and eight. Next to cards in sequence come cards only one or two points apart, as the cards of the opposite player may fill up the gap, and convert them into sequence cards. Of single cards, a five is the most likely to score, inasmuch as there are sixteen tenth cards to four of any other denomination, and the chances of its forming part of one or more fifteens are therefore considerable. The cards which are least likely to make for crib are king and ace, inasmuch as nothing save queen, knave can convert a king into a sequence card, and nothing save a two and three can convert the ace into a sequence card. The best cards for the non-dealer to throw out are therefore a king or ace, and some second card so far removed from the first that the two cannot form part of the same sequence. King or queen, with nine, eight, seven, six, or ace, are good "baulking" cards; likewise two or ace, with seven, eight, nine, or ten. The non-dealer should never throw out a knave if he can help it, as the start may cause it to score a point for "nob." In like manner, the non-dealer should avoid laying out two cards of the same suit, as he thereby runs the risk of a flush in crib.
In the play of the hand the guiding principle should be to give to the adversary the fewest possible opportunities of scoring. Bearing this in mind, it will be seen that the best card to lead at the outset is an ace, two, three, or four, as the second player cannot make fifteen, and the chance of doing so will revert to the first player. A five, on the other hand, is a very bad lead, inasmuch as, from the greater number of tenth cards in the pack, it gives the second player the best possible opportunity of making fifteen. If the leader holds duplicates of
ace, two, three, or four, one of them is a very good lead; for if the second player should pair, the leader will have the opportunity of making a pair-royal. Failing pairs, if the leader hold two cards which together make five, i.e. two and three, or ace and four, it is good to play one of them, when, if the adversary play a tenth card, the leader will be enabled to make fifteen. Likewise, if the leader hold a six and a three, a four and a seven, or a three and a nine, the first card of the couple is a safe lead, for if it is paired, the second will make fifteen. On the other hand, should the second player play a tenth card to the lead, the first player may pair it with perfect safety, for no pair-royal can be made without overpassing the limit, thirty-one.
As regards the second player, he will generally do well to make fifteen if he can. If a low card, i.e. a four or less, has been led, he has no choice, in the majority of cases, but to leave the fifteen to his adversary; but he should carefully avoid playing such a card as will enable the adversary to score not merely the fifteen, but a pair or sequence in addition. On a four led, for instance, it would be very unwise to play either a six or a five, as in such cases respectively, a five or six played by the first player would give him both fifteen and a sequence. On a three it would be equally wrong to play a six; on a seven a four; on a nine a three, or on an ace a seven; for a like card played by the first player would give him both fifteen and a pair. Again, it is in general unwise for second player to play a close card (i.e. next or next but one to the lead), as he thereby gives the adversary the chance of a "run." If he is in a position to continue the run, he may of course play
a close card with impunity. The points of "five" and "twenty-one" are to be avoided, as a tenth card played by the adversary will in such case make him fifteen or thirty-one. Similarly, it is bad play to make fourteen or thirty (i.e. one short of fifteen or thirty-one) with an ace; to make thirteen or twenty-nine (two short) with a two; twelve or twenty-eight with a three; eleven or twenty-seven with a four; as in either of such cases, should the adversary be able to pair, he will thereby score four holes. The only exception is where the player chances to hold two deuces or aces, in which case it will be worth while to make twenty-nine or thirty respectively with one of such cards, on the chance of the opponent holding no deuce or ace, in which case the first player will himself gain the advantage of the double score.
Some discretion is needful in pairing the card first led, as the first player may be aiming at a pair-royal, and the temporary gain of two points may be counter-balanced by six to the adversary. Where, however, the player holds two of the card led, it may be paired without hesitation. The chances are much against the dealer's being in a position to make a pair-royal, and if he should, it can be capped (unless the card be over seven) by the double pair-royal of the last player.
A further point to be considered, in deciding whether to make a pair or sequence, is the state of the score. It is calculated that the non-dealer, at five-card cribbage, should make, on an average, six in hand and play; the dealer eleven, or a shade more, in hand, play, and crib. When each has dealt once, they should stand abreast at seventeen
to eighteen, and so on throughout the game. The player who has maintained this average is said to be "home," and a player who is in this condition at an advanced state of the game, should run as few risks as possible; should avoid pairing, play wide cards to avoid sequences, and so on. This is known as "playing off." If, on the other hand, he is behind his proper position, his chance of winning will depend, in a great degree, on his making more than the average number of points in play. In such case, he should embrace every opportunity of making a fifteen, a pair, or a sequence, even at the risk of giving opportunities to the enemy. This is known as "playing on." As there are sixteen tenth cards in the pack, and ten out of fifty-two are dealt, the probabilities are in favour of the players holding originally three between them, and this probability should be borne in mind, as the so doing will often help the player to a thirty-one. Suppose that the leader starts with queen, and that the other player has no tenth card, but has a seven and a four, an eight and a three, or a nine and a two. In such case it is good policy to play the seven, eight, or nine. If the first player again plays a tenth card, the second will be enabled, with his small card, to score thirty-one. If the second player have no tenth card in his own hand, the probability of his opponent holding more than one is proportionately increased.
It may be useful to illustrate these elementary principles by the play of a couple of imaginary hands. Let us suppose that A (elder hand) has the queen and six of hearts, nine of clubs, eight of diamonds and seven of spades. And B (dealer) the
ace and ten of hearts, ten of clubs, five of spades, and four of diamonds.