Proposed ratings changes

[Steadman]

I splooged my pants reading this thread

[bugga]

Something else I wanna say. People suddenly want Babe Ruth to be higher in wb/jav. Last season he played some with me on Blind and he was a total 'failure'. Remember that he is with a team that carries his ass hard. When a team wins average 50-30 a 6 star doesnt often fall out. When it comes to changing ratings you have to keep in mind how a player would have done on another team.

Why did you add me a medal for that league? I think your blind irrational hatred has grown after the league.

Babe Ruth played well in TSL recently:
http://www.trenchwars.org/tsl/index....layer&id=60456

People are right to advance me to a seven, absolutely. I certainly want to stay as a 6 to get more playtime. If I was set up as a 7, I wouldn't bother showing up.

[Voth]

I think all the ratings were fine and shouldn't be changed.

Don't forget - there are only five ratings: 6*,7*,8*,9* & 10*

That means there is a 20% distribution in skill within each rating.

The expectation that everyone who is rated, for example, a 6* will play the same is statistically naive.

There will always be a range of ability within each rating - and it's up to captains to sniff out that value. That's what makes captaining fun.

this

[Ricko]

I got massively bummed this year as captain, but all in the rules but I wont be Captaining again unless their are some changes to the picking, its a shame because I have felt the strongest ive ever been in warbird and ended up with only a + 9 rating over deaths in my favourite ship to show for it.

sorry bud

[Attacks]

I got massively bummed this year as captain, but all in the rules but I wont be Captaining again unless their are some changes to the picking, its a shame because I have felt the strongest ive ever been in warbird and ended up with only a + 9 rating over deaths in my favourite ship to show for it.

[MHz]

I think all the ratings were fine and shouldn't be changed.

Don't forget - there are only five ratings: 6*,7*,8*,9* & 10*

That means there is a 20% distribution in skill within each rating.

The expectation that everyone who is rated, for example, a 6* will play the same is statistically naive.

There will always be a range of ability within each rating - and it's up to captains to sniff out that value. That's what makes captaining fun.

[Rab]

Isn't that exactly how it works now?

[Vys]

I think splitting the sections into tiers would be the way forward.

Tier 1 - 10* and 9* = 2 can be played in warbird and javelin and 3 can be played in base
Tier 2 - 8* and 7* = 2 can be played in warbird and jav and 4 can be played in base
Tier 3 - 6* and 5* = 1 can be played in warbird, javelin and base.

Use the end of season statistics to calculate players ratings.
1 - 20 = Tier 1
21 - 60 = Tier 2
61 - = Tier 3

We could look in more details about this and which tier is the cut off point.

[Omega Red]

There’s always gonna be those few guys who have improved or just got slept on and seem broken in hindsight.

Taking into consideration if adjusting ratings after a season will affect the players ability to play in the future should be irrelevant. This depends entirely on factors that happen in the future such as how your captain drafted at your slot and the show rate of other equally rated teammates. The main factor in adjustment should be performance in my opinion.

The alternative leads to apathy from opposing captains and correctly rated players and a degenerate league experience.

It’s a hard thing to do to be able to correctly evaluate players especially now that there’s less activity and visibility. I know everyone wants to just show up on sundays and play and have fun including myself but it’s a balance.

[ronaldinho]

I'm aware my voice don't count or almost, and I understand it, but still, I go with Ogron. Major Crisis is highly capable, he proved being a good staff, a good captain and a good player. If we go from Henry Saari to Major Crisis, we improve the league. He played a part on sign ups, on ratings, on stats and follow-ups, he is motivated and wouldn't scare to tell captains they can't do this or that.
I go with Mythril too. While I understand Unlimited's feeling, us newbies have many weapons to improve. I received shittalk since day one, and while I still don't understand how a total newbie can be shittalk that much, the situation now is totally different. I have many weapons to improve, TWD, Elim, and all other 2nd tier leagues, I don't use them, so don't improve and so understand why on a DT I have no chance to play.

