Results 1 to 10 of 10

Thread: TSL Finals

  1. #1
    House Baratheon Claushouse's Avatar
    Join Date
    Dec 2002
    Location
    Canada
    Posts
    3,070

    TSL Finals

    The TSL regular season is slated to end on Thursday, August 3rd, 2017. TSL All-Stars (Top 10 in Warbird, Top 10 in Javelin, Top 18 in Base) will be announced on Friday, August 4th, 2017. TSL All-Stars will receive a trophy on TWD with 1 to 3 gems inside, depending on how many leagues they were an All-Star in.

    This means everyone still has a full three more weeks of games to qualify, or about 40-50 more games.



    The TSL Finals will take place on Sunday, August 6th, 2017.

    TSLD Final: 2:00pm EDT
    TSLJ Final: 3:30pm EDT
    TSLB
    Final: 5:00pm EDT

    The Finals will be a best-of-three in each league. Winners will be awarded a blue, yellow, or red ribboned medal in TSLD, TSLJ, and TSLB respectively.

    Even if you narrowly miss becoming a TSL All-Star, you can still win a Finals medal if players are missing or lagging.

    Good luck!







    First Draft of Selection Process -- May Change

    I'll discuss with the finalists on who they feel are two sensible choices for captains in each league, and they will select from the available TSL All-Stars. The lines will remain intact across all three rounds. Any TSL All-Star unable to show will be replaced by the next highest player in the TSLD and TSLJ Standings. In TSLB, the two Multi-Ship All-Stars are the subs/alternates. If more basers are required, the next highest in overall value in standings will be selected, provided they can play the needed ship at a high level.

    The highest seeded of the two selected captains will choose first, with each captain taking one player at a time. In base, captains are not allowed to poach (i.e., taking both terrs to deprive the opposing team of one) and are required to draft for need. It's a Final, not a Game Theory class.

    If, for some reason, captains cannot be agreed upon or refuse, the two highest seeded players will be mandated to captain both sides under risk of being replaced in the Final. In base, the highest seeded player and the next highest seeded player of the same ship will be captains should we be unable to agree on two captains.

    Last edited by Claushouse; 07-14-2017 at 06:46 PM.

  2. #2
    Registered User
    Join Date
    Jan 2010
    Posts
    684
    how does pick order work for finals

  3. #3
    House Baratheon Claushouse's Avatar
    Join Date
    Dec 2002
    Location
    Canada
    Posts
    3,070
    Quote Originally Posted by zidane View Post
    how does pick order work for finals
    Highest seed of the two captains goes first.

    I think we'll do 1/1/1/1/1/1/1/1, but I'm not opposed to 1/2/1/1/1/1/1/1/1/1 like League of Legends if people prefer that.

  4. #4
    Registered User Pressure's Avatar
    Join Date
    Feb 2005
    Posts
    1,387
    My only comment: in a league wherein the bot has been randomizing games for the whole season, why would finals be any different? Shouldn't there be some factor of randomization, since first picks could swing the tide for the other 7 players? Not sure why it would be contingent on captain regular season performance, but maybe you have already thought about this?
    4:Best> pressure was a terr for halloween 4:Best> guy had 4 dudes on him attached

    4:Siaxis> what time is it in zimbobwe para?

    4:Children> every morning i wake up, look in the mirror and thank god i'm not paradise

    4:Best> Mighty's forehead has wifi

    4:Mighty> u guys ever eat a ketchup sandwich?

    4:sphinx> mighty fought a raccoon in an alley for some bbq chips

    4:RaCka> so what if paradise is dressing up as a slim jim for halloween?

  5. #5
    Registered User
    Join Date
    Jan 2010
    Posts
    684
    if it's the latter the highest seed should be able to choose whether 1st or 2nd then

  6. #6
    House Baratheon Claushouse's Avatar
    Join Date
    Dec 2002
    Location
    Canada
    Posts
    3,070
    Quote Originally Posted by Pressure View Post
    My only comment: in a league wherein the bot has been randomizing games for the whole season, why would finals be any different? Shouldn't there be some factor of randomization, since first picks could swing the tide for the other 7 players? Not sure why it would be contingent on captain regular season performance, but maybe you have already thought about this?

    1) Because the bot is far from perfect

    2) If a bot picks bad teams in the regular season, it's 1 out of 100 games. It's not a huge deal. Your luck will even out over time. The bot is fine over a large sample set, as you'll play on roughly the same amount of good teams, bad teams, and even teams if you play enough games. If it picks bad teams in the Finals, it ruins the Finals as it is a 1-game sample set.

