hide answers

Dan Barlow:

It's too early in the game to be basing my decision on the board position. I pair in both cases. In the long run, pairing everything (unless you can score 15-2) pays off.

John Chambers:

In this situation you should be at least at hole 44*. So you're about 12 holes short of where you need to be. On a 2 lead I would play a 4 from 2-3-4-4. If your opponent plays a 3 you may be able to gain multiple points if your opponent is trapped. True, your opponent may peg but you have to make up 12 holes. It's a chance you have to take. Or if your opponent pairs your 4 you can get a pairs royal for six points. You could pair the deuce but being 12 points behind your position, two points won't mean much, and your opponent may get better position with a pairs royal.

On an 8 lead I would pair it holding 8-9-10-10. You're probably saying "but you just said not to pair"! Because of the cards this is a different situation. Let's face it, if your opponent leads an 8 and you play a 10, you've set up your opponent for a three card run and a go with no recourse. That's four points. If you pair and he does get have the other 8, of course he gets seven points. But I feel there is a good chance he might not have the third because you each have an 8. That means that in eight cards (yours and opponent's) there are three 8s and in the other 40 cards there is only one 8. I'll take the chance. If it does work you have a good chance to hold your opponent from pegging and gain a little on your position.

DeLynn Colvert:

Holding 2-3-4-4 on a 2 lead, I'd play the 4. If paired you score six. If a 3 is played, pair the 3.

Holding 8-9-10-10 on an 8 lead? Play the 10, and if not paired, it is a safer lead back in the second sequence. Only the three 9s can beat this (or the two 10s). This is a "board" play. If playing all-out offense, pair the 8, but defense says play the 10.

George Rasmussen:

On a 2 lead, I would play a 4 from 2-3-4-4. I have it backed up, and I cannot afford to give pone a triple here for six points. If I give up six points here by pairing that deuce, my opponent is likely to be in great shape for the next deal.

On an 8 lead I would play a ten-spot from 8-9-10-10 for the same reasons. An 8 lead is generally an indication of other middle cards (6-7-9). I am holding a 9, which makes the play of the 10 safer than other choices. I can't afford six points on a triple. That Second Street zone is too close to take that chance. With a real cagey player the 8 lead might also mean that pone is holding 2-3-4 or 3-4-5 along with that eight-spot. In that case, when they make count 22 with the play of the 4, close the count with the 9 for 31. In the latter case, they will likely make the count 23 with the play of the 5. Close the count with the 8 for 31-2.

Michael Schell:

This is a classic situation. You have a double run (or a four-card run) and face a lead against an undoubled edge card. You must choose between pairing (thus risking a triple), or playing one of your end cards.

I start by checking the board. At 32*-36 I'm -12 to pone's +2. The J cut makes me -10, and the ten-point hand (worth two more than average) makes me effectively -8. Defense is probably still in order, but if I have a promising crib (maybe I dropped a pair of 5s!) an exchange of pegs could actually get me into position by the end of the deal.

For offensive or balanced play, pairing the lead is always a no-brainer in these cases. You get two immediate points, and though you'll occasionally be tripled, mathematically you'll come out ahead in the long run, despite the tendency of most players to favor leading from a pair where possible. For many years, the conventional wisdom held that you should also pair when playing defensively, on the grounds that you were "trapped" and would do better to minimize the number of losing cards. HALSCRIB, though, has taught me that there a bit more to it. Specifically, I would pair the 2 in the first case, but would refuse with a 10 in the second case.

The difference is that if you refuse the 8, you risk a pair and a run, whereas if you refuse the 2, you risk a pair, a run, and a 15-2. This is what's missed by most analysts who tell you to always pair away. Compare the objective risk here:

Scenario:	Objective risk:
	Pair	Refuse
2-3-4-4 against a 2 lead	(2 cards ∙ 6 points) =        12	(4 cards ∙ 2 points) + (2 cards ∙ 2 points) + (3 cards ∙ 3 points) = 21
8-9-10-10 against an 8 lead	(2 cards ∙ 6 points) = 12	(2 cards ∙ 2 points) + (3 cards ∙ 3 points) = 13

Pairing the 2 is clearly superior than refusing with a 4, even allowing for pone's likely bias in favor of leading from a pair. In the case of the 8 lead, the objective risk is about the same, but factoring in pone's leading bias, I agree with HALSCRIB that a 10 will give up about ¾ pegs less in the long run. Remember this idea: when holding a double run and facing a lead against your undoubled edge card, pair the lead unless you're playing off and won't risk a 15-2 by refusing the pair.

As you can see below, the bot decided to go ahead and pair the 8 anyway at this board position — not out of any tactical disagreement, but because it decided that more offense was called for.

Phyllis Schmidt:

In the first scenario, I'll play a 4. I don't want to give away six points if I play a 2 and the opponent has a 2 and I couldn't get away from further runs. If he does get the run I would play along and take what ever run was available to me.

In the second scenario I would match the 8. There's not much room for movement there. Playing any card is risky, so I'll take the two points.

Peter Setian:

Almost without a choice, I would definitely play aggressively in both pegging situations, which would mean pairing the 8 lead with 8-9-10-10 and probably pairing the 2 lead with 2-3-4-4, even though playing a 4 might produce more pegging points as play continues (an eight point crib seems almost certain given my toss and the cut, especially if pone trips the pair).

HALSCRIB:

In both scenarios, my projected endgame scores will be 115* with me dealing and pone at 114 with first count. On average, pone will peg and then count her hand for a win with three holes to spare. Considered in this light, both scenarios are identical and pose the same question: should I pair or not. The humans will probably couch their replies based on knowledge of whether the opponent habitually leads from a pair. Unfortunately, I won't have this capability until version 7. So my current answer lies in the expected average pegging gain for each player based on hand frequencies and board position which suggests that opponent would lead from a pair if available.

Here are my calculations holding 2-3-4-4 facing a 2 lead:

	Average pegging:
	Net	(pone/dealer)
2	+2.04	(2.73/4.77)
3	-0.34	(3.34/3.00)
4	+0.91	(3.04/3.95)

Note that my average pegging paring the 2 includes two point for the pair. Now for 8-9-10-10 versus an 8 lead:

	Average pegging:
	Net	(pone/dealer)
8	+1.76	(2.76/4.52)
9	-1.49	(3.97/2.48)
10	+0.51	(2.02/2.53)

Based on these numbers, I will pair in either case so that on the last deal I increase my chances of pegging out. If my opponent triples me, the game is still far from totally lost.

Panelists

Dan Barlow won the 1980 National Open Cribbage Tournament, and made the 1985 All American Cribbage Team. His cribbage strategy articles appeared in Cribbage World for many years, and can be seen on the ACC Web site. He also provides strategy tips at MSN Gaming Zone. He has written seven books on cribbage, two of which have been glowingly reviewed in Games Magazine. All, including his latest book Winning Cribbage Tips, are available at The Cribbage Bookstore.

John Chambers was one of the original founding members of the ACC. He is a Grand Master, winner of seven major tournaments, and author of Cribbage: A New Concept, He also directs three annual tournaments: the Ocean State Cribbage Classic, New England Peer Championship and Charity Cribbage Challenge.