Thousand tournaments

30 hands are played against different opponents. The approximate duration of the tournament is 1 hour. The same cards are used on different tables. At the end of each hand, the player's score is compared with the score of other players who played with the same cards, and the user receives 0 - 100 points accordingly. At the end of the tournament, the users' points are summed up, and the final ranking of the tournament is created. The number of wins in the hands can be considered as another criterion. The minimum number of participants in the tournament will be 9 (to create an adequate comparison point with other tables).

Now that the sasku tournaments are working without any problems, we could think about a thousand tournaments. The original plan was to evaluate the bidding and playing of one hand. It seems to me that the amount of randomness there is too high to evaluate one hand (those 3 cards in the middle). Maybe it would become too much of a game of chance. One option would be to draw players into tables, deal the same cards in different hands, let the game play out to the end (up to a thousand), and then look at/evaluate the results of different positions. Any thoughts?

I agree that 3 cards in the middle increase randomness. But that's part of the game. In the long run, luck fades away and skill becomes decisive. I certainly don't advocate playing to a thousand. After the 10th-15th hand, the situations are so different that the current score affects the bidding/game strategy. This is especially noticeable when the 800 mark is exceeded. So the results obtained at different tables are not so easily comparable. Moreover, tournaments would then start to drag on, reaching a thousand takes very different times at different tables. Such a format would be conceivable if the starting time could also be different at different tables. So when 3 players get together, they "start the tournament" and the exact results will be revealed later. I advocate a scoring method where the evaluation of all hands is independent of each other. Exactly like in Saskus now. How to evaluate the results obtained in a hand? Let the distribution scores be: 120, 30, 0 100, 20, 0 One way would be to find the arithmetic mean score of the table and subtract it from each player's score. Then the sum of the players' scores at each table is 0: 70, -20, -50 60, -20, -40 These scores could simply be compared with each other. In each comparison, a win gives 2 points and a draw 1. So the points would be given: 2, 1, 0 0, 1, 2 So in this example, 30 on table 1 is "as good" as 20 on table 2. But 0 on table 1 is worse than 0 on table 2. There are certainly other ways to distribute points.

I would only compare the results of the same position. So for three tables: 1) 0: 120 1: 100: 2: 0 2) 0: 90 1: 100 2: 100 3) 0: 0 1: 100 2: -100 the results would be: 1) 0: 100% 1: 50% 2: 50% 2) 0: 50% 1: 50% 2: 100% 3) 0: 0% 1:50% 2: 0%

Yes, it can be done that way. However, it seems to me that the 1:100 on the 3rd table could be worth more than the 1:100 on the other tables. Hence my suggestion to reduce the average.

If the stars are in your favor, tomorrow (Saturday, 27.02, Vint.ee's 9th birthday!) evening at 9:00 PM will be the world's first online thousand tournament!

As I understand it, when calculating tournament points, what counts is how many trick points the user gets from the hand, not how many trick points his opponents get at the table. 1) 0: 120 1: 100: 2: 0 2) 0: 90 1: 100 2: 100 3) 0: 0 1: 100 2: -100 In this example, all players in 1st position get 50% for their 100. How does this change the strategy of tournament thousand compared to regular thousand? 1. If I am on defense, then in regular thousand the first priority is to take down the contract. Not in tournament thousand! Here the first priority is to collect as many points as possible. The contract is irrelevant, because it does not affect my tournament result. 2. If I am the declarer, then my only priority is to play the contract to the full. Everything else is irrelevant. 2a. After getting the board, I have to give 2 cards to each opponent. In regular thousand, it is wise not to give queens-kings to avoid the risk of a new trump. If a good board comes, and I can fill my contract with the first tricks, then in tournament thousand, I can freely deal queens-kings. Especially to friends. If they can make a trump, then I can give them a good result, without affecting my own result. If the players are cooperative, then it is possible that all 3 players get 100% from one table, 300% in total. This is possible, because different tables make different numbers of trumps. And besides, the extra points won by declarer do not count in any way. So the total amount of trick points is different at all tables. How to explain the situation that all 3 players "played better simultaneously" with their cards compared to the same players at other tables? === If we subtract the arithmetic mean of the table from each player's result, then the differences between regular and tournament thousand are smaller. At one table, 3 players can get a total of 100%...200% instead of 0%...300%. Ideally, each table should be distributed exactly 150% (at least the average result of opponents in tournaments of other games is 50%). So arithmetic subtraction is not an ideal solution either.

Yes, there is a point here. I'll do this subtraction of the average. Then the defending player has an incentive to take from the trump taker (so that the average is lower) and the trump taker has an incentive not to give tricks to his opponents (so that the average is lower).

The question arises, what to show in the standings?

The current table format is not very clear. The current 2-dimensional table could be divided into two 1-dimensional tables. One table could display the results of player Me in all hands. The second table could display the results of one hand in all tables. The hand could be selected by clicking on the result in the first table.