    3) The bot struggles with distributing the right amount of basing ship-types to both teams

    4) I think we'll be able to achieve a better finals using Captains. Captains are slow, hold up the flow of TSL, and most are unwilling in the regular season -- having the bot select and playing fast is much better. But in the Finals, this isn't an issue.

    5) I agree there could be issues in some leagues depending on who gets the first pick, but I'm not sure there's an ideal solution. I don't see it mattering much in warbird, but whoever gets Zidane in jav (or if Zid caps and gets 1st pick) could unbalance things. If you have feedback on a solution or better way, I'm open to it.

    I'll put the Selection Process in italics and at the bottom for now, and see if we can find something better. If not, we'll use captains.

  7. #7
    House Baratheon Claushouse's Avatar
    Join Date
    Dec 2002
    Location
    Canada
    Posts
    3,070

    TSL Finals Pairings

    So, I've thought about this a bit, and this is what I came up with.

    Instead of captains or completely randomized bot teams, how about this:

    I will manually pair players of even strength together. If I find good pairings for everyone all the pairings will be 1 player for 1 player. If not, I'll group larger segments of players together.

    TSLB, for example:

    Thix & Ogron
    Trasher & Spezza
    skyforger & nosydawg
    mcvicar & mythril


    I'll run the pairings by Finalists, then if most are okay with them, I'll randomize them live on stream. So I'll put Thix and Ogron up in terr, and randomize them. Whichever name pops up first goes to Freq 0, the second goes to Freq 1. Then I'll randomize Trasher and Spezza, and they'll either go to Freq 0 or Freq 1, etc.

    If there is an imbalance, I will group more than two players together to even the sorting out. For example, Zidane is going to be an imbalance in shark, as he doesn't have a top 5 all-time shark to pair with (Mattey, a2m, Cranium, etc.)


    https://gyazo.com/3f0af53258e10db3dd9f8b70a211fe62

    What I can do is pair multiple players together, so if there are 3 elite spiders, the team without Zidane gets two, and Zid's team gets one.

    In jav, we can separate Zidane and Ease, in warbird, Iron and Mythril.

    So the teams would be random, but I'd pair players together so that the teams are fair.

  8. #8
    Registered User
    Join Date
    Jun 2017
    Posts
    45
    Like this alot!

    I'd also consider doing the finals in 3 rounds and having the pairings/segments randomized for each round. TSL Champions crowned by $value or wins count.

  9. #9
    House Baratheon Claushouse's Avatar
    Join Date
    Dec 2002
    Location
    Canada
    Posts
    3,070
    Quote Originally Posted by skyforger View Post
    Like this alot!

    I'd also consider doing the finals in 3 rounds and having the pairings/segments randomized for each round. TSL Champions crowned by $value or wins count.
    I thought about this initially, but it produces too many major issues.

    1) Some players would "win" after 2 rounds, removing incentive to try round 3
    2) You could conceivably medal in 2 rounds, then play soft so your friends on the other team get their medal, or because you don't like the players on your team in round 3.
    3) It's anticlimactic and confusing to both spectators and players. Round 3 ends... who won? who lost?

    I think it's better to simply keep things simple -- you're a TSL All-Star, you get to play in the Finals as a reward, and get a chance to win a TSLD, TSLJ, or TSLB medal to cap your season off.

    Aligning players directly with their team's fate in a regular final keeps incentives direct -- win.

    A final is always a small sample size, and while randomized teams over a large sample like the regular season (with 100 games) means that, over time, the best players will always emerge -- the luck of randomized teams evens out over a long period and you'll be on "good" teams roughly as much as "bad" teams. Taking a more hands-on approach for one 3-game series seems like the best strategy to ensure fair teams.

    The real prize of TSL is qualifying as an All-Star over a full season with random teammates.

    The Finals is more of a reward for that achievement, and a good way to cap off the season, showcasing the top players from that season in a Final (although given there will only be 5 medals in WB/Jav given out for winning the Final, it's certainly a worth fighting hard in the finals for!)

  10. #10
    Registered User
    Join Date
    Jun 2016
    Posts
    43
    I like your pairing idea. What about a combo of pairing and picking? So captain 1 picks ogron then thix auto on other team. Captain 2 picks trasher so spezza auto other team, etc

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •