the people complaining the most, because even if they win them sometimes, they don't enjoy playing at a disadvantage all the time.
- It really skews what BR means in the first place. Being a good coach should be about winning all types of games. If BR is in the matchup forumla though, maintaining a high BR is entirely about winning TS-down games. Even worse, for an erratic coach (or even just a coach of erratic teams like elves) who can win games outmatched but can also lose even with a big advantage, it's completely unstable. They could equally maintain a high BR or a low BR, just depending on how they do in their first few games.
- It also skews team management. The team management punishment for high BR coaches gives unwanted feedback into the BR system and again makes it unstable. It also allows the tactic of using teams designed to lose games and lower their BR, to allow for easier matchups for the coach's other teams. It's cherry picking all over again.
Of course how strong these factors may turn out to be in practise is not clear. They're strong philosophical problems but may or may not make a big difference in practise.
I have nothing against a bias in the scheduler towards pairing coaches with similar BR, if it can handle it. People do prefer to play people of a similar skill level, independantly of whether or not they want a handicap when they don't.
All in all, I think if people want a 50-50 chance of winning every time, and want ineffective skill choices on fluffy teams, they can do that much better already in [R]. [B] should be about people forcing themselves to play matchups people wouldn't normally take but on a fair playing field for everyone. The scheduler forumla should just be as simple and fair as possible, while doing its best to preserve every aspect of the spirit of blood bowl.
Edit1: Incorporated some of the points made below |
Last edited by Irgy on %b %06, %2009 - %00:%Jan; edited 1 time in total |
|
westerner
Joined: Jul 02, 2008
|
| Posted:
Jan 05, 2009 - 15:38 |
|
Good review and ideas...
Regarding incorporating BR into the scheduler:
-there's going to be inherent TS variability in matchups just due to what's in the box. This can be perceived as BR-bias when it's really luck of the draw.
-most cases of reported BR-bias have shown an average of less than 5TS difference.
-I haven't seen anyone in favor of a pure 50-50 system. Some are in favor of a system that moves *toward* 50-50 by giving a slight TS edge to the weaker coach. Given the 15TS cap, I believe the current system effectively works this way.
-the issue about teams designed to lose games and lower BR is a potential exploit, but as yet I've not seen too may examples of it. |
_________________
\x/es
Last edited by westerner on %b %05, %2009 - %15:%Jan; edited 1 time in total |
|
JanMattys
Joined: Feb 29, 2004
|
| Posted:
Jan 05, 2009 - 15:49 |
|
| Irgy wrote: | | One thing I will say though is that [B] really lacks a reward for getting your TR high. In [R] and [L] and probably most other places, getting an uninjured, high TR team is the main goal in many ways. That missing reward is another issue in its own right, but I mention it because it impacts on some of the issues below. |
Strongly disagree on this one.
High uninjured TR might be a goal in R, but just because there are Majors to win.
In B, all that matters is the W/T/L record.
I disagree that the main point of DivB is that everybody can build a team on equal terms. On the contrary, team building is exactly what Blackbox is NOT about. In DivB all that matters is how victorious your team is in a random-pairings environment, because that's a measure of your skills as a coach.
Team development for team development's sake is a cancer that must be extirpated once you leave an Open division: team development leads to cherrypicking in R and to coaching Dorfs/Orcs/Khemri (i.e. easy-to-develop-and-keep-alive teams) in B. |
_________________
 |
|
westerner
Joined: Jul 02, 2008
|
| Posted:
Jan 05, 2009 - 15:59 |
|
| JanMattys wrote: | | On the contrary, team building is exactly what Blackbox is NOT about. |
Then, why bother with it at all? Why not just have a division that uses one-off, preconstructed teams of varying TS levels? |
_________________
\x/es |
|
JanMattys
Joined: Feb 29, 2004
|
| Posted:
Jan 05, 2009 - 16:02 |
|
| westerner wrote: | | JanMattys wrote: | | On the contrary, team building is exactly what Blackbox is NOT about. |
Then, why bother with it at all? Why not just have a division that uses one-off, preconstructed teams of varying TS levels? |
What do you mean, why bother?
BB is more than just rolling dice, you know...
My team is my team, my players are my players, my skill rolls are my skill rolls, my fluff is my fluff, and my wtl record is my wtl record.
Still, I don't give a damn if my 20/0/0 humans are TR 250 or TR 120, at the end of the 20 games streak.
edit: but again, Westerner, you consistently play TS down in R so I guess you are overly-protective of your pixels and teams. I can see why we have different views on this matter. No offence intended, it's just personal views on the game. To me, a healthy and "strong" team that didn't gain its status through mud, sweat, blood and courage isn't really worth anything. On the other hand, a crippled team, fighting for its own existence in every game and continuously fighting for survival in close battles to me is worth all the respect I can give, even if its TR is low. |
_________________
|
|
westerner
Joined: Jul 02, 2008
|
| Posted:
Jan 05, 2009 - 16:26 |
|
| JanMattys wrote: | | edit: but again, Westerner, you consistently play TS down in R |
Actually I haven't played much R lately. Mostly B. In fact, at this point, I've played more B than R.
Stereotypes FTW! |
_________________
\x/es |
|
Eddy
Joined: Aug 04, 2004
|
| Posted:
Jan 05, 2009 - 16:27 |
|
| westerner wrote: | | JanMattys wrote: | | edit: but again, Westerner, you consistently play TS down in R |
Actually I haven't played much R lately. Mostly B. In fact, at this point, I've played more B than R.
Stereotypes FTW! |
You playing more in B than R doesn't change the fact that you do (or don't, i haven't checked nor do i care =P) "consistently play TS down in R". |
_________________
'The generation of random numbers is too important to be left to chance.'
Robert R. Coveyou |
|
JanMattys
Joined: Feb 29, 2004
|
| Posted:
Jan 05, 2009 - 16:31 |
|
| westerner wrote: | | JanMattys wrote: | | edit: but again, Westerner, you consistently play TS down in R |
Actually I haven't played much R lately. Mostly B. In fact, at this point, I've played more B than R.
Stereotypes FTW! |
But again, Westerner, you consistently played TS down in R.
Happier? I don't see how this changes the meaning, anyway.
It's not called stereotypes, it's called statistics.
And in B you coach Dorfs, probably THE team for those who hate their pixels to die too often.
As I said, it wasn't a personal attack, just a reason why we see things different.
To me, high TR is not an accomplishment in itself. It's a nice thing to have if it happens, but I will never artificially seek for compromises that will let me reach high TR more easily than Nuffle wants me to.
High TR is not a goal. It's a state of things. And for many teams (for example humans, or Vampires, or Norse, or Zons) is not even a particularly favourable state of things.
No need to derail the good review by Irgy. I just posted because I disagree with one of the basic premises of his review, and I think my point makes sense. |
_________________
|
|
treborius
Joined: Apr 05, 2008
|
| Posted:
Jan 05, 2009 - 19:10 |
|
| Irgy wrote: |
TR:
I'm going to use this as a starting point for the comparison, so not much to say other than: It's not as bad as you might think.
TR is already what the game uses for limiting winnings, it's what a lot of leagues use to keep teams in roughly the same range, and people already have to manage their teams for TR so it's not an extra thing for everyone to worry about. It won't always give fair matches, but it's mostly fair in the long run, and has a simplicity and purity about it. |
i disagree strongly - you obviously have never tried to coach Ogres.
even when you look at a perfectly healthy team, its TR is usually 15..50% higher than its TS (w/o any players mng or spare cash) and surely their TS isn't too low 
this is the reason why Ogres in any TR-based league are even more unplayable than they already are in a TS-based environment.
| JanMattys wrote: | | On the contrary, team building is exactly what Blackbox is NOT about. In DivB all that matters is how victorious your team is in a random-pairings environment, because that's a measure of your skills as a coach. |
i disagree: When people like to build teams i don't see why they shouldn't do that in DivB as well as in any other div.
i don't oppose your style of playing and your view on the game either, but to me teambuilding is one aspect of the game and i wouldn't want to miss that.
| JanMattys wrote: | | Team development for team development's sake is a cancer that must be extirpated once you leave an Open division: team development leads to cherrypicking in R and to coaching Dorfs/Orcs/Khemri (i.e. easy-to-develop-and-keep-alive teams) in B. |
i agree, that a lot of teambuilders cherrypick in R, but that's true for a lot of win-oriented players as well
that's actually why i like divB: no cherrypicking 
as far as choosing high-AV-races is concerned: true, it's easier to coach / build them while winning is probably easier with low-AV-teams, but that's what makes this game so great: a lot of different choices and aspects to enjoy |
|
|
Irgy
Joined: Feb 21, 2007
|
| Posted:
Jan 06, 2009 - 00:39 |
|
A few replies
Westener:
I agree with the point I think you're making that the perceived problem with the BR bias is much bigger than the reality of it, and all the negative factors I talk about are not necessarily huge. I still think they're problems, though maybe more from a philosophical perspective.
JanMattys:
I think you make the point that high TR shouldn't need to be a goal or to be rewarded at all, and I think that's a fair point. Maybe the lack of reward for high TR isn't a problem so much as just a difference.
When I talk about "team building" being important in [B], I'm not talking about getting a high TR low injury team for winning tournaments. I'm talking about maintaining a good balance of skills, and keeping your team in working order as often as possible. To keep a good wtl record, you need to play well on the field to win games, but you also need to choose a good balance of skills, manage spps and injuries, and protect key players. These aspects are what I'm saying [B] gives a level playing field for, by forcing people to play teams that will kill and injure their key players.
treborius:
You're completely right about TR and Ogres. That's why I would hope that my preferred option incorporates most of the ways that using TS helps teams like Ogres by giving a fairer evaluation of the base worth of the players. In some cases it's not even that the players are badly valued but just that the roster on the whole is weak, and I'm not opposed even to giving entire races a systematic boost just to balance them out, as long as it was done right. |
|
|
westerner
Joined: Jul 02, 2008
|
| Posted:
Jan 06, 2009 - 00:53 |
|
| Irgy wrote: | | When I talk about "team building" being important in [B], I'm not talking about getting a high TR low injury team for winning tournaments. I'm talking about maintaining a good balance of skills, and keeping your team in working order as often as possible. To keep a good wtl record, you need to play well on the field to win games, but you also need to choose a good balance of skills, manage spps and injuries, and protect key players. |
I largely agree with this, to which I would add: I enjoy trying to make my teams total pool of skills as efficient as possible, and analyzing what skill choice would be best for a given player. I also like seeing players develop into "Top 10" rankings, understanding that this is just a milestone on the way to death or retirement. |
_________________
\x/es |
|
spelledaren
Joined: Mar 06, 2004
|
 Posted:
Jan 06, 2009 - 01:32 |
|
Where is the thumbs up function? Argh!
This seems like a good discussion, I might even add something useful to it later on! |
_________________
FUMBBL! |
|
Frankenstein
Joined: Aug 02, 2003
|
| Posted:
Jan 06, 2009 - 01:49 |
|
In my opinion, permanently injured players (i.e. not mng) should fully count towards TV, for both fluff and balance reasons. |
|
|
westerner
Joined: Jul 02, 2008
|
| Posted:
Jan 06, 2009 - 02:01 |
|
I understand the fluff reasoning, but why the balance? |
_________________
\x/es |
|
|
|
| |
BB2020 - Kick team m...
Search
