JackassRampant
Joined: Feb 26, 2011
|
 Posted:
Sep 29, 2021 - 18:46 |
 
|
EUREKA!
Rather than gauging match suitability by ∆G, use ∆(2W+T)! Rookie-season protection vs teams more than one season old and up more than 15% TV. (EDIT: optional rookie protection, a checkbox on your team that you can toggle on/off for your first season before it goes away.)
That way, a team on a losing streak gets easier games, at least in theory, while more successful teams get harder games. And unlike gauging by CR, (EDIT: accumulated wins) in season is pretty volatile and won't lead to repeated matchups. |
_________________
Lude enixe, obliviscatur timor.
Last edited by JackassRampant on %b %29, %2021 - %20:%Sep; edited 3 times in total |
|
koadah
Joined: Mar 30, 2005
|
| Posted:
Sep 29, 2021 - 18:47 |
|
| mister__joshua wrote: |
| koadah wrote: | | ClayInfinity wrote: |
Majors should be for teams with 15 or 14 wins in a season...
|
That is a bit harsh. Prioritise by record and seed the top 16/32 would be better IMO. |
I agree with the principle of having a 'Season' in Competitive replace what was previously Major Qualifiers. The mechanics of it could be discussed, but seeding sounds like the fairest way. You could then have a tracker (like the box trophy page) showing the current qualifiers for the next Major and their seeding position, so coaches know the record they are trying to beat to qualify. |
I don't. 
What I meant was... If the tournament allows 128 or 256 teams then it would be the 128 or 256 with the best record that applied that get selected. The top 16, 32 or even all of them could be seeded.
Restricting Majors to only the top coaches would be bad for the community IMO. |
_________________
O[L]C 2016 Swiss! - April ---- All Star Bowl - Teams of Stars - 2 more teams needed |
|
The_Murker
Joined: Jan 30, 2011
|
| Posted:
Sep 29, 2021 - 20:16 |
|
+1 to the OP. Well written and thought out.
+1 to the forum guys who responded. So nice to see an idea not get pooped on in a disrespectful way.
I agree that seasons gives an excellent opportunity to eliminate massive TV differences in Box that many dislike. I agree very much that match making based on games played in current season seems like a great idea. 'Fair' with respect to how much building a team may have done since redraft, and well within the spirit of league play within the 'team tiered' world of BBowl.
Some form of rookie protection for teams in their very first season seems reasonable. A toggle switch might be do-able, but might be complicated for those new to the site. I suggest making it mandatory to activate AT LEAST 2 teams in Black Box. Assuming most of us agree that diversity in BlackBox is a good thing, the idea is that activating a Season 2+ team with your Season 1 team would prevent the new team getting selected against a relative monster. It would be your choice to get around this safeguard if you wanted to activate two Season 1 teams. Fair enough. |
_________________
Join the wait-list. Watch the action. Leave the Empire. Come to Bretonnia! |
|
Gartch
Joined: Sep 07, 2012
|
| Posted:
Sep 29, 2021 - 20:46 |
|
| JackassRampant wrote: | EUREKA!
Rather than gauging match suitability by ∆G, use ∆(2W+T)! Rookie-season protection vs teams more than one season old and up more than 15% TV. (EDIT: optional rookie protection, a checkbox on your team that you can toggle on/off for your first season before it goes away.)
That way, a team on a losing streak gets easier games, at least in theory, while more successful teams get harder games. And unlike gauging by CR, (EDIT: accumulated wins) in season is pretty volatile and won't lead to repeated matchups. |
Hi Jackass Rampant.
I'm sorry I don't understand what are G, W and T in your formula.
I am not against the idea of pairing based on teams results, but Chris explained that in the past, each time pairing was based on coach level, it resulted in way less coaches and matchs in Black Box.
In other words a vast majority or coaches don't want a pairing based on coach level (I suspect team results will be strongly related to coach level).
I am in the minority on this topic and must accept majority's opinion (I think having less match in black box is worst). |
|
|
mister__joshua
Joined: Jun 20, 2007
|
| koadah wrote: |
I don't.
What I meant was... If the tournament allows 128 or 256 teams then it would be the 128 or 256 with the best record that applied that get selected. The top 16, 32 or even all of them could be seeded.
Restricting Majors to only the top coaches would be bad for the community IMO. |
That’s not different, is it? Unless I’m somehow massively missing your point. I’m not advocating for limiting participation in tournaments, only that the tournament entry requirements could be based on some metric achieved during the preceding season instead of the current requirements which are typically based on high TV.
