TWDT Season 28 Announcement

[LeeRomeno]

Almost 2 weeks since the announcement, I'd like some more detail:
- how many of each tier can a team have?
- what is the "change of game order"?
- are we meant to have separate base and duelling signup lists?

there is separate talk about tiers.
Regarding change of game order. This is really something that i have heard in many discussions, that we should maybe try some league that does not have wb-jav-base order. There are arguments that base games are late and many euro basers have hence stopped playing or have issues showing for those games. As base is anyways in trouble, i had an idea of going for jav-base-wb order. I do realize base games take most time, but moving them a bit earlier could perhaps increase some activity there. Anyways, this is just an idea, any feedback on it, is welcome.

As for separating wb/jav and base. I think last tw cup showed it was very successful change. First of all it gives cap chance to draft purely based on what he thinks the freq needs, rather than trying to balance in finding those multishippers rather early on and leaving specialists later. Secondly people can sign up to exactly what they want and it gives better understanding for staff as well, how many squads are viable for each category. There are very selected few pilots who bother mentioning they only want to play this ship or that, but it usually happens when league goes on and people just dont show (either early to be around for wb, or leave before base).

Most probably due to technical issues and things going slowly we need to move back the starting dates a bit as well, to give more time to prepare.

[zidane]

Draft was my hypothetical example looking at MMR ratings, the player doesn't matter. You're still proposing to make the #1 player the same cost as the #16... who do you think will be #16 in jav/wb?

One team will have Mythril costing the same in wb as Zapata or Cripple or JZ or whoever (all top 15 in WB Cup). but have no fear Zapata's captain will have first crack at the 7 and 8s to make up the difference!!! Good luck man lol. extremely delusional if you think that's an even trade.

you've lost the plot completely too, the 9 and 10s are almost never being benched in DT too. It happens sometimes in Base if a captain has drafted horribly. It's usually the mid rated players or 6s who are benched, especially the inactive ones.

[LeeRomeno]

Draft was my hypothetical example looking at MMR ratings, the player doesn't matter. You're still proposing to make the #1 player the same cost as the #16... who do you think will be #16 in jav/wb?

One team will have Mythril costing the same in wb as Zapata or Cripple or JZ or whoever (all top 15 in WB Cup). but have no fear Zapata's captain will have first crack at the 7 and 8s to make up the difference!!! Good luck man lol. extremely delusional if you think that's an even trade.

you've lost the plot completely too, the 9 and 10s are almost never being benched in DT too. It happens sometimes in Base if a captain has drafted horribly. It's usually the mid rated players or 6s who are benched, especially the inactive ones.

I have no idea who will be the top 16 player in jav/wb, it really depends on signups, the more top players sign up, the higher quality that player will be.
You are mistakening tier slot being anything to do with "cost". Its alternative way of trying to balance the freq. Surely, whoever gets myth, is lucky, but he will also get that 16th player to back myth up, while whoever picks 8th will get 8th and 9th. Also you need to think about caps. If cap is rated top tier, this means that a) they will most likely not get first pick as there is handicap towards mid-tier caps b) they will not get second pick, as there are 2 top tier slots per freq.

So in a nutshell, its alternative way how to balance freqs, but its still a draft, there is order and that order itself already balances things out.

[zidane]

your explanation makes no sense, how are you going to ensure the #1 is paired with the #16, or the #8 with the #9? you're going to reorder the draft as the picks are done for every single round to somehow reach lineup balance to solve the problem you're introducing? you're creating way more problems than you are fixing and this sounds like a complete trainwreck.

your whole system implies that the captains will pick the best player available at each pick, and even then it's not going to work. for all we know pick #6 and #11 could be 19* total whereas pick #7 and #10 could be 17* and then that captain is pretty much doomed to fail.

[LeeRomeno]

I think representing this tiered system as moving away from the star cap is a pretty huge misunderstanding. There will still be a cap since the flexibility of lineups is something that will be preserved. Call it a salary/value/star/tier cap - it doesn't matter, there will still be a limit to how much value you can add to any lineup.

The only differences I see in this new proposal is we're replacing the word star with tier and reducing the number of tiers from nine to three - that is the problem. I'm not going to argue that nine is the perfect number to represent different levels of skill and impact in subspace, but I'm 100% certain the number is not three.

Placing players into three tiers would not make ratings more accurate, it would do the complete opposite. The skill gap which used to be represented in more than three tiers will not magically disappear just because the additional tiers to place people in longer exists. There is an obvious difference worth representing and balancing for in the rating system between a 9* and 10*. Same as 7.5* and 8.5*. Someone being .5* off in their rating has been enough to give a lineup an edge in the star cap era. If someone is a full 1* off, that's considered criminal around here. Yet this system has a 1* difference baked into every tier. How can you create an even playing field for all teams in a capped system if a 10* player, who has a higher expected output and impact on winning, pays the same cost towards the cap as a 9* player?

