If you follow cribbage much, then you're probably familiar with the writings of Dan Barlow. Perhaps you've read one of his books (Cribbage for Experts, Miracles on Fourth Street, Fun with Cribbage), or perused his online Pegging Pointers columns at the Web pages of MSN Gaming Zone and the American Cribbage Congress. Barlow is a former contributor to Cribbage World (the ACC's house organ), and also once published a short-lived independent cribbage magazine. I've always enjoyed his work, most of which features trenchant advice on pegging and endgames. His articles on Fourth Street tactics are particularly important, making him possibly the game's leading endgame theoretician. I appreciate Barlow's cogent, informal writing style: his cribbage books are easily the most readable of those currently in print, although this judgment must be taken amid the rather meager literary standards of this quaint pastime of ours.

Barlow has written a new book, which Sterling Publishing has just released as part of its Official American Mensa Game Book series. It is handsomely produced, in a digest-sized paperback edition which the publisher has given the unfortunate title Play Cribbage to Win, thereby inviting confusion with similarly named works by DeLynn Colvert and Joe Wergin. Although Barlow's previous books, all self-published, have been geared towards advanced players, this latest effort — the first of his books to be available through a commercial publisher — appears aimed at a more general audience. It includes chapters on the history and rules of cribbage, a series of exercises on hand counting, as well as a glossary and short articles on cribbage variants, computer cribbage and the ACC. All of these are perfectly adequate for someone new to the game. But my review will focus on the meat of the book: five chapters on discarding, pegging and endgames, followed by an illustrative game.

Discarding has never been Barlow's strong point, and Play Cribbage to Win is quite weak in this department. There are no discarding statistics nor any other kind of analytical tool. There is no mention of basic mathematical applications such as average hand and expected average. Barlow offers no specific guidelines or mnemonics to assist with discarding decisions, only broadly formulated advice like the following:

"If it's your crib...why not toss in something useful, like a pair, two cards that add up to 15, or touching cards?"

Lacking a sound mathematical basis for his discarding choices, it's understandable that he commits a few faux pas. Here's one of them:

"Holding 2-4-5-6-8-9 [as dealer], you can save 4-5-6-9, the most points possible. But the 8-9, tossed into your crib, may lead to a bigger payoff than 2-8."

It's true that on average 8-9 gets more points in your crib than 2-8. But a quick calculation should dispel any notion that it gets you the most total points:

Keep:	Toss:		Average  hand:	Chambers  average:		Average crib:			Expected average:
						Hessel	Colvert	Ras	Hessel	Colvert	Ras
2-4-5-6	8-9		8.46	9.54		4.74	4.6	4.77	13.20	13.06	13.23
4-5-6-9	2-8		9.93	10.50		3.58	3.6	3.82	13.51	13.53	13.75

Another example is 2-5-8-9-Q-K as pone. Barlow suggests you keep 5-9-Q-K, tossing 2-8 to your opponent's crib. You might want to run the numbers on this yourself. If you do, you'll see that tossing 2-8 here is simply awful. Much better is to keep 2-5-8-Q, which gets the same average hand as 5-9-Q-K, but with far less risk in the crib.

Not all here is bad, though. Barlow correctly suggests sacrificing two points to toss yourself 6-Q (instead of 2-Q) from 2-3-3-3-6-Q. That 2-3-3-3 combo is powerful, and keeping it together gets you about one point more in the long run. Another interesting hand is 3-6-7-8-9-J (where underlined cards are of the same suit). Here you're pone, and you're playing on. Which hand should you keep to maximize your offensive potential: the 3-6-7-8 flush or the 6-7-8-9 run? The former is worth nine points going in, but the latter gets sixteen points on any mid-card cut. Barlow recommends keeping the flush, which gets about a point higher average hand. It turns out that the flush gets the best expected average too, since tossing 9-J gives up only a trifle more than tossing 3-J. But you wouldn't know that from this book — it doesn't have discard tables.

Barlow is on surer footing when the subject is pegging. He devotes one chapter to the opening lead and its response, and a second chapter to the remainder of the play. The latter includes the following passage on three-on-one situations, which typifies Barlow's direct, common sense approach to pegging tactics:

"Sometimes as the play of the hand progresses, you'll find yourself holding three cards, while an opponent holds only one. This happens whenever you deal, your opponent's second card puts the count over 21, you say, go, and your opponent plays one more card.

