This section is intended for those new to the game. Basically, we'll look at discarding and pegging play, with a hint or two about positional play.
Discarding Strategy
Beginners usually find discarding more difficult than pegging because there are more choices. There are 15 ways to discard from a hand, while you never have more that four choices when pegging. Besides that, there are lots of considerations that affect your decision. You want your final hand to score well. You want your discard to either help or hinder the crib. You want to keep good pegging cards. Finally, you want to weigh all this against the game situation, which dictates whether you should be cautious at the moment, or aggressive. This is too much to handle at once, so let's break up the problem into manageable pieces.
For starters, let's look at discarding from the perspective of making your final hand score the best possible. This is usually the most important consideration in discarding. Hand scoring for dealer and pone (that is, excluding peg play scoring and crib scoring) accounts for over 60% of the points in cribbage.
The unknown in the final hand score is the cut card. A computer can average in the effect of the cut when it assesses a discard. For you and me it's a different story. We rely on experience and rules of thumb. A good rule of thumb here is just to look for the four cards that by themselves add up to the most points. The scoring rules covered previously for five-card hands work just as well with four-card hands. You can estimate about a quarter point per jack for possible "nobs". It still may seem like a lot of work to evaluate all the possible four-card combinations in a hand, but after playing awhile, you'll soon notice that most of the combinations aren't even worth looking at. At this point, just try to narrow down your choice to two or three possibilities that score the highest.
Next, consider the effect the discard has on the crib. As dealer, naturally you want to throw good cards, since you get to score the crib later. What are good cards? A pair or two cards that form a 15-count are good cards, since they are a sure two points. Other good cards are two cards consecutive in rank, like 6-7 or 3-4. These increase the chance of runs in the crib. Also, a 5 with any other card, a 2-3 combination, and an A-4 combination are all good discards. 5's are easily the most important cards in the game of cribbage. The abundance of 10-count cards (face cards and tens) in the deck means that 5's will be part of 15-count card combinations (which are worth two points) more frequently than any other rank.
As pone, you take the opposite view. You want to avoid throwing the card combinations that help the crib. If you can, you would like to throw cards like 10-K, 9-K, or 9-Q. As a rule of thumb, almost never throw a 5 into your opponent's crib.
For a detailed comparison of discards, see the tables in the Programming Notes section.
Now look back at the two or three four-card combinations that score highest and check the corresponding two-card discards for their effect on the crib. If you still have two choices that seem pretty even, a good rule of thumb for breaking the tie is to save the hand with lower ranking cards. These tend to improve pegging potential. Another tiebreaker is opponent tendencies. If your opponent is very conservative, and rarely throws good cards in your crib, you should tend to avoid wasting your own good cards by throwing them there also.
Another good rule of thumb when discarding is never to break up a "double run". A double run is a combination like 2-3-4-4 or 9-10-10-J, i.e., a combination of two runs and a pair. This guarantees you at least eight points and has potential to score much higher with a favorable cut.
Let's look at a few examples, starting with an easy hand. Suppose as dealer you hold 2-3-4-5-9-10. There are three ways to save five points: 2-3-4-9, 2-3-4-10, and 3-4-5-10. The corresponding discards are 5-10, 5-9, and 2-9. Since it's your crib, the 5-10 discard is best, since it's two sure points and 5's are great crib helpers. If you are pone with the same hand, then the 2-9 discard is best.
Now suppose as dealer you hold 2-3-4-4-7-7. Keeping either 2-3-4-4 or 4-4-7-7 saves eight points. The corresponding discards are 7-7 and 2-3. These are both good discards. Save the double run. It's a good rule of thumb. But what if you are pone? You hate these kinds of decisions. The safest discard is 4-7. In situations where extreme caution is dictated, many players will choose this, even though it means breaking up a double run. If it's early in the game, throw the 7's.
Finally, a trick hand. As dealer or pone, you hold 2-2-3-3-6-6. It seems clear cut, since 3-3-6-6 is worth eight points. You might be surprised to find that experienced players will frequently discard the 6's. The rule of thumb fails us here, since 2-2-3-3 has tremendous scoring potential. Sometimes two birds in the bush are better than one in the hand. If cribbage were too easy, it wouldn't be nearly as interesting.
A last word of encouragement on discarding. If you are a beginner, it can be frustrating to play an opponent who never needs more than a couple seconds to discard. First, he has probably been playing for years. He should be a little faster at it. Second, realize that this can work to his DISadvantage. There may be some type of hands he discards one way just because he's always done it that way, even though it's not the best way. You are new to the game and still think before acting. Make it work to your advantage.
Pegging Strategy
Although fewer choices make pegging seem simpler, good rules of thumb are harder to come by. Opponent tendencies should have a larger influence on pegging decisions than on discarding decisions. Keep this disclaimer in mind when reading on here.
Pegging play, by the way, is the weakest part of the program's game. It is a little too predictable and does not follow the above advice -- it does not try to adjust its style according to your playing tendencies.
A 4 is generally considered the best lead by pone. This prevents the dealer from reaching a count of 15 on the next play and forces him to make a play that gives the pone the opportunity to reach 15. As pone, however, you should occasionally decline to lead a 4, to keep your opponent off guard.
Leading from a 5-combination as pone is also common (e.g., leading a 3 if you also have a 2). Almost a third of the deck is made of 10-count cards, so dealer likely has one or more. If he plays one, pone can score two points. The dealer knows this, however, and will often decline to play a 10-count card next.
A 5 is a poor lead. You might be tempted to lead a 5, for example, from 5-J-Q-K at the end of a game if you are 11 points from going out. You are trying to lure your opponent into playing a face card so you can pair up. The unusual lead instead will trigger all kinds of warning signals in your opponent (even if you "cleverly" pretend the 5 slipped out accidentally). Instead, play one of the face cards, and hope that your opponent will routinely (and incorrectly in this instance) play a 5 to score two points. Then you may pair the 5 for your eventual winning points.
If you are dealer, you should generally assume that pone has a 5 and try to trap it. The pone almost never throws a 5 in your crib, so he is more likely to have a 5 than any other card by rank. He will hold a 5 after discarding about four hands out of every ten. As pegging proceeds, the play of 10-count cards, 4's, or 6's by pone should reinforce this initial assumption, while other plays weaken the assumption.
Though obvious to say so, remember what you discarded and what was cut. Those are three known cards your opponent cannot hold. You'll sometimes hear a real-life opponent say "Pair the deck" as his first lead pairs ups the cut card. This has several possible implications. He may be playing the percentages to reduce your chance of pairing his play. That is, he is including the extra known cards in his play decision. Or, psychologically, he may be trying to heighten your anxiety that he got a favorable cut. On the other hand, because he could have made the lead without comment, he may actually be trying to lure you into pairing the card. This should sound a little like the poison duel in "Princess Bride". Just make the simple assumption that he is playing percentages.
As the pegging winds down, and you have seen a few opponent cards, you should assume the opponent's remaining card or cards fit best with what he has shown. E.g., if your opponent has shown 6-8-9 already, guard against a 7 first, then against a 6 or a 9 next. If you play a 2, and he surprises you by pairing the 2, don't consider that as losing points. When the hands are scored, his total will be poor and that will more than offset those two pegging points.
From the opposite point of view, if you've got a discouraging busted hand, stay sharp in the pegging. Because your cards don't fit well together, your opponent likely will be mis-guessing what you have. You'll have a good chance to salvage some points.
A good piece of pegging advice for beginners is NOT to make scoring plays automatically. While you may end up making the play anyway, you should always do a damage assessment first and consider how your opponent will follow up. Give your opponent credit. Particularly early on in the pegging, when he has more freedom in card selection, he will be trying to guard against your possible scoring plays.
The toughest decision in pegging is whether or not to pair an opponent's card. You stand to gain two points, but risk giving up six points. Four points is a painful swing in the pegging. Tournament players are overly cautious in this respect, probably thinking that skill will carry the day anyway. Beginners, on the other hand, frequently pair up without a second thought. Probably the best advice here is to watch your opponent's style of play. If your opponent tends to play from his pair early in the pegging, then you should tend to decline the early pairing opportunity. Positional strategy, discussed next, affects both discarding and pegging.
Positional Strategy
Experienced cribbage players often use the phrase "playing position" (or "playing shape"). This means playing aggressively or cautiously depending on the game score, i.e., on the peg positions on the board. I'll use an analogy here. As a driver, you should know enough not to keep your eyes on the bumper of the car in front of you. You should look well ahead for potential accident situations.
In cribbage, just watching to see whether you are a couple points behind or ahead of your opponent is like staring at the bumper in front of you. You may think you're in a close game and you may even gain a couple points as you turn the corner onto home street, but then kapow! An accident. There you are at 103 points and your opponent is at 99. But he's dealing next and you are pone. Think about it. Who's probably going to win?
The dealer on a typical hand scores 16 combined points. The pone typically scores 10 combined points. So he is likely to be around 115 as pone next hand, while you sit glumly dealing at 113. Then you think back a couple hands to the furious pegging play where you both scored a couple of times on run combinations and you had a net gain of a couple points. What if you had blocked the runs by playing your king? Take away 10 points from your score and 8 points from his and the score would be 103 to 107. Then you could have had a real battle down home street.
This is what positional play is all about. Think ahead. Try to maneuver so you will be dealing when you have 95-100 points. Sometimes it will take risky pushes in the pegging. Sometimes it will mean breaking up your hand to throw real clunkers into your opponent's crib.
The final tip here comes from computer simulations. Most cribbage games require 8, 9, or 10 hands to play, with 9 hands as easily the most common. Not surprisingly, the first dealer has a significant statistical edge in 9-hand games, since he has the extra deal. This suggests that the first pone in a game should do his level best to steer the game toward an 8- or 10-hander by risky or cautious play. The very best players probably do this instinctively.