*cap thoughts for next DT

[RageRitual]

I keep seeing people complain about the * cap, increase the star cap, every season.

Obviously this season we should have had more teams, but what is done is done.

I'm not sure about duel leagues but at least in base, everyones ratings are way too inflated.

A few suggestions with the *cap:

1) Having x rating in dt used to mean something. For example one rating was for a pubber, another for a twd'r, another for twl contender, etc. We should have clear reasons behind the ratings to avoid all these opinion-based ratings. Maybe not really using the old system but a similar system in place might help avoid a lot of the ego and opinions on individual ratings.

2) only x amount of players should be allowed the highest few ratings in each ship, and go from there.

3) what might be easier instead of clearing everyones ratings out, increase the highest rating from 10 to 11 or 12 and increase the *cap as needed with it.

Just a few quick examples of inflated ratings:

hulk is only worth 0.5* more than markmrw or a2m?

sarger and spezza are only a half star better than lockdown?

dwopple and joe half a star better than morph?

I remember suggesting half stars awhile back, they got implimented and it helped, I honestly think having a wider range for ratings might help too.

-EDIT- Just a closing thought here, two 6* players are not worth an 8* let alone a 10*. Mathematically they are worth 12, so why is 10* the cap.

[MousE]

Sorry Dwop

[viop11]

v1nt1g1

[apt]

i think my first dt was just a couple years ago and i remember like rly having a lot of issues with star system. its flawed. especially when it comes to high tier rankings like 9 star and above.

i think once ur in ten zone, ur implying u have a player that can play all 3 ships at an elite level.

having a 10 star spid who can only play 1 ship sufficiently, or a 10 terrier, 10 shark is wild. its too late now to be revamped i suppose, but off the bat a star system in base would of had to have been built on a per-ship-basis on not a blanket star ranking, which can only work in a 1 position/ship sport/game like javs and wbs.

then you're also completely relying on subjective observation of players, which is v much reliable on hearsay, groupthink and confirmation bias. not a good system tbh.

MMR has its flaws ofc, like weaker players constantly being carried by better players because theyre friends, thus inflating a weaker players rating. but its the lesser of two evils and a much better alternative.

[Dreamwin]

Star cap has worked decent for the most part when it has been set high enough. Caps that are motivated to win will often put time and effort into developing low stars in their teams because the lowstars are often the key part which wins you titles in this format(or to put it better: dont cost you titles). This is very healthy for the zone in general.

The issues are just highlighted this season when the expectations didnt meet the reality in terms of player activity. It has also been too long since the last high quality league and I think we could really use a TWL after this one.

Despite this, I feel tempted to try going old school with:

Dueling:

Max 10*
Max 10*
Max 9*
Max 8,5*
Max 7.5*

Base Max 10*
Max 10*
Max 10*
Max 9,5*
Max 9*
Max 8,5
Max 8*
Max 7,5*

We could also limit the amount of 10* and 9* stars that can be drafted in 1 team / 1 division. This would help to spread top tier talent evenly. I think this part is something that havent been suggested before(?). Of course it would make things more complicated but experienced caps can definitely handle such restrictions.

[apt]

Star cap has worked decent for the most part when it has been set high enough. Caps that are motivated to win will often put time and effort into developing low stars in their teams because the lowstars are often the key part which wins you titles in this format(or to put it better: dont cost you titles). This is very healthy for the zone in general.

The issues are just highlighted this season when the expectations didnt meet the reality in terms of player activity. It has also been too long since the last high quality league and I think we could really use a TWL after this one.

Despite this, I feel tempted to try going old school with:

Dueling:

Max 10*
Max 10*
Max 9*
Max 8,5*
Max 7.5*

Base Max 10*
Max 10*
Max 10*
Max 9,5*
Max 9*
Max 8,5
Max 8*
Max 7,5*

We could also limit the amount of 10* and 9* stars that can be drafted in 1 team / 1 division. This would help to spread top tier talent evenly. I think this part is something that havent been suggested before(?). Of course it would make things more complicated but experienced caps can definitely handle such restrictions.

fair pts, but the last pt probably just provides more difficulty. as players can't show, won't be as active enough, esp as everyone gets older and everyone is pumping out babies now* and ish.

prob this is the best we can do, ppl fieldin many complaints but end of day y'all've kept a 20 year old game at the brink of death alive.

i am keen on seeing dwopples MMR strategy tho, i'm a fan of automation over human error. at this point in the game a self-regulating system of competition is the best chance of keeping us moving fwd. similar to how twd and allowing widespread borrows has kept the game alive, just wish we had that option a long time ago.

