Who are your opponents? You are sitting E/W at table 8 in a 12 table game. Are your opponents the N/S team # 8? Or are they the 11 other E/W players sitting in the room? A good Duplicate (or Match Point) player realizes that his true opponents are the 11 other teams sitting E/W. Your SCORE must be better then theirs. And at some point during the auction or play, you ask: "What is happening at the other tables." The answer to this question can effect bidding and playing strategy.
         Your N/S opponents are playing opponents; they are just vehicles to SCORE points with. What is crucial is the number of E/W pairs you can SCORE better than. You do not win if you get a high SCORE. But you gain only if your SCORE is better than the SCORES of the other E/W pairs. And it does not matter by how much, or by how little. This can be done by playing or defending 1 # or 7 No Trump. The maximum (or minimum) number of Match Points is the same for each hand played with a particular board.
        Match Point scoring is on a relative basis. It is your SCORE compared to your true opponents' SCORE. Therefore overtricks do count. 
        If possible keep away from the minors and always look for a major suit fit. It may often be worth while to play in a major suit with 7 trump than in a minor with 8 trump.

PART SCORE: Most Duplicate bidding is done in the clinches; battling for a Part Score. How high to go; when to double; and when to defend. This is where knowledge of how to SCORE comes in. 
          The vigorous fight for a Part Score is the most competitive part of Duplicate Bridge bidding. Since you can go down 1 or 2 tricks and still get a good score, you will tend to be more aggressive. Your playing opponent (N/S # 8 as in the above example) bids and makes 3 of a minor (or 2 of a major) for a SCORE of 110. That is, you lost 110 and they are plus 110. But if you bid, let us say, 3 $ and go down 1 trick, you will just lose 50 (100 if vul). You will get a better SCORE going down 1(-50 or -100), than letting N/S make a Part Score (-110). And if you are the only one in the room to do this, then you will get a top. 
         Beware of going down and losing 200 points (the death number). Most Part Score contracts SCORE between 110 and 190 points. So don't give your opponents a 200 number. For if you are minus 200 (vul: down 2 tricks or 1 doubled), then you will most likely get a bottom or near bottom. 

GAME: In Duplicate bidding you are not rewarded for the SCORE you achieve. If you bid and make a Game along with everyone else in the room, then you will  get an average along with everyone else in the room. You must outscore the other pairs sitting in your direction by any margin. 
        If you think everyone in the room with your cards will be in Game, then bid it. If it does not make, you will have plenty of company and get near an average. But to be in a Game that is risky and does not make, you will get a bottom or near a bottom. That is because most of the teams are in a Part Score getting a plus SCORE, while you are getting a negative SCORE. So bid Game if your are fairly certain that it is can make and that other teams will be there. If the Game has a 50 % chance or better of making, then bid it.
        Keep away from the minors. That is, if possible play in No Trump or the majors. 5 # or 5 u will SCORE the same as 3 No Trump and 20 points less than 4 of a major. If the choice is between the major and No Trump, choose playing in  the major: it is safer and will usually outscore No Trump.

SLAMS: As with bidding Games, bidding Slams should be done only if there is the chance that the Slam could be made. Again, the Slam should be bid if there is a 50% or better chance that it could make. In a conservative bidding field,  however, that percentage might be raised to 60%. Going down in a Slam when everybody bid and made Game will result in a bottom. 

SACRIFICES: A sacrifice bid is bidding over the opponent's game (or slam) knowing that you will not make your contract. You hope that your loss will be less than the opponent's gain. It is better to go down 3 tricks doubled, not vul for a SCORE of - 500, then let the opponents bid and make a vul game for 620. Several things should be happening before a sacrifice bid is made. 
      w  Your opponents must be in a game or a slam.
      w  Either you or your partner must have a long suit with     some shape (singleton or a void).
     w You should have some idea as to how many points you  will lose playing the contract, as compared to how many points the opponents will gain if they played.
        Some sacrifices are obligatory because everyone in the room will take them. If you think that is the case, sacrifice. Also note, that your opponents might bid over your sacrifice bid. In this case they might be at too high a level and not make their contract. Pushing opponents to a level higher than they would care to go, is another reason to consider sacrificing. 
          In most cases do not sacrifice against a slam. Again the numbers tell the story. If you go down 800 or more (a reasonable amount for a sacrifice against a slam) and other pairs did not bid the slam but bid and made game for a 690 or a 680 number, then you will get a bottom. If you can really be sure that every one in the room will be in the slam, then your minus 800 will be a good match point score. Not that easy to evaluate. So most of the time try to set the slam.

PENALTY DOUBLES: It is important to keep in mind that in Duplicate the amount of gain or loss is not your primary concern. What is important is the number of pairs whose SCORES you beat or whose SCORES will beat yours. Doubling opponent's Part Score bid,  is sometimes a good strategy. This is true if you will get more points if you set them one or two tricks, then if you played and made a contract a Part Score contract your way. 
          Sometimes the penalty from doubling the opponents at a Part Score contract will SCORE more than if you bid and made game. These considerations will determine what position you and your partner will take. Duplicate players are calculating players. 
          Doubling of a No Trump contract or a slam is offten lead directing. But it could also be because you feel the opponents are up too high and will not make their contract. Or that the other teams sitting in your direction will be playing the hand and getting a positive SCORE. 
          Do not let the opponents steal the contract. If your team has the balance of power, that is, more than half the high card points, then it must either play the hand or double the opponent's bid. Victory or Death!