"If, among your three remaining cards, you have a pair, you will be tempted to lead from that pair, hoping the opponent will be forced to pair you with his last card and that you can then triple.

"Take a look at the following example:

John		You
8		J (18)
6 (24)		Go
7 (31-2)		?

"Your remaining cards are Q-Q-K. If John's last card is a K, you are better off playing a Q now. And if John's last card is a Q, you are still better off playing a Q, as you will triple. So you reach for a Q. But wait! Is there any card John might be holding, which would make playing the K a profitable play?

"An A! If John's card is an A, playing your K will allow you to play your Qs consecutively, scoring 31-4, rather than 31-2. Which is John more likely to be holding with 6-7-8? An A, which improves the hand, or a K or Q, neither of which does? Probably an A. It can also be argued that if John had a face card, he'd have played it on your J, hoping for a go.

"Here's another example:

John		You
8		J (18)
8 (26)		Go
4 (30-1)		?

"Your three remaining cards are 9-9-10. Playing a 9 will pay off if John's last card is a 9 or a 10. And it very well could be. But it looks to me like he might have a 3 there. If so, you'll want to play the 10, which also pays off if John has an A or a 2.

"To summarize, when holding three cards, including a pair, to your opponent's one card, decide whether your opponent's card is more likely to match your pair, or be a low card — so low it'll allow you to play your matching cards consecutively. If the low card seems more likely, hang onto your pair."

Barlow is at his best when he writes about the endgame, and Play Cribbage to Win touches on several of the ideas that Barlow has explored in detail in his other books. This includes two critical concepts — holding for specific count and the endgame two-on-one — that are largely ignored by other writers. Additionally, Barlow devotes several pages to pegout situations, in which both players are within pegging range of home.

A good example of Barlow's treatment of endgame strategy and tactics is the following deal. It comes from the sample game, which, as is usually the case, is the most instructive part of the book.

PONE  (119):
Barlow

	A-3-6-9
	play:
	A  7  9  2  6  3  3 (31-4)
	crib:	9-K ?-?		cut:	J
	2-3-7-10

DEALER  (117*):
"Lorraine"

Barlow starts this deal two points away from home. Lorraine, his imaginary opponent, is four points away. Barlow is dealt A-3-6-9-9-K, and decides to keep A-3-6-9. The 6-9 guarantees Barlow a victory if neither player pegs out, and it lets him lead the 3 in complete safety — if dealer scores on it, Barlow will win immediately with his 9. The 6 would also be a safe lead, since it's covered by the 3 and the 9. But if dealer cannot score on a 6 lead, she will not hesitate to play a ten-card (if she has one), putting the count above 15. She might hesitate, however, to play a ten-card on a 3 lead, since these very often come from a magic five combo (an A-4 or 2-3). What would she play in lieu of a ten-card? Well, perhaps a 6 or a 9, either of which Barlow could then pair. Of course, there's no guarantee this will happen, but part of what characterizes expert play is the ability to create extra opportunities to win.

(Despite all this, I personally would have kept 3-6-9-K here, planning to lead the 6, then drop the 3 on a ten-card reply. The problem with leading the 3 from A-3-6-9 is that if dealer replies with a 7 or 8, you're now forced to play your A, giving dealer a shot at a pair or a 15-2. This kills you if dealer is holding something like A-4-8-9 or A-5-7-8. As it turns out, my plan would have lost in this particular case — after cutting a J, I'd have led the 3, which dealer would have paired — but nevertheless I think it would do a tad better in the long run.)

A wrench gets thrown into the works when Barlow cuts a J, moving Lorraine to within two points of home. At this point, most players would simply curse their bad luck and lead the 3 as planned. But Barlow realizes that a change in score often dictates a change in strategy. Since there is no longer any value in covering the opening lead — dealer will win immediately if she scores on it — leading the A is now just as safe as leading the 3. Is there any reason to favor one over the other?

Yes! Because if Lorraine can't pair the lead, Barlow will then have an opportunity to win with his next card. Leading the 3, he'll win if she replies with an A, 2, 6 or 9. Leading the A, however, he'll win on a 2, 3, 5, 6, 8 or 9 reply (in other words, anything but a 7 or a ten-card). Thus, Barlow leads the A.