That's why I also don't see how this makes the league more inclusive. There is a massive disadvantage for players who arbitrarily sit at the lower end of each of these three tiers. They have to compete against players on their own team for playtime who may sit on the higher end of a tier, and then even if they do get playtime, they're expected to outperform their rating by what was previously considered a two-tier skill gap.

Playing lineups which are 2* below the cap has been a great way to lose games in TWDT. A three-tiered system would misrate more than half the playerbase to the point a 2-4* difference between lineups will not just be possible, it's going to be guaranteed. It's also a mathematical certainty that every team will not be able to meet the star number of the best drafted team in a snake draft. We now have more data than ever on who impacts winning with thousands of games played in MMR. I'm not sure why we would move to a system that implies we actually know less than ever about the differences in impact between players in the game, specifically for warbird and jav.

Thank you for feedback, as i replied already to zidane, I do not see this as just renaming star to tier. Its another system how to balance the gameplay, with separation of duelling and base, we will have larger number of duelling teams and smaller number of base teams (but with more people to have subs available).

With less people per team (this needs to be of course balanced, so we dont face no--shows), lets say 8 or 9, even the lower rated players within the tier have a chance to fight for spot. That fight is imho more fair, then purely based on someones opinion that you are 6 6.5 or 7. Last draft league we had yogi as 7, sunny d and tj hazuki as 6. Yogi had 0 games, sunny and tj both managed to play in 6 games. They are all in reality in same area in mmr ratings, similar in contribution to game as well, but 2 players were lucky enough to have that 6 vp, while 1 was forced on bench.
Same thing in mid-tier ratings. Banzi (rated 7.5) got 15 games on power vs rojo (rated 8), who got 2 games in first week. Banzi may be better player anyways, but that flawed rating basically guaranteed rojo, not to really have any gametime in league, with tier system, its up to him to prove he can do better.

Systems have tendency of trying to optimize and go too precise over time. We need to keep in mind that this is ancient game with diminishing playerbase, none of us (after soul left) are doing this professionaly. We may have mmr bot that optimizes the freqs and gives 55% chance of some freq winning, but the game can still end lopsided with 50-30 score for that 45% just because someone didnt feel like playing. We can try putting all the data we have and trying to figure out that exact value of some player, but it will never be precise. Its not even precise in sports like football, where buying player for 100+ mil can just as easily turn into massive flop, although there was team of 20 data analysts behind the purchase analyzing every muscle movement of the player for 20 years before. The tier system will divide the players into tiers and generate the competition for spot on freq within the tier, its not perfect of course, but i find it still more honest towards the players vs 2-3 guys in stream rating you and deciding if you have a guaranteed spot to play or 0 chance to play. You have 2-3 players in your tier on your squad you need to be better, id take that chance every day.

[LeeRomeno]

your explanation makes no sense, how are you going to ensure the #1 is paired with the #16, or the #8 with the #9? you're going to reorder the draft as the picks are done for every single round to somehow reach lineup balance to solve the problem you're introducing? you're creating way more problems than you are fixing and this sounds like a complete trainwreck.

your whole system implies that the captains will pick the best player available at each pick, and even then it's not going to work. for all we know pick #6 and #11 could be 19* total whereas pick #7 and #10 could be 17* and then that captain is pretty much doomed to fail.

First of all, as we have separate teams for the leagues, each cap can focus on picking the best lets say wb player on list. We have top 16 pilots rated as top tier. Out of those lets say 5 are caps. The 3 caps who are mid-tier, get handicap, those 3 will then pick (hopefully) the 3 best wbs in draft. Then the top tiers will get their 1 pick (as they are themselves already top tier) and round is finished by the 3 mid-tier caps getting their 2nd top tier player.

I dont think argument of caps doing horrible draft really matters here. Exactly same has happened in several draft leagues with vp system before. It is always possibility and i really hope we get strong caps who avoid mistakes like not picking lets say myth or ease and letting racka get him in 4th round. But system itself will try to balance the order so that the game-deciding players are spread out as equally as possible.

[LeeRomeno]

I do realize that there is alot of confusion as several elements of the league have not been posted properly. I will try to finish the first draft of rules for draft and setup by end of the week, so it will be more clear.

[Tiny]

Lets talk logically about this.

If we go to 3 tiers, this makes the top of those tiers the most important and valuable players. Those people will get the most playing time and be picked first and rightfully so. With the system we are about to do, (using old system ratings for clarity) 7* 8.5* and 10*s will be the cream of the crop. Having 3 tiers won't change the fact that everyone knows who is better than everyone else and lumping 3 tiers into 1 will just hinder captains from making up star power differences. People will still get benched and playing time will stay the same, except teams will be even more heavily skewed. The larger tier system allows for more flex in lineups, for example, i was capping as a 9* before and able play more mid tier players, where 10*s had to play more lower players, but in the end made it pretty even. I will not cap as a 9* against 10*s, as I would be a top tier player in this new system and immediately be at a huge disadvantage.

[Rab]

No hate but I'm struggling to understand Lee's posts tbh. Don't know if it's me or if English isn't his first language. Bit of a problem when trying to discuss stuff where the detail matters.