[apt]

also super strange question, but. what's wrong with just abolishing a rating system for TWDT all together. whether it's MMR or Star Rating. There are less and less players wanting to get into the fray and the ones that are interested are satisfied enough playing TWD matches here and there when they get the chance. Esp in basing most of the super new players get frustrated at their poor performances in TWBD that they request to be subbed, they ain't even trying to play a league setting any time soon.


why not just run a straight draft without ratings.

[zidane]

too many people tiptoed with signing up until the last minute. Another major factor is that so many people said they would barely show, which hasn't been the case. there's no doubt 8 (or even 10) teams would have been possible and these star caps would seem less egregious. There is no precedent of redoing a draft or changing the cap after the season has started though.

having said that I definitely preferred the 69* cap. The rating inflation is real, the reality is there are no new players so far more ratings are increasing than decreasing. The goal should be to be able to add 8 active and properly rated players. This season is just a combination of factors making it so each team is being forced to bench several good players in each league

[Zeebu]

also super strange question, but. what's wrong with just abolishing a rating system for TWDT all together. whether it's MMR or Star Rating. There are less and less players wanting to get into the fray and the ones that are interested are satisfied enough playing TWD matches here and there when they get the chance. Esp in basing most of the super new players get frustrated at their poor performances in TWBD that they request to be subbed, they ain't even trying to play a league setting any time soon.


why not just run a straight draft without ratings.

i dont see newbs just requesting to be subbed. typically they get endlessly berated for being bad by all the good players who want to see them subbed out.

[Jones]

i dont see newbs just requesting to be subbed. typically they get endlessly berated for being bad by all the good players who want to see them subbed out.

Zeebs is awesome.

Mouse was so mad he told me to start killing repless sharks in mid just to improve my stats. He is a twlb champion.

Edit, i also agree with mouse that basing draft should be eliminated and just use twlb squads instead. It would kill the zone faster, naturally, but think about the possibilities here.

[Exalt]

too many people tiptoed with signing up until the last minute. Another major factor is that so many people said they would barely show, which hasn't been the case. there's no doubt 8 (or even 10) teams would have been possible and these star caps would seem less egregious. There is no precedent of redoing a draft or changing the cap after the season has started though.

having said that I definitely preferred the 69* cap. The rating inflation is real, the reality is there are no new players so far more ratings are increasing than decreasing. The goal should be to be able to add 8 active and properly rated players. This season is just a combination of factors making it so each team is being forced to bench several good players in each league

While it would take a lot of time, would it really hurt if a large group of players (not just like 3 or 4 guys on a stream, but not 100 bozos trolling either) redoing the * ratings in a deflationary manner before next TWDT happens?

If we took intelligent players from each respective league and had them rate people in that league based on clear guidelines that started with something like 6* = player who doesn't play this ship and has no real intention of getting better at it and 10* (or higher) = top 1-3 player in this league in the game, would it really hurt? I'd go even further and make it a true 1 to 10 system, where 1* is Cyclone and 10* is a Hall of Famer. The majority of the playerbase would be at 5* instead. Adjust the * Cap from there.

The issue I had with the last rating adjustment I saw (the stream from last season's TWDT) was that it was only like 3 or 4 people doing ratings, and for instance, Racka would rate someone in WB and Ogron would overrule him. Setting aside Ogron's ability in WB or intelligence level in that league (which are admittedly higher than average), Racka should never have been overruled in any situation whatsoever in WB unless it was really egregious. He was though. This happened in other leagues as well, because groupthink and biases and such would make people cave to their peers who almost certainly have less knowledge in that league than the main guy who was there to represent it. This from the very start caused doubt in the ratings, and since ratings didn't really change at all this season, it makes it even worse now.

On top of that though, Racka alone cannot accurately rate every single WB (or Ogron rating every baser, etc) out there, since that's expecting too much. Basing things on past TWDT or TWL seasons alone is also not working, as player skill is not a flat line that stays the same year to year. There's very few players in the game who maintain a high-level of skill in the game for all this time. Most players either get better or worse. Just look at Riverside being an 8.5* WB as an example. In what universe is this really true outside of the one season Sweet won TWL 10 years ago? How is it possible that once a player reaches a certain * level, they never decrease to where they accurately are years later?

The other issue as I see it is that TWDT is always drafted in a way where 3 leaguers matter more than the best 1 leaguers. I still have never understood why this is the case, as less than 5% of the population is even competent at all 3 leagues, let alone a top player in them. Many basers out there dislike dueling leagues and many dueling league players dislike basing. The leagues are so different in the skillset and knowledge level that any normal player moving from one league to the other usually lowers the quality of those games dramatically, yet due to * caps, every TWDT season has this happen just so rosters can work. The low * player from another league dislikes being forced to play in the other league, while they may not even get a chance to play in their favorite league because the * caps force through lower * players from the opposite league to take their spot. How is this fun to anyone on any side? The top players in that league don't enjoy playing with bad players who don't understand the nuances of their respective league, while the lower * player dislikes having to play as well. It isn't fun and has no real reason to exist.The only reason this was ever set in place back in 2006 or whenever the first TWDT season happened was to get "new" players to enter leagues. We all know for a fact that new players don't exist at this point in this game's lifecycle. Why is this being catered to now? It isn't fun, and that's the only reason to play this game to begin with. Fun. Even when there is money being donated by Wirah or Dad, it isn't enough money to matter unless you live in some African jungle like Zapata.

Separate the dueling/basing leagues into different drafts, get rid of the low * cap that caters to "new" players that don't exist (6* players should not be required to have lineups work in any game), and have a larger committee of relevant and smart players in each league to redo the * ratings with deflationary numbers that better match skillsets. .5 * difference isn't enough for a lot of player skill gaps, and frankly, a lot of players are rated too highly for their current skill level. There are way more overrated players than underrated at this stage, and it really hurts the league and those players themselves in many cases.

Edit: Also, if the MMR thing actually works as intended in the future, we can try a season with that as the basis as well. That remains to be seen though, so it's not really realistic to suggest it yet.

[Jones]

While it would take a lot of time, would it really hurt if a large group of players (not just like 3 or 4 guys on a stream, but not 100 bozos trolling either) redoing the * ratings in a deflationary manner before next TWDT happens?

If we took intelligent players from each respective league and had them rate people in that league based on clear guidelines that started with something like 6* = player who doesn't play this ship and has no real intention of getting better at it and 10* (or higher) = top 1-3 player in this league in the game, would it really hurt? I'd go even further and make it a true 1 to 10 system, where 1* is Cyclone and 10* is a Hall of Famer. The majority of the playerbase would be at 5* instead. Adjust the * Cap from there.

The issue I had with the last rating adjustment I saw (the stream from last season's TWDT) was that it was only like 3 or 4 people doing ratings, and for instance, Racka would rate someone in WB and Ogron would overrule him. Setting aside Ogron's ability in WB or intelligence level in that league (which are admittedly higher than average), Racka should never have been overruled in any situation whatsoever in WB unless it was really egregious. He was though. This happened in other leagues as well, because groupthink and biases and such would make people cave to their peers who almost certainly have less knowledge in that league than the main guy who was there to represent it. This from the very start caused doubt in the ratings, and since ratings didn't really change at all this season, it makes it even worse now.

On top of that though, Racka alone cannot accurately rate every single WB (or Ogron rating every baser, etc) out there, since that's expecting too much. Basing things on past TWDT or TWL seasons alone is also not working, as player skill is not a flat line that stays the same year to year. There's very few players in the game who maintain a high-level of skill in the game for all this time. Most players either get better or worse. Just look at Riverside being an 8.5* WB as an example. In what universe is this really true outside of the one season Sweet won TWL 10 years ago? How is it possible that once a player reaches a certain * level, they never decrease to where they accurately are years later?

The other issue as I see it is that TWDT is always drafted in a way where 3 leaguers matter more than the best 1 leaguers. I still have never understood why this is the case, as less than 5% of the population is even competent at all 3 leagues, let alone a top player in them. Many basers out there dislike dueling leagues and many dueling league players dislike basing. The leagues are so different in the skillset and knowledge level that any normal player moving from one league to the other usually lowers the quality of those games dramatically, yet due to * caps, every TWDT season has this happen just so rosters can work. The low * player from another league dislikes being forced to play in the other league, while they may not even get a chance to play in their favorite league because the * caps force through lower * players from the opposite league to take their spot. How is this fun to anyone on any side? The top players in that league don't enjoy playing with bad players who don't understand the nuances of their respective league, while the lower * player dislikes having to play as well. It isn't fun and has no real reason to exist.The only reason this was ever set in place back in 2006 or whenever the first TWDT season happened was to get "new" players to enter leagues. We all know for a fact that new players don't exist at this point in this game's lifecycle. Why is this being catered to now? It isn't fun, and that's the only reason to play this game to begin with. Fun. Even when there is money being donated by Wirah or Dad, it isn't enough money to matter unless you live in some African jungle like Zapata.

Separate the dueling/basing leagues into different drafts, get rid of the low * cap that caters to "new" players that don't exist (6* players should not be required to have lineups work in any game), and have a larger committee of relevant and smart players in each league to redo the * ratings with deflationary numbers that better match skillsets. .5 * difference isn't enough for a lot of player skill gaps, and frankly, a lot of players are rated too highly for their current skill level. There are way more overrated players than underrated at this stage, and it really hurts the league and those players themselves in many cases.

Just start twl

[Turban]

way too much inflation in ratings based on personal bias. you honestly can't have this many 10-stars for instance. here's some ratings quickly adjusted based on my opinion, and yes i don't give a fuck if someone is butthurt about losing half a rating or more. star cap should be based on average rating of top 70 (or 75) players in a 6-team format which in this instance would have been somewhere around 69 to 69.5 star cap.

https://puu.sh/JazSm/049ea2860a.png

[Dwopple]

Seeing a lot of interesting and thoughtful points in here. The one bottleneck not being mentioned is the very real state of the availability of our already small volunteer development team.

Theorizing too much drastic change from the tools and formats we have today may be very likely to never come to fruition. This level of theory is awesome, and I only bring it up as a preemptive caution and to set expectations.

This is not a knock on our devs but more a reminder people's personal lives should come first and most of what gets created happens during their evenings or weekends. Change in any amount is likely to be slow moving.

Exalt's ideas may very well be possible with the current tools with minor tweaks, but we'd need dev input.

I also do not have much optimism on the scale of input being hypothesized, while admittedly having a pessimistic biasness from the last 2+ years lol.

One last thing, maybe 2. The MMR idea is novel and exciting (and not my brainchild, please thank qan, BIET, and Rab for the current direction of this to be hopefully tested in a TSL format at some point), but is also running into the same bottleneck explained above. Real life should always come first.

That said, there is some planning and discussion happening on "what's next" to address much of what's being discussed in this thread but will let the person whose idea it is share the specifics when ready.

Long live TW

[Riverside]

That said, there is some planning and discussion happening on "what's next" to address much of what's being discussed in this thread but will let the person whose idea it is share the specifics when ready.

Please, I ask and recommend that your team continue to seek community input, in various and authentic ways.

Ideas are great, but:

Along with collecting and considering opinions for qualitative reasons, polls can help indicate public perception sometimes, despite perceived voter fatigue, the margin of error re accuracy, etc.

I dislike stealth mode planning for these Leagues. Unless im missing something: some reasons to share early and often:

• Limiting re-occurances of what caused the last 2-3 month delay between TWDT seasons, despite the post-season HYPE that came with last TWDT
• Recruiting volunteers
• Overall community engagement (which helps with activity)

Dwopple,

I appreciate that you've continued to listen to many voices, even when there are disagreements with a set of choices or decisions.

And we have disagreed; publically and privately. As people with different view points and lived experiences, it may happen again. Imo, that can be a helpful tool for communal growth.