Lorraine replies with a 7. Barlow then plays the percentage 9 (you'll recall he tossed another 9 to the crib, leaving six losers for the 9 compared to seven losers for the 3 or 6). Lorraine drops a 2. Now Barlow has another tricky choice: play the 3, making the count 22, or play the 6, making the count 25.

We know that Lorraine has no A, 8 or 9, so put yourself in Barlow's place and consider what might happen if you play the 6 here:

• If Lorraine says go, you'll drop the 3, peg one point and win unless her last two cards are 10-10, J-J, Q-Q or K-K
• If Lorraine plays a 2, you'll win. Your 3 will get a go, and her last card will put her in the stinkhole
• If Lorraine plays a 3, you win immediately with your 3
• If Lorraine plays a 4, you'll lose. She'll get a go on this play series (or a 31-2 if she also has a 2), then peg out with last card
• If Lorraine plays a 5, you'll lose. She'll get a go here, then one point for last card
• If Lorraine plays a 6, you lose immediately

There are a total of eleven losers (4, 5 and 6) if you play the 6. If she plays anything else, or says go, you'll probably win. What might happen if you play the 3?

• If Lorraine says go, you'll drop the 6, peg one point and win unless her last two cards are 10-10, J-J, Q-Q or K-K
• If Lorraine plays a 2, you'll win. Your 6 will get a go, and her last card will put her in the stinkhole
• If Lorraine plays a 3, you lose immediately
• If Lorraine plays a 4, you lose immediately
• If Lorraine plays a 5 or 6, you'll lose unless her last card is a 2 (it obviously isn't going to be a 3 or 4). In that case, you'll steal last card and leave her in the stinkhole. Otherwise she'll peg one point for the go, then one more point for last card
• If Lorraine plays a 7, you'll lose. She'll peg one for the go, then one for last card

There are ten certain losers (3, 4 and 7) plus seven likely losers (5 and 6) if you play the 3. You only win if she plays a 2, or plays a 5 from 2-5, or if her last two cards are 10-J, 10-Q, 10-K, J-Q, J-K or Q-K. Clearly the 6 is the correct play here.

Barlow plays a 6, then pairs Lorraine's trapped 3 to win. Her last card was a 10.

Conclusion

Play Cribbage to Win is a fine introductory text, perhaps best thought of as a sampling of the concepts of skilful play that Barlow has written about for the past several years. Barlow covers some important aspects of pegging and endgame play, and gives many interesting examples. The book is an enjoyable read, and there are quizzes every few pages, making it a lot of fun to work through. The typesetting and layout are reasonably good, but I personally would have preferred smaller type and larger margins (leaving more room for notes). A few typos and outright errors have crept into the book, perhaps the inevitable result of commercial publishing nowadays. But one of the things I've appreciated about Barlow's self-published books is their lack of gaffes.

The primary downside to Play Cribbage to Win is its scope. It is not a comprehensive treatment of the game. Missing are statistical tables, a discussion of averages and odds, and coverage of psychology and cheating. The treatment of discarding is cursory, and certain pegging tactics, such as J traps, are not covered at all. There is also no mention of board strategy, except as it applies to the endgame. Considering that most of today's top-rated players regard board strategy as the centerpiece of modern play, this is a significant omission.

If you're a novice or an intermediate player, I think you would do well to read Play Cribbage to Win in conjunction with DeLynn Colvert's Play Winning Cribbage (still the best single book on cribbage) and John Chambers' Cribbage: A New Concept. Together, these three books will give you all the essential knowledge you need to move to the next level.

If you're an expert player, or if you've already read Barlow's other books, you'll find yourself going over familiar territory. Nevertheless, it's always worthwhile to review important concepts, particularly when they're presented through new illustrative hands and examples (as far as I can tell, none of the material in Play Cribbage to Win is recycled). You may disagree with the text from time to time, but perhaps you'll also pick up some new ideas — I know I did.

Whatever your level of play, this book is well worth having, and well worth reading. And at $7.95, it's hard to knock the price.

- October 2000

See the article A course of study for more reading and study suggestions.