I agree with zid/Tiny/others - star system is better than tier system - and everyone already understands it. Props to Tiny for writing clearly.

I think separate drafts is against the theme of twdt - being able to draft multi-ship wisely is kinda the point - but I'm not against trying it.

I think moving base earlier is a good change. If you're having separate drafts you need to keep jav and wb next to each other so the duellers aren't waiting around with a big basing gap between their events.

I'm fine with shorter seasons, I wasn't a fan of that in cup, but it seemed that most people preferred it.

Taking on running a league always results in getting hated. Making a big change even more so. Making 4 big changes is league-running on insane difficulty.

I suggest making just 2 changes (or even just one of these): short season + move base earlier (jav-base-wb or base-jav-wb), and give up on the other changes.

I also suggest banning wiibimbo from this forum, he doesn't even play the zone but shows up in every thread criticising ppl.

[wiibimbo]

No hate but I'm struggling to understand Lee's posts tbh. Don't know if it's me or if English isn't his first language. Bit of a problem when trying to discuss stuff where the detail matters.

I agree with zid/Tiny/others - star system is better than tier system - and everyone already understands it. Props to Tiny for writing clearly.

I think separate drafts is against the theme of twdt - being able to draft multi-ship wisely is kinda the point - but I'm not against trying it.

I think moving base earlier is a good change. If you're having separate drafts you need to keep jav and wb next to each other so the duellers aren't waiting around with a big basing gap between their events.

I'm fine with shorter seasons, I wasn't a fan of that in cup, but it seemed that most people preferred it.

Taking on running a league always results in getting hated. Making a big change even more so. Making 4 big changes is league-running on insane difficulty.

I suggest making just 2 changes (or even just one of these): short season + move base earlier (jav-base-wb or base-jav-wb), and give up on the other changes.

I also suggest banning wiibimbo from this forum, he doesn't even play the zone but shows up in every thread criticising ppl.

Next you will tell me to die because I have a heart condition, right? LOL

[Dwopple]

Next you will tell me to die because I have a heart condition, right? LOL

Wiibimbo, please stop derailing this discussion. You dishing out personal vendettas isn't necessary or welcome in this thread. Thank you.

[wiibimbo]

I gave you constructive criticism on the last page but I guess your tiny brain didn't see it, eh Dwop? Stop acting like you are somebody and leave us normal people alone.
PS. I don't think you know what a personal vendetta is. I'm just here having some fun vsing people whose IQ is below 50. So, pretty much everyone lol (except Mythril).

And for your information, the last time I got banned is because somebody else did something to me and not vice versa, so your credibility in saying anything to me is compromised. I wouldn't trust with anything what you fuks say here even if you were the last people on Earth. So keep playing your illusionary draft league for 20 people and keep telling yourself how good you are but know this: NO U CAN'T GET PREGNANT IF U CUM IN GF'S ASS.

Wii out

[Dwopple]

I gave you constructive criticism on the last page but I guess your tiny brain didn't see it, eh Dwop? Stop acting like you are somebody and leave us normal people alone.
PS. I don't think you know what a personal vendetta is. I'm just here having some fun vsing people whose IQ is below 50. So, pretty much everyone lol (except Mythril).

And for your information, the last time I got banned is because somebody else did something to me and not vice versa, so your credibility in saying anything to me is compromised. I wouldn't trust with anything what you fuks say here even if you were the last people on Earth. So keep playing your illusionary draft league for 20 people and keep telling yourself how good you are but know this: NO U CAN'T GET PREGNANT IF U CUM IN GF'S ASS.

Wii out

The "normal" people in this thread are having productive discussion. You let your pride get in the way for the 3rd time. You can continue your need to flame me, Rab and vys to discord or in game. Thanks for understanding.

Thanks everyone for keeping the discussion going. Clearly some polarizing changes, which is understandable.

I'm interested if anyone has some outside thoughts or ideas on improving the subjectivity of ratings, more importantly the misratings. I do think misratings in the lower-mid tier players has an extremely adverse effect on playtime. We do want to improve the match quality without forcing captains to shove a last round 6* in because of player(s) being misrated.

No doubt there's a lot of MMR and previous season data to refer to. Do any ChatGPT or statisticians have interest in volunteering to raise some objective findings?

Thanks again for the lively discussion. Much appreciated.

[zidane]

A good way to prevent needing those random 6* who don't play the game is to have a higher star cap. I believe we ran 43/43/69 a couple seasons ago and I'm fairly sure there was a lot less of that going on as a result. 1 star can make a massive diff for that stuff

[Rab]

I'm interested if anyone has some outside thoughts or ideas on improving the subjectivity of ratings, more importantly the misratings. I do think misratings in the lower-mid tier players has an extremely adverse effect on playtime. We do want to improve the match quality without forcing captains to shove a last round 6* in because of player(s) being misrated.

No doubt there's a lot of MMR and previous season data to refer to. Do any ChatGPT or statisticians have interest in volunteering to raise some objective findings?

Is there a sheet of current star ratings somewhere which doesn't include tax?