In this review, I discuss Modern Chess Openings 14 (MCO), revised and updated by Nick deFirmian. DeFirmian did much of the writing (roughly 32%, by my estimation), and undoubtedly helped to write other contributor's sections. Before anything else, I should say that MCO 14 is a vast improvement over its predecessor #13 in almost every respect. It has moved into the modern age with extensive use of databases and analytical help from computers. MCO's other contributors are John Fedorowicz, John Donaldson, Elliot Winslow with Steve Brandwein, and Bruce Leverett, all theoretically inclined and competent analysts. Fedorowicz handles Indian System specialties like the Benko Gambit (about which he has written a book), the King's Indian, and the Grunfeld. John Donaldson has been writing books and articles about the Slav Defense for years. Winslow and Brandwein collaborated on the Queen's Gambit and romantic King's Pawn openings, and Bruce Leverett authored the section on his specialty, the English Opening.

I am rather surprised that the one-volume comprehensive encyclopedia continues to be most players' gateway into the world of openings. But that is definitely the case. Thus far, and probably for several years, electronic versions of such a book are falling far short of the depth of, and intelligence behind, the printed versions. The real question, for those who want to study openings, is the right balance between the comprehensive approach and specialized individual books. There clearly exist large groups of players who prefer either or both. A book like MCO has the advantage of putting all the material in one place, with a human being deciding on best lines and mostly eliminating the rest. This is something that a computer, even with advanced calculating and assessment ability, still can't even approach doing; it is far stronger as a practical competitor. DeFirmian, who worked to prepare Deep Blue against Kasparov, points this out, explaining that "chess openings are very difficult for computers unless they simply repeat human moves." Still, to supplement books, magazines, database searches, and human judgment, deFirmian used computer analysis, primarily to eliminate tactical oversights.

MCO begins with a brief introduction to notation and an example of what simpler lines a near-beginner might want to start with. The book is divided into five major sections (Double King Pawn Openings, Semi-Open Games, etc.). Most individual subsections have a fairly short, popularly accessible, introduction to the history and strategy of the variation in question, with the exception of major openings, where several subsections are introduced together. This attempt to appeal to inexperienced and intermediate players is consistent with the broader, less intensely analytical approach we will discuss below.

There are two versions of the 14th edition, one published by McKay (henceforth MMCO) and one by Batsford (henceforth BMCO). As far as I can tell, the prose, moves and analysis are identical, with the extra pages in MMCO attributable to slightly more space between notes, blank pages separating sections, a condensed Index in BMCO, and the like. Differences in appearance will be discussed shortly.

Although I take this step with great trepidation (fearing howls of protest from both sides), I want to compare MCO with its nearest competitor, Nunn's Chess Openings (NCO). This is the best way to clarify certain issues, and in any case the reader should benefit from such a comparison. I took a lot of time comparing the two, so as to eliminate hasty judgments, but there are always things to miss, so I beg forgiveness in advance. I should say up front, however much it looks like hedging, that both books have advantages and disadvantages which will probably make them more appealing to different sets of readers.

Let's start with appearance and readability (in the visual sense of the term). It is remarkable that the Batsford version of MCO is so much more readable than the McKay version. BMCO is physically wider and a bit taller, so that it can fit the same material on a page using a larger font for the main moves. More importantly, the Batsford type is darker with much better contrast against its white background. In appearance and readability, it is therefore superior to MMCO, and also to NCO, but for somewhat different reasons. NCO uses a smaller font and a more crowded page. Surprisingly, with what is often two to three times the analytical material per page (see below), NCO's readability is quite competitive with MMCO's, due to NCOs darker type and the color of its pages, which is not as white as BMCO's, but not as yellow as MMCO's. On the other hand, both MCOs have a great deal more space between the main moves and between moves in the notes, which definitely makes for more comfortable reading. So to me, the conclusion on this issue is that BMCO is far and away superior (easy on the eyes, that is) to either MMCO or NCO. But NCO is at least comparable MMCO, with the latter's open space only slightly more important than NCO's superior contrast. As a point of information, MMCO uses "traditional" algebraic notation with letters (e.g., 32.Qxc3), whereas both BMCO and NCO use figurine algebraic (the piece given by a pictorial representation, the rest of the move by letters and numbers).

Now we move to a subject that means a great deal to advanced players and most tournament players, but perhaps relatively less to the reader who plays less frequently, namely: the amount and character of the material presented on any given opening. This is a very complicated issue, but I can make a few general statