|
Frequent questions |
|
|
Homepage Contents | |||
  | |||
|
Software
Besides Windows
Useful information
Prepared games
|
Why study a deal? A question of feeling
When we play cards, we are not always as attentive as we think. There are the pace of the game, the emotions linked to the desire to win, what is at stake on the deal. And in the rapid flow of information that is reaching us every moment, there are losses, that's normal. We do not remember that a card was played, it was a Diamond and we saw a Heart, we no longer remember a bid, in short, our memory and our attention are not perfect. Hence the impression that some machine players make mistakes. Strategy errors of course, but also breaches of the rules of the game, or even curious decisions that make one think of cheating. All these events cause us an unpleasant feeling: that of being disadvantaged, deceived by the opponents, and for team games, of having a partner who is (it is obvious to us) a terrible player!
And yet it is not difficult to escape this negative feeling that spoils our games. A few clicks are enough. We simply have to study the deal and find the moment when the decision of a player has bristled us, even scandalized. That's all, it's easy, and it can change everything in our analysis of the game.
I always tell this story because it really happened to me. I saved a deal one day with the comment: "But why does North ruff, I'm winning the trick!!!" (with a string of exclamation marks). I was very unhappy with the playing engine which was making a huge mistake. Some time later, I open the file where the deal was saved. And I saw the mistake again. Indeed, my partner North had ruffed with a low Heart a trick that I was winning. What a mistake!!! Except that, looking at it more closely... Spades were trumps! My partner had not ruffed, he had discarded! How did I go so wrong, I still wonder. But it is a reality. This kind of thoughtlessness happens to everyone.
You have at your disposal all the necessary tools to verify that your impressions are very often misleading. Once you have played the entire deal, you can see all 4 hands at the very beginning of the deal, and study step by step how it went. To do this, use the progress buttons that are enabled as soon as a deal is over.
Find out about their keyboard shortcuts via the Progress menu.
Then go down step by step by a simple left click, or go up by a right click. By keeping the Control key pressed, you can move from trick to trick. This is how you will first eliminate all the situations where you thought there was a mistake, whereas the player in question had in fact no other choice.
To better understand the strategy of machine players, you will need to know the intention of the player when playing. To do this, click the Explain decision button. A short text will then tell you what the player is trying to do. In bridge, the player's game plan will often be detailed, sometimes with several steps in preparation.
Machine players do not see other player's cards. But before playing, each of them reconstitute the hidden hands, taking into account the bids and cards previously played. You can see what the player on lead knows about the other hands, by the Player's visualization button. Here is an example of visualization for mixed bridge and belote:
In Bel Atout, visualization at 70% is less certain than at 90%. Many signals when discarding are placed at 70% because a player often has little choice of which card to pitch. Some signals will still be transferred to the 90% visualization if confirmed by a second card. A player uses his visualization at 90% or 70% depending on the situation. He uses his 100% visualization for safety plays.
In SimiliBridge, visualization at 70% is reserved for urgent situations. Some signals are interpreted at 90% and 70% differently, given that the player does not always know whether his partner has signaled an urgent situation or not. In the next plays, the player will use his visualization at 70% if he analyzes an urgent situation, and if he can think that his partner had spotted it himself when making his signal.
There is however a risk in studying a deal while seeing all 4 hands. The risk is to analyze playing mistakes that seem obvious, without being aware that the machine player could not see all the hidden cards when making his decision. It is the feeling with all cards face up. We all got caught up in it, me included. When we see the four hands, we can easily guess the winning line of play. The player should have led that suit, he should not have drawn trumps, he should have won the trick or on the contrary played low. And from there, everything was simple and easy. The player made a huge blunder. How can one play so badly!?
It happens quite often to me to save a deal which I classify in the big obvious mistakes. And when I set to work to correct it, I suddenly realize my error. The player does not have the good vision of the hands. From his point of view, the line of play is not so obvious, and even it can seem dangerous to him. If I manage to improve his deductions so as to give him the correct view of the hidden hands, I see the end of it at little cost. But in most cases, I have to look for other arguments, much more complex to analyze, involving a necessity hypothesis (without this hypothesis, no chance of winning), or notions such as urgency or security plays. It sometimes takes me several hours to correct a mistake that at first seemed obvious.
We must therefore be wary of analyses that we do at the end of the deal, when we know all the cards and we reason only about the winning line of play. If we say "You had to play this card because it was the one that won", it's not good. We have to go further in the analysis, and explain how the player could find the right card to play from his personal view of the hands. Otherwise, we are led to establish strategy rules that vary according to the deals. On one deal, we will lecture our partner by saying "You should have drawn trumps!", while on another, we will repeat vehemently to him "You should not have drawn trumps!". Hence, our poor partner will be totally destabilized by these contradictory instructions, because from his hand, he will never be able to know in which play situation he is.
We must also reason about probabilities. If a strategy makes us win in 90% of cases, we must adopt it, while accepting that in the remaining 10% of cases, it will make us lose. Because if we cannot distinguish the cases where it makes win from those where it makes lose, we have to put the odds in our favor. Of course, a player who does not adopt the strategy may win in 10% of cases. And he will be able to show dozens and dozens of deals where he was right not to adopt the strategy. But all the same in the long term, out of the astronomical number of possible deals (see below), this player will lose.
In bridge, there are 53 644 737 765 488 792 839 237 440 000 possible deals
I think you have understood, from reading the lines above, that it is useless to report to me a playing mistake without sending me as an attachment the deal relating to your questions. Without the deal, I can't say or do anything. I don't know if you're wrong or not in your analyses, I don't know your rule preferences, I don't know exactly what happened before the decision you're challenging, and I'm not sure you're describing the situation properly. What is the machine player's vision when playing? What were the cards played before the disputed play? Sometimes, playing an 8 instead of a 7 can have changed a player's entire vision. You will never be able to describe all this to me with good reliability.
Fortunately, there is a button that solves all these problems with a few clicks. It is the Save deal as button, which allows you to save the deal to a file, with all the information I will need to analyze the playing mistake. By default, this file will have the .bel extension for Bel Atout, .bri for SimiliBridge, .pre for Élysée and .whi for Far Whist. Here is a brief description of the procedure you can follow:
Then you just have to send me a message and attach the file you saved. Your deal files will be in your Documents folder, in the sub-folder:
When studying a deal with all cards face up, you have control of the 4 hands. You can therefore change a bid to study a different contract, or change a card played to study another line of play. You can do anything you want on the deal. The results obtained on a different contract or another line of play will be displayed on a colored background, as a reminder that these scores do not count for the game. Here are the other tools you can use to make your study easier:
When playing around a table with friends, in a tournament or on line, we rarely have the leisure to redistribute the 4 hands, and study everything that happened on the deal, with a clear vision of the hand of each player before playing. So take advantage of my programs to think more about the different strategies. Learn to recognize your mistakes, to question some of your playing habits, and to conduct more rigorous analyses of situations. Practice respecting your partner, understanding the difficulties he faces, and finding the right arguments to help him play better. It is by studying the deals that you will become a good player and a good partner. And maybe also… by the comparison !
|