And then having some way to follow this so you could always see who had qualified. |
_________________
"Yeah, well, you know, that's just, like, your opinion, man." - The Dude
Mr. J's LRB7 / Forum |
|
Verminardo
Joined: Sep 27, 2006
|
| Posted:
Sep 29, 2021 - 21:07 |
|
I think matchmaking based on "games this season" makes sense and would be fun. That being said, Fumbbl has always had a very peculiar meta. Some coaches were on Box Trophy runs, some coaches played for ranking, some played for badges and top lists, some played to challenge themselves with motto teams, and many played to prepare their teams for tournaments and in particular, Majors. And of course yet others just played pick-up games one at a time for the fun of it. I have done all of the above except for the badges and top lists. And we will see how each of those turn out in the new rules.
So I guess I'm saying, I don't think TV matchmaking is much worse than it really always has been. I'm guessing it'll be a while still until we get seasons implemented and until then it'll do. But once seasons are implemented "games this season" seems like the logical thing as a starting point, although some observation and modification will probably be required. |
|
|
koadah
Joined: Mar 30, 2005
|
| Posted:
Sep 29, 2021 - 21:23 |
|
| mister__joshua wrote: | | koadah wrote: |
I don't.
What I meant was... If the tournament allows 128 or 256 teams then it would be the 128 or 256 with the best record that applied that get selected. The top 16, 32 or even all of them could be seeded.
Restricting Majors to only the top coaches would be bad for the community IMO. |
That’s not different, is it? Unless I’m somehow massively missing your point. I’m not advocating for limiting participation in tournaments, only that the tournament entry requirements could be based on some metric achieved during the preceding season instead of the current requirements which are typically based on high TV.
And then having some way to follow this so you could always see who had qualified. |
OK. I misunderstood you. In my mind "Qualifiers" are the tournaments that some Majors have that reduce the field down to the last 16 or 32.
I was responding to a post that said you'd need 14 or 15 wins in a season to join a Major.
I agree with using season record instead TV. |
_________________
O[L]C 2016 Swiss! - April ---- All Star Bowl - Teams of Stars - 2 more teams needed |
|
koadah
Joined: Mar 30, 2005
|
| Posted:
Sep 29, 2021 - 22:35 |
|
|
MattDakka
Joined: Oct 09, 2007
|
| Posted:
Sep 29, 2021 - 23:00 |
|
| Gartch wrote: |
I don't understand what are G, W and T in your formula.
|
I guess G stands for "Games played", W for "Wins", T for "Ties". |
|
|
JackassRampant
Joined: Feb 26, 2011
|
| Posted:
Sep 29, 2021 - 23:19 |
|
Right, Matt. 2 matchmaking points for each win in your current season, plus one per tie. A 0-0-5 team would have the same scheduling values as a 0-0-0 team. So teams that are young in their season and teams that aren't doing well this season would match up, while the teams having a good season would see other teams doing well, especially later in the season. |
_________________
Lude enixe, obliviscatur timor. |
|
JackassRampant
Joined: Feb 26, 2011
|
| Posted:
Sep 29, 2021 - 23:39 |
|
This gets around the problems of CR-driven matchmaking by being based on the team's status within the season, which is volatile and may vary from team to team, not coach to coach. |
_________________
Lude enixe, obliviscatur timor. |
|
Mingoose
Joined: Jul 28, 2016
|
| Posted:
Sep 30, 2021 - 00:34 |
|
I don’t think any of the suggestions in this thread are bad, but I don’t understand how any of the suggestions solve the issues around TV.
The premise is that TV doesn’t equal team strength, and the game is designed to compensate for TV differences in a meaningful way to make matches equal. Therefore there is no need to create matches with equal TV.
The reality is that people will game TV no matter what you do, because inducements matter, and gaming TV is *easy*. Rolling random secondaries saves 20 TV each. That is a big delta.
Maybe I am not understanding the premise properly, but arranging games by season progression won’t stop TV optimization. It is a design flaw inherent in a poor GW redesign and nothing can stop it other than extreme hate fouling of TV optimizers. |
|
|
JackassRampant
Joined: Feb 26, 2011
|
| Posted:
Sep 30, 2021 - 01:18 |
|
TV-driven matchmaking exacerbates this. BB2020 was designed around a non-TV-driven matchmaking system. OP's argument is that we should follow suit to avoid some of that added perversity. There will always be gaming. |
_________________
Lude enixe, obliviscatur timor. |
|
mrt1212
Joined: Feb 26, 2013
|
| Posted:
Sep 30, 2021 - 02:36 |
|
| Mingoose wrote: | I don’t think any of the suggestions in this thread are bad, but I don’t understand how any of the suggestions solve the issues around TV.
The premise is that TV doesn’t equal team strength, and the game is designed to compensate for TV differences in a meaningful way to make matches equal. Therefore there is no need to create matches with equal TV.
The reality is that people will game TV no matter what you do, because inducements matter, and gaming TV is *easy*. Rolling random secondaries saves 20 TV each. That is a big delta.
Maybe I am not understanding the premise properly, but arranging games by season progression won’t stop TV optimization. It is a design flaw inherent in a poor GW redesign and nothing can stop it other than extreme hate fouling of TV optimizers. |
It isn't meant to stop it, full on stop it - it's meant to avoid potentially goofy situations from prior matchmaking where a contingent of coaches went out of their way to optimize around in TV in ways that were just a pain in the neck to deal with if you weren't in on that racket.
The thing about a lot of where we're at right now is that we have conjecture on how people will optimize for the game with skill selection and TV management - to the extent where I really do think that using the cleanest measure already built into the rules - seasons, is a totally fine way to do it. We don't know how crappy meta gaming can really get, especially if it becomes unavoidable via some matchmaking method.
Meta emerges after time even though we all have a list of how we might make the most of it. I don't think we have clearly identified intolerable Nuffleshite just yet.  |
|
|
happygrue
Joined: Oct 15, 2010
|
| Posted:
Sep 30, 2021 - 03:13 |
|
| Rawlf wrote: | But in FUMBBLs Competitive division, TV is also used for matchmaking and considered the main indicator for team strength. And that is a very big problem.
...
TL;DR
Let's get rid of TV beyond what the rules need it for. Let's get our tilted minds back in line with the intent of the rules. |
I think Rawlf has framed the issue nicely. I tend to think solutions that pitch a "better" indicator of real team strength than TV - and there are many such plans that have merit - are not going to succeed, because the shift the problem to a new metric that will be gamed in similar ways, but might be "a bit harder to game".
I tend to speculate that a large motivator of those folks that are, we might agree, gaming the system for wins in dubious ways is to get "a higher rating" or perhaps just flat win rate. There is a lot of overlap with a high win rate and a high CR, so all the incentives in current matchmaking are pushing in that direction. You want maximal chance to win with minimal TV, to rank in this system.
Anyone trying to do "something else", like play a suboptimal race or play around with creative new skill combos (and inflating their TV in the process) is bucking the entire system. That's creating a tradeoff between "fun innovation" and "looking like a coach to be feared" by the metrics in question.
I know I'm not the first with this idea, but I would suggest it might be interesting to tackle this kind of issue not by changing anything about the current system, but to add a different layer on top that optimizes for a different function. For the sake of argument, let's call it the Shame Index. Er, I mean, the Challenge Score (CS). This could be a metric that takes into account things like
1) playing matches in some of the ways deemed elsewhere in this thread to be desirable - teams with 2+ seasons playing a teams in their 1st season would suffer, for example.
2) Playing coaches above/below your current CR.
3) Playing as/against lower tier teams.
4) Considering recent team performance (say, average of last season performance and current season prorated, or maybe just "last 10 games") - this would try to be an indication of how well constructed a team currently is.
5) Historical matchup win rates by race (and perhaps ideally by TV as well)
No doubt other things could be added. CS could be displayed near CR on coach pages, and on the game finder in a similar manner.
The point of this would be to let everyone looking for a match see "Ah this is a legend coach playing a top tier race... but it seems they do a lot of picking on new coaches. Interesting!" It would also let those looking for games see "Ah, this coach is playing snotlings with moderate success and middling CR - but look at that CS - they spend a lot of time seeking hard matches and are probably a dangerous opponent!"
The point of doing this would be: there really is not any compelling reason why people shouldn't be allowed to only play the best races and seek to win a lot by choosing favorable matchups in an open environment. That's the point of an open environment. But it also seems reasonable for the site to provide more information to those seeking games about their opponent. This is especially new folks who can't just glance at some teams and numbers and figure this out for themselves.
The advantage of doing things this way is that it targets those folks who actually care about things like rankings and are being assholes in an attempt to game themselves higher. Those who don't really care much about CR - well, they can ignore CS too. But for those that are optimizing CR/wins at all cost, doing so using some of these silly tactics would then tank CS. Basically, you can't optimize play for both unless you take a lot of hard matches and do well in them.
Naturally, CS could and presumably would be gamed too. But what I think a metric like this might do is: give people reason to take harder, more interesting matchups. It might make people stop and think about match fairness more when selecting a match. And doing that feels like it could directly combat some of the most blatant sort of abuses detailed in this (and many other) threads.
TL;DR We can't stop munchkin pickers, but we can make them squirm a bit! |
_________________
Come join us in #metabox, the Discord channel for HLP, ARR, and E.L.F. in the box!
 |
|
|
|
| |
Gnomes are trash
Search
