Molossus 2: a 3h 6x 1m Tourney

Molossus 2: A 3h 6x 1m Tourney

Molossus 2 is a wonderful opportunity to play some serious, long time control, games against players of similar skill as well as spectate the same. Each game is split into two sessions, each scheduled at mutually agreeable times by the two players in order to make scheduling easier.

You can check out Molossus 1 under the “Previous Rounds” dropdown below, get a more comprehensive overview of the tourney under “More Info”, or read the entire post for the full picture. In any event, please feel free to comment down below, even if you aren’t able to participate.

More Info

All pairing will be posted when the Tourney starts, with each player assigned an opponent for each week running from Sunday to Sunday. Each week, the players will contact their respective opponents to schedule that week’s game at a time that works for both of them. If they can’t find a time, they can agree to play one or both sessions during a subsequent week of Molossus 2: as long as it’s not holding up the completion of the tourney, we would much rather have a good game played a bit later than no game at all.

Molossus 2 will take either 3 or 5 weeks to complete, depending on the number of players. The players will be divided into one or more groups of 3-6 players of similar skill. The players in each group will get to play once against every other person in their group. Ties in individual games or in the final standings will simply be considered as ties: no tie-breaks.

The pairings and results are here. It’s a wiki, so feel free to edit in your session times and results.

The name “Molossus” comes from a poetic foot of the same name which consists of three long syllables (stressed syllables in modern English meter).

Please comment in post or pm if you believe I have overlooked something in these rules.

Previous Rounds

Molossus 1

Sealed Tourney (aka “Molossus 0”)

Game Rules

All games should be unranked and played using NZD Rules and 7.0 komi on OGS. Time control should be set to byo-yomi with a main time of 3 hours for each player and 6 periods of 1 minute each.

Allowed Aids


  • Game Chat/Malko Log
  • Paper Kifu
  • Pencil/Paper, or equivalent (these may hold the current or earlier game state(s), but should not contain any future game states during the sessions)

Allowed between (but not during) sessions:

  • Playing out variations on electronic and/or physical boards


  • Consulting with other humans regarding the game
  • Using AI (including but not limited to LeelaZero, Katago, and score estimators)
  • Using position and/or game databases (including but not limited to Waltheri and joseki or fuseki dictionaries)


Each week, each participant should contact their designated opponent via forum and/or OGS pm. The two players should make mutual reasonable effort to arrange a time UTC at which they can both begin each of their two sessions, optimally by the end of the following weekend.

Upon reaching agreement, one of the players should update the pairings post accordingly, both so that the players can check it for reference, and so that passersby and other players can plan to spectate the game as it’s being played.

Inability to Schedule Sessions

While players are highly encouraged to make all reasonable effort to schedule and complete their game within the week, leeway will be extended if the following conditions are met:
a) they can show reasonable effort on their part to arrange the game in a timely manner
b) they are not postponing the completion of Molossus 2 for more than a week by their delay.

If one player is unresponsive to attempts by the other to arrange the game, the second player may request an adjudicated win by forfeit.

If both players are unable to arrange a mutually agreeable game time despite mutual effort, they can request an adjudicated draw.

Sealed Move

Each game will be played over two sessions. The first session will conclude, if not ending due to the game being completed, with a sealed move, which will only be revealed to the opponent at the beginning of the next session.

The first session will be played for three hours, at which point the players can continue playing on for as long as they wish, but they now both have the option every turn to seal a move instead of playing, which will bring the session to a close when done. Thus noone is forced to continue playing the first session past 3 hours longer than it takes them to come up with a move to seal after at least 3 hours have past.

For more details on how this works in practice, see “How to Seal a Move” below. For more information on how to start each session, see “Boilerplate”.

How to Seal a Move

When one player (during session 1, 3h+ in) decides to seal a move, instead of playing it on the board, they should enact the following procedure, in order:

  1. Type the sealed move into the Malkovich Log (Accessed by left-clicking on the word “chat” to the left of the chat input box. It will be visible to spectators but not the opponent until after the game. (alternatively, you may use a hash to conceal your move and paste the hash in the game chat)).
  2. Pause the game. (OGS will still think it is the sealing player’s turn)
  3. Inform the opponent in game chat that you have sealed your move.

It is the responsibility of the player sealing the move to ensure it is unambiguous and legal. It is recommended to ctrl-click on your intended move to put it in the malko log, and check by hovering over the sealed move in the Malko log that it is correct, before pausing the game. If you sealed the wrong move or changed your mind before having paused the game or informed your opponent of your intent to seal, just say so in the malko log. It is only once you have paused the game that you must live with the move you have sealed, so double check before pausing the game.

Unpausing by either player before they reconvene for session two is grounds for adjudicated loss by forfeit. Play of moves (with the exception of the sealed moves at the beginning of the second session) while the game is paused is disallowed.

When both players arrive for the resumption of the game, the player who sealed a move should play on the board their sealed move, and then unpause the game. No moves other than the sealed move may be played while the game is paused.

If one player fails to show up, the players may work it out for themselves or ask a moderator or the TD for a win by forfeit to the present player at the present player’s discretion.

Thereafter the game continues normally to conclusion.

To my knowledge, we have not yet had a session 2 that went to more than 3 hours, but it is possible if both players are taking their time. If your session 2 lasts longer than 3 hours, I would be interested to know, along with your opinion on if that was a problem that could be remedied by a slightly longer (15-30 minutes) session 1, or if it was an unusual case and the current session times are good.


When the first session begins, one of the players must make a note in chat containing the following information (exact wording not required as long as the information is clearly communicated):

This game for "Molossus 2" is being played over two sessions using a sealed move between the sessions. This is session one, starting at <time>.
After three hours, one player will seal a move and pause the game. The game will resume and be played to completion at <resumption time/date>.

One of the players should update the pairings post with a link to their game so that other players can view it.

When the second session begins, one of the players must make a note in chat containing the following information (exact wording not required as long as the information is clearly communicated):

This is session two, which will last until the game is complete, starting at <time>.

This procedure is intended to ensure that the players are on the same page, as well as explaining what is going on to future and current spectators.


Autarchic Obfuscation

Rights of the TD:

  • To update the rules, such as to codify precedent or to prophylactically handle ambiguities or vaugities in the rules.
  • To make final decisions regarding cases in the following situations:
    • The case is appealed at the discretion of the moderator handling the case.
    • Two or more moderators both having valid claims to original jurisdiction in the case have arrived at different rulings and are unable to reconcile the discrepancy among themselves.
    • A moderator’s ruling on a case has failed to adhere to reasonable application of the rules, in the TD’s estimation.
    • The TD chooses to exercise original jurisdiction on any case, esp. as pertains to an ambiguity or vagueness in these rules, at their discretion.
  • To assign original jurisdiction to a moderator of their choice in the case that no moderator has yet had it.
  • To decline appellate jurisdiction of a particular case in favor of the decision of the moderator having original jurisdiction.
  • To adjudicate individual games as wins, losses, or draws.
  • To disqualify players from the tourney.
  • To appoint moderators at discretion, contingent on the candidate’s agreement.
  • To play in the tourney.

Rights of Moderators:

  • To make rulings regarding the application of these rules to specific tournament situations.
  • To adjudicate individual games as wins, losses, or draws in accordance with these rules.
  • To appeal cases to the TD at their discretion.
  • To play in the tourney.

Rights of Players:

  • To contact a moderator or the TD (or to ping them all, or all online ones, or any other subset) to ask for a ruling regarding a dispute between them and one or more other players.
  • To ask the moderator with original jurisdiction over a case which they are involved in to appeal the case to the TD.
  • To ask the TD to review a case.
  • To play in the tourney.

In any case involving one or more moderators and/or the TD, a disinterested moderator or TD should take original jurisdiction. If no such person exists, the TD should find a 3rd party to moderate the case, making all reasonable efforts to make such party amenable to all interested parties.


In order to join, pm me (@Samraku) either here or on OGS, or post in this thread, and I will add you to the list, in alphabetical CWM FJORD BANK GLYPHS VEXT QUIZ order.

The tourney will begin on Sun., 4 Oct., 2020 with the following participants (so far):

  • chartreuse
  • Watcharin47
  • GOkyle
  • Lyle
  • Samraku

Pairings, Schedule, and Results

Week 1

  • Watcharin47 (B) vs. @Samraku (W) result
    • Session1
    • Session2
  • @Lyle (B) vs. @chartreuse (W) B+F
    • Session1: Sunday, 10/11, 1pm EST (5pm UTC)
    • Session2
  • @GOkyle bye

Week 2

  • Watcharin47 (B) vs. @GOkyle (W) result
    • Session1
    • Session2
  • @Lyle (B) vs. @Samraku (W) result
    • Sunday, 18th Oct., 2020, 1pm US ET (5pm UTC)
    • TBD
  • @chartreuse bye

Week 3

Week 4

Week 5

Group A Running Scores

A win is worth 1 point, a draw 0.5, and a loss or bye 0.

chartreuse: 0 (withdrawn)
Watcharin47: 1
GOkyle: 1
Lyle: 1
Samraku: 1


Week 1


Week 2


Lyle —(username to sign up for tourney)—

1 Like

I’ve added you. :slight_smile:

Welcome aboard, GOkyle! :smiley:

May I know the ranks of the participants? If they’re similar to 7kyu or a few stones stronger I might be interested.

1 Like

Lyle is 19k, Watcharin47 is 7k, I am 7k, GOkyle is 2k, and chartreuse is 1k. Also, if we get 7 or more total participants, the players will be split into two or more groups by strength.

While I would ideally wish to start this at noon on Monday UTC, in order to ensure that it does not start before the end of Oct. 4th in any time zone, I will be at work at that time. Therefore the pairings will be posted later in the day.

The pairings will be for all weeks, so official beginning of each week is not particularly important from here on in.

Molossus 2 has can now begin in earnest. I hope to enjoy some great games both as a player and as a spectator, and I wish you all the same as well.

Since the pairings would not change significantly between 5 and 6 players, I will add up to one other person who applies by Oct. 11th (so during week 1) on a first come first served basis to the Tourney. Adding more than that would drastically alter how I would have done the pairings, since 7 was my cutoff for splitting into two groups, and thus I cannot accept more than one, unfortunately. The only change to the pairings if we do get a sixth player (@ほっと茶? @S_Alexander?), will be to slot them in against the player with a bye each week, randomizing B/W as with the other pairings.

Checking in! I’ll be scheduling my game with @Lyle asap


I sent you the challenge. :black_circle: :white_circle: @chartreuse

Week 2 has started; good luck to the players!

@chartreuse has withdrawn; I’m looking forward to playing him again in the future, perhaps next January in Molossus 3? We’ll see what the future has in store.

I am sorry for my ill attention regarding this Tourney. I am interested in finding a way to revive it, possibly using a format which allows participants to decide ahead of time the number of games they will sign up for instead of having to play them all or not sign up.

That said, Molossus 3 is on indefinite hiatus, but I’m definitely up for some long time control games if anyone’s interested in that in the meantime.

As much as I’d love to announce a new round starting in a few weeks (to allow time for signups), I think two things need to be true before I can do that.

Firstly, I’m currently playing more Arimaa than Go, so at the risk of sounding all new-agey, I need to really be “feeling” Go: I need to be on a Go kick. I still love Go, I still want to study and improve more in the future, but I’ve been largely taking a break from it these past few months in deference to work, school, and Arimaa. It’s happened before though, and I know it will happen again: my interest in Go will rekindle; perhaps not even too far out.

Secondly, I need to revisit the format. Not the time control as I’m happy with that; not the rules (at least not majorly) as I think that incremental adjustment is more appropriate there, but the format itself. Right now it’s multiple groups of individual round robin mini-tournaments. I don’t think this is by any means a bad format, but there is one shortcoming which I think may render a change beneficial on the balance: each potential participant must commit, in the current format, to play up to 5 games over the course of as many weeks, or decline to participate. If I ever considered allowing players to choose how many games they played, I must have dismissed it, but I’ve become increasingly convinced over the last 2-3 months that that is the way to go.

clyring recently restarted a running thematic Arimaa event with endgame starting positions: Endless Endgame Event. While allowing players to choose when signing up for the next 4 week period how many games they wish to play in that period does mean that there is no overall winner each week, it is counterbalanced by the fact that someone only able to commit to one game in that period can participate right alongside someone able to play twice a week. While the opportunity to play serious long time-control games is enough of a draw that I would participate regardless, clyring also provides continuity between rounds by keeping WHR records for the EEE games.

This, possibly without the WHR, is essentially what I’m considering, possibly with some automatic rest weeks in between given the long nature of Molossus Tourneys. If I want WHR, I would need to figure out how to calculate it, which should be tractable; there are some open-source programs to do it, I believe.

So for now the current plan is to use an EEE-inspired format for a new reverse komi running tourney: “Lagrange Points”. Time controls will have a default that either player can insist for any reason or no reason at all be used, but players will be free to agree to play by any other fischer or byo-yomi time controls they and their opponent agree upon before the match. I hope that this will allow players who like more extreme time controls to play them when they run into players who feel likewise or are just flexible, while not requiring anyone to commit to anything too out of the ordinary. Currently I’m pretty sure 30m + 15s/move fischer will be the default: games which go to counting should last a bit over 2 hours if both players are using all their time.

I think this will appeal to a larger range of players than Molossus 0, 1, and 2 were able to, and hopefully we can find some good games.

1 Like

Groin is organizing a place to find long time control games which may appeal to people who enjoyed Molossus.

1 Like