[Mythril]

Just some quick rating changes that I feel are obvious:

- Steadman 8/10/8
- Beam 10/6/6 (I actually don't feel he's broken at the 6* in jav/base. He would get no playing time in any other spot, maybe 7* as a shark. Definitely 10* wb after the insane season he just had)
- Lews 8* warbird
- Rodney 9* shark at a minimum. He's a huge luxury for me at the 8* in base.
- DW needs to be dropped back down to 8* in warbird. With more warbirds like racka, myself and attacks capping, 9* is a big disadvantage for his warbird team.

As already stated, the only reason we had any kind of updated ratings was because major crisis took initiative a couple days before the draft. Myself, zid and major skimmed through and revised as accurately as we thought possible. I'm in touch with warbird, zid with jav, and major/zid are really in tune with all of the basers. It worked out pretty well for the most part. The most questionable ratings came from FA (Lews is not a 7*), as those ratings were being debated literally 5 minutes before the draft. That was the only time any abuse could have really happened considering captains knew their drafting spot and which positions they needed filled in their rosters. Even then, all of the caps were online at that time.

And still, it's a lot of guesswork. How do you rate someone who's inactive? Or how do you rate inactivity in general? It's a monster variable in TWDT. Turban for example, is a top 2 pick in any season if he's active. But what if you're at the #2 drafting spot? He's not worth drafting at #2 in a snake draft because you still have a wide selection of cornerstone, high impact multi-leaguers at that spot who you won't have a chance at when the draft snakes back to you. We also can't force Turban to commit to an exact amount of games played considering we can't manufacture interest or availability in subspace for anyone. Every captain has to weigh these things and play the odds. If you're on a chat with Turban as RaCka is, it's obviously going to have an impact on his showrate. But then do we penalize racka for being friends with turb? Or DW for being in contact with all of the inactive euros? I got lucky when I drafted TJ hazuki and he said he had rodney's facebook and could get him to logon, so I stole him in the later rounds when everyone else assumed he wouldn't show. These things will always happen, but as captain, part of the fun is in trying to play these strategies to find any small advantages. When you look at the wins throughout the season, things played out very evenly with 4 teams fighting for their playoff spots during the final week.

Ratings will always need tweaked and it can't be a zero-sum game considering players don't improve/decline on an even basis. There's still an advantage in having an eye for up and coming talent. Pineapple express for example is imo on the verge of being very good, maito_chi is a bit 'broken' at the 7*, fatrolls is an 8* minimum if he shows, which he never did. But someone like fatrolls is also unproven, so it's unfair to grant him the 8* without a season of evidence. Steadman was also in that position. Not until this season would he have been considered an 8* in warbird.

Things won't ever be completely fair or all inclusive as everyone would hope for. There's too many moving parts and the population is squeezed into a strict amount of teams. TWDT is the new face of the competitive scene, so there's no throwaway games where it's not in the cap's best interest to add players for their own enjoyment. With only 6 teams, many players will be stuck on the bench for entire seasons. There's no way around it.

The incentives all point to winning as many games as possible. As with anything competition based, it's on the players to improve and earn their spots in these lineups, and the conversation should be about creating the best atmosphere for that improvement in arenas like elim and twd. Either that or turn TWDT into 4v4, 6v6 and force more teams where captains are actually required to utilize more of the zone's players. But you can already imagine the push back with that. A big problem has been our reluctance to move past 5v5 and 8v8 games that never happen outside of leagues as our population has decreased. But since that's the case, a single league can't benefit everyone and we have to stop pretending it should.

[Zeebu]

3) There's no point in making it truly 1-10 when there are only 3 tiers of star slots. You only need to differentiate between four classes of players. Even then, the 8-Star slot (that I created) is debatable. There should definitely be three tiers (6-Star, 7-Star, and Max), and maybe four tiers (6-Star, 7-Star, 8-Star, Max). Any more than that is too much. You could make a legitimate argument for eliminating the 8-Star slot and having three Max slots. I prefer having the 8-Star, but it's not an absolute necessity. Remember that before I took over TWDT in 2014, there was only one low-star slot, and in the playoffs you could play with 8 Max slots. Obviously increasing the low-star slots has been a boon with the decline in population, but there's an upper limit as to how much you limit participation from veterans.

what about getting rid of the 'slots' altogether?

[ogron]

1) Camel by Committee has proven to be the absolute worst ratings in multiple TWDT seasons. I've tried it before and it was unanimously considered awful.

a) Most people are bad judges of talent. There's very few truly skilled, active players who are capable of producing good ratings
b) Committees are bad partly because of a), but also because you run into huge consistency issues. People have different ideas of what constitutes a 6 or 7 star, so even if everyone was a good rater, it still produces bad ratings because they're all averaged together
c) Captains with votes intentionally lowball themselves and other players they plan to pick and that may be of value only to them

2) Variable ratings are a terrible idea, way too much work, punish captains for good picks, and would be rife with abuse (like we see in elim)

3) There's no point in making it truly 1-10 when there are only 3 tiers of star slots. You only need to differentiate between four classes of players. Even then, the 8-Star slot (that I created) is debatable. There should definitely be three tiers (6-Star, 7-Star, and Max), and maybe four tiers (6-Star, 7-Star, 8-Star, Max). Any more than that is too much. You could make a legitimate argument for eliminating the 8-Star slot and having three Max slots. I prefer having the 8-Star, but it's not an absolute necessity. Remember that before I took over TWDT in 2014, there was only one low-star slot, and in the playoffs you could play with 8 Max slots. Obviously increasing the low-star slots has been a boon with the decline in population, but there's an upper limit as to how much you limit participation from veterans.

Finals Thoughts

There is no magic bullet for ratings. The actual solution is having 1 veteran TWL player do the master ratings, while getting input from experts in all 3 leagues. That way the 1 person can uniformly apply the 6, 7, and 8 star ratings so they're consistent with each other, while also getting invaluable input from experts who make sure no one slipped through the cracks.

That methodology has unequivocally produced the best ratings and least complaining (there will always be some complaining and 1-2 misrated persons).

The problem wasn't the system this season. The issue was the people doing the ratings. It's supposed to be a huge undertaking that takes weeks and requires lots of hours, and given the sheer laziness on display by TWDT ops, that clearly didn't happen.

The list should be going through 10 iterations, you're supposed to be on a Google Docs sheet with an expert in each league poring over all 150-200 ratings with a fine tooth comb, looking over activity and TWD stats, checking previous TWDT and TWL performances, talking to the best players, etc.

IMHO, Major Crisis has proven to be the most competent contributor this season, and should be in charge next TWDT. He's run a successful TWDT in 2011, and did a great job all season with various tasks. He's a good enough judge of talent and administrator.

I doubt he'd do it with all the cancerous, power-hungry worms that have embedded themselves into staff, but it would benefit the league tremendously.

[Major Crisis]

i dont think its unfair or anything, i just think its a bit of a busted system. draft for skill and activity and then get ahold of the folks with broken ratings. its how this league has always been. i just think theres a better way to do it.

Like I mentioned before, this was done because Henry lacks in coordinating leagues. A proper coordinator would have gotten more people together and maybe even let them rate a list instead of going through the list together on a chat. It went this way, and yea, it had its busting ratings. I am sure zidane knows as well that the ratings and all werent done perfect, but then again it never will be.

I think we can even defend all the ratings who have been 'too low' for many people. For example Ro, ro wasnt that active before TWDT started and we all had a doubt if he was gonna be active. Last season he played (fall 2016) he disappeared midseason. How can we judge that he should maybe be a 9 after all? Babe Ruth shouldnt be a 7 in wb or jav. Yes, he has done decent on Power, but lets be real... He had to play with Ease, Vys, Tripin/Lockdown and an experienced 7* as Lews Therin. I doubt many 6s would have done much worse. The only thing that could be done is all experienced TW players like that move to the 7 spot, so we get the EG/DSB ppl at the 6* spot. But then we have a bunch on 7 who would never get to play anyway. There will always be questionable ratings or rating system.

[Zeebu]

Uhhhhh pretty sure ratings were posted days before the draft lol. This isn't some big conspiracy. A few people in the dt chat looked at old ratings from last season then made changes if the majority agreed. We all were looking at the same ratings before the draft it wasn't some secret affair. There's also some crazy theory where sometimes a player plays better or worse than expected and their ratings look questionable in hindsight... but who knows

i dont think its unfair or anything, i just think its a bit of a busted system. draft for skill and activity and then get ahold of the folks with broken ratings. its how this league has always been. i just think theres a better way to do it.