In rubber bridge, Chicago or team games, the Declarer's prime objective is to make the contract. Overtricks are of little concern. But in Duplicate,  the Declarer is competing against all the other teams in the room playing in the same direction. And to get more Match Points the Declarer has to get a higher SCORE than these other teams. This could be as little as 10 points. Note: 3 ! making 3 for a SCORE of 140 will beat 3 # making 4 for a SCORE of 130. Also note that if every team sitting in your direction bid 2 $ and make 2 for a SCORE of 110; and you bid 2 $ and make 3 for a SCORE of 140, then you get a top. Match Point scoring is relative scoring: the overtricks count. . . . in most cases.
          The bidding is over; your left hand opponent has made the opening lead; and Dummy comes down. After seeing the Dummy and reviewing the bidding, the Declarer asks the KEY QUESTION: should I try to make the contract at all costs and not consider overtricks; or should I try for overtricks and not worry about making the contract?  The bid and the cards determine the answer. Some conditions and solutions can be found in the following paragraphs.
         You in a good contract that no one else, or at least very few teams, will be in? If so, try to make the hand at all costs. You can even indulge in taking a safety play (conceding a trick to guard against bad splits). Do not take chances, make the contract. This is especially true if you are doubled or in a slam. Making these contracts will get Match Points. 
          You are in a bad contract. Let's say you are playing in 3 No Trump while you suspect everyone in the room will be in 4 $. How do you know? Look at Dummy: a 5 card Spade suit opposite 4 Spades in your hand and 3 losers. Teams playing this hand will score 420 points while your 3 No Trump contract will score 400 points for a bottom or near bottom. You must make the over trick at all costs and score 430. You must even risk making the contract for the sake of the over trick. In this case minus 50 is the same as plus 400; both a terrible Match Point score. 
          If in the above example, the overtrick is not really there no matter what you do, then you have to evaluate the hand differently. You must assume the other Declarers, due to bad distributions and splits, did not make their contracts. (If they did make their contract, then you will get a bottom.) In the above example 4 $ will not make if there is a 5 / 0 trumps split and if the Club finesse loses. Make this assumption and Play your contract of 3 No Trump accordingly. Remember what Mr. Marx said: "you have nothing to lose but your Match Points."
          You are in a good contract that everybody in the room will be in.   Now try to play the hand as the other players in the room would do it. Do not shoot for a top by doing something outlandish. Shooting is for cops and cowboys, not bridge players. If you must shoot, shoot for an average. And sometimes that might  involve jeopardizing the contract for an overtrick. 

OPENING LEADS: If the lead appears to be normal, then go ahead with a normal line of play. But if the lead is an unusual killing lead then drastic steps must be taken. You must recover the trick(s) lost by the opening lead by taking chances and making usual plays. If the opening lead gives you a trick, then try to maintain the edge.

GOING DOWN: Not every contract is made. If this were the case players would just bid and score. (In Bridge heaven every contract is made by the Declarer and set by the Defenders. Sounds awfully dull.) If you are doubled, you have to consider vulnerability and what the opponents can do if they played the hand. Example: you are doubled in a Part Score not vul, and the opponents are not vul. Then going down 1 for a lose of 100 will usually salvage some Match Points and might even be a top. Down 2 for a loss of 300 will be a bottom if the opponents do not have game; a top if the opponents have game. Down 3 for a loss of 500 is a bottom. Again vulnerability and scoring determine strategy.
        Another Declarer's KEY QUESTION: take a chance and try to make the contract but risk going down for a large loss; or play it safe and take the small loss?  The answer my friend is not blowing in the wind, but is at the table and in the cards. 

In rubber bridge, Chicago or team games the prime objective of the Defense is to defeat the contract. Overtricks are of little concern. The Defense is prepared to give up an overtrick for the chance to set the contract. But in Match Point play the Defense is competing against all the other teams in the room playing in the same direction. 
           After seeing Dummy and reviewing the bidding, the Defenders have to ask themselves the KEY QUESTION: to go all out and try to defeat the contract; or focus on restricting the Declarer 's  overtricks. The bid and the cards determine the answer. Some conditions and solutions can be found in the following paragraphs.
          If the Declarer in an excellent contract that no one else in the room will be in, then the strategy is to try to set the contract at all cost; do not worry about overtricks. Example: the contract is a game or a slam with relatively few points between the Declarer's hand and Dummy and few pairs are likely to make this bid. Now go all out to set the contract. Overtricks be damned!
          The Declarer seems to be in the right contract; that is everyone in the room will be there. (Evaluating at Duplicate is something like a popularity contest.) Now the Defense has to restrict the Declarer from making overtricks. That is keep the overtricks to a minimum. Zero is a good minimum.
         Another Defender's KEY QUESTION: when does the Defense take their winners? Even if you cannot set the contract make sure get all the possible tricks. A long suit in Dummy or a suspected one in the closed hand says its time to "cash out", that is take your winners. Overtricks do count.

OPENING LEADS: desperation leads can and often do result in burning (losing) a trick. This might be allowable in rubber bridge, but not in Duplicate. Going "against the percentage" might set the contract, but most of the time it will give the Declarer a gift of an overtrick. Save gifts for Christmas. In  Duplicate the overtrick is usually worth more than the contract. So keep the opening leads conservative, frugal and standard. Of course this does not mean the Defender should abandon inspiration or intuition. Good luck.