The DailyGammon Tric Trac Tourneys

What is a TTT?
  • A TTT, or Tric Trac Tourney, is a round-robin of cubeless one-point games with a scoring system designed to encourage aggressive, gammon-seeking checker play. Gammons are not just counted but are crucial to success in a TTT. This is detailed in the Scoring section below.
  • The DailyGammon TTTs come in various flavours. The forum-based ones come as "regular" TTTs, "stratified" TTTs and the occasional "Mega TTT". The Christmas TTTs are a seasonal version of the regular ones.
  • A regular TTT has 12 players, and hence everyone gets 11 games to play. The time control is once-a-day for these and all matches are expected to be started together and played in parallel.
  • A stratified TTT is one of a set of four where each has 12 players out of a sign-up set of 48. Like the DailyGammon stratified tourneys, the set of players is partitioned according to skill level. With DailyGammon tournaments the rating is used. For stratified TTTs this will also be the case however it is hoped that a better metric may be developed in due course, in particular using TTTs results of the players.
  • A Mega TTT has as many players as want to play. Typically has been 70-100 players which means a lot of matches! The time control is a more relaxed once-per-two-days and a generous period of time is alllowed for getting matches started.
  • The TTT concept came from the able mind of one Mike "Mad Monk" Main, who was dedicated and highly respected British Tournament Director.
How it works
  • A TTT starts with a sign up thread in the forum. For regular TTTs, as many players as want to play post their name. Usually there are more than the 12 required for a single TTT and so a set of TTTs will be played. If there aren't enough players to completely fill the TTTs then players will be invited to sign up for a double or triple dose of TTT fun.
  • Although these are individual TTTs, the players will be assigned to groups so that the groups are balanced and reasonably comparable. This allows the presentation of tables of statistics which contain all the players in the TTT set. Players who have entered more than once will have multiple entries in these tables.
  • Once sign-ups are closed, the web site will be prepared and the tourney started. Each player gets to make half of the invitations for their matches, the other half being the responsibility of the opponents.
  • The collection of match results is done automatically on an irregular basis by the TD and the results fed to the website. Traditionally, players have been required to report their match results in the forum thread. This is no longer necessary although it still occurs.
Getting started
  • Each player will be provided with a personal page which is to be used for inviting their opponents to a match.
  • In a Mega TTT the invitations may be spread out over a period of up to 6 weeks. The invitation pages will only have a third of the invitations enabled at the start. After 2 weeks another third will be enabled. The rest will be enabled 4 weeks after the start. This is to help players who would like to pace their matches to be able to do so without having their message queue blocked by invitations that are waiting to be accepted.
  • If two or more matches are started between any pair of players, the one with the lowest match id will be deemed the official one and all others will be deemed a friendly match.
  • These friendly matches, along with any arising from other incorrectly accepted invitations (made between groups or by a non-tourney player clicking an invitation button) will have the name of the tournament but will not belong to it. Such matches may be played out or resigned; it's entirely up to the players.
  • If you have duplicates and resign the official match by mistake then you will forfeit that match unless it is clear that it was a mistake and that you gained no advantage. Please be very careful about resigning when you have duplicates. You can check which is the official match by going to the match table page and hovering your mouse over the cell for yourself and your opponent. The link in the status bar will show the match id. If you don't have a status bar then clicking the cell will go to the match page and you can have a look at it.
  • Once underway, each player will also have a personal match table. The full match table for a Mega TTT is very unwieldy and scanning it to check matches is a challenge. It'll probably be much easier to check the match tables of individual players.
Scoring
  • One of the big differences between TTTs and ordinary tournaments is that points are awarded even when a player loses. This seeming quirk is what gives the TTT format its power.

    Imagine playing two ordinary single-point games. If you play to win and you win both then you gain 2 points. Now consider that you play extra aggressively for gammon and in one game it works but the other is lost. You get 2 points for the gammon and 0 for the loss. Overall it's the same 2 points as before.

    Now consider the same situation with TTT scoring. You play aggressively and win one gammon but lose the other game. You get 2 points for the gammon and ½ for the loss, for a net result of 2½ points. Clearly, even though you play more riskily in going for the gammon and lose a game, you come out with more points than by playing conservatively!

    This is the essence of TTT. The risk versus reward in going for gammon is such that and you can afford to be - and should be - extra aggressive in seeking gammons.
  •  
    Outcome Points
    scored
    Original
    scoring
    Win single 1.0 1
    Win gammon 2.0 2
    Win backgammon 2.7 3
    Lose single 0.5 ½
    Lose gammon 0.5 ½
    Lose backgammon0.3 ½
     
    Win by timeout at least 1 3
    Lose by timeout0 unknown
     
  • This scoring system is slightly different to the original TTT scoring. Backgammon wins are rewarded less and backgammon losses are compensated less.
  • The original scoring system's 3 points for a backgammon win has been reduced to 2.7. The TTT format is about gammons. High gammon rates are due to the aggression, skill, and luck of the players. With backgammons the luck component is much higher. Some backgammons are the result of skill but most are fortuitous and some may even be given away by the opponent.

    This scoring system's adjustment to the points for backgammon wins reflects this additional luck component and is intended to allow gammons to win out in tied situations.

    Consider two players, one with 1 win and 1 backgammon, the other with 2 gammons. Using the original scoring, each will have 4 points and be tied. With this scoring system, the backgammon win plus a single win (3.7 points) doesn't measure up to the two gammon wins (4 points). The player who more successfuly followed the TTT ethos of going for the gammon scores better here.
  • A backgammon loss is compensated less than a single or a gammon loss in order to discourage the recklessly giving away of backgammons. Giving away a backgammon, whether negligently by not caring about it, or deliberately, such as "being nice" to the opponent, is unfair to all other players in the TTT.

    Perhaps a better way of looking at this is that saving a backgammon wins 0.2 points.

Tie-breakers
  • In the event of two players tying for first place, the head-to-head result will be the decider.
  • If three or more players tie for first place and no winner emerges from the head-to-head results then a mini-TTT will be played to decide between them.
Timeouts and resigning
  • It is important that under no circumstances should you time out. If there's any chance that you may do so then please arrange in advance for someone to ready to step in and play your TTT games for you. You can let all your other matches time out but not the TTT ones!
  • This is because multiple timeouts for a player who already has several results can drastically affect the scoring. It not only penalises those opponents who played faster, especially the ones who lost, it will probably alter placings and may even change who wins the tournament!
  • If a player times out then I will evaluate the last position's percentages with GnuBg 2-ply cubeless analysis and apply the TTT equity calculation to the opponent's percentages. That equity will be increased by half a point, rounded to the nearest tenth of a point and finally subjected to limits (a minimum of 1 and the maximum of the score for a backgammon). This value will be the score for the match.
    This is another reason not to timeout. At the moment it's a time-consuming manual process and so I don't want to have to do any of these! ;o)
  • Examples
    In the following, the Win Rate includes wins by gammon and backgammon and the Gammon Rate includes backgammons, this being how GnuBg presents its evaluations.

    Equity = Wins + Gammons + Backgammons * 0.7 + Losses * 0.5
    Score = Equity + 0.5, rounded and limited to (1 .. 2.7)

    Some backgammon potential
    Wins = 98.7%, Gammons = 96.3%, Backgammons = 12.3%
    Equity = 0.987 + 0.963 + 0.123 * 0.7 + (1 - 0.987) * 0.5 = 2.0426
    Score = 2.5426, rounded and limited = 2.5

    A likely gammon
    Wins = 96.5%, Gammons = 77.7%, Backgammons = 0.2%
    Equity = 0.965 + 0.777 + 0.002 * 0.7 + (1 - 0.965) * 0.5 = 1.7609
    Score = 2.2609, rounded and limited = 2.3

    Decent gammon potential
    Wins = 82.5%, Gammons = 30.3%, Backgammons = 0.3%
    Equity = 0.825 + 0.303 + 0.003 * 0.7 + (1 - 0.825) * 0.5 = 1.2176
    Score = 1.7176, rounded and limited = 1.7

    An even game with some gammon potential
    Wins = 51.5%, Gammons = 18.9%, Backgammons = 0.5%
    Equity = 0.515 + 0.189 + 0.005 * 0.7 + (1 - 0.515) * 0.5 = 0.95
    Score = 1.45, rounded and limited = 1.5

    A likely win but no gammon potential
    Wins = 87.6%, Gammons = 0%, Backgammons = 0%
    Equity = 0.876 + (1 - 0.876) * 0.5 = 0.938
    Score = 1.438, rounded and limited = 1.4

    An unstarted money game
    Wins = 50.0%, Gammons = 13.7%, Backgammons = 0.6%
    Equity = 0.50 + 0.137 + 0.006 * 0.7 + (1 - 0.50) * 0.5 = 0.8912
    Score = 1.3912, rounded and limited = 1.4
  • If a player times out in too many of their games then I reserve the right to remove them from the tournament. As mentioned before, timeouts can very much upset the balance of the whole TTT. So can removing a player. However, although it's not necessarily the best, to me it's conceptually cleaner to consider them as having never played at all. The decision of whether to remove or keep a timeouter will be made if the situation arises.
  • Resigning has the same effect as timing out. You should never resign a TTT game.
The small print
  • The Tournament Director reserves the right to amend these rules in any way that circumstances may require.
  • Signing up for a TTT does not a guarantee a place. The Tourney Director may refuse entry to any player, with or without a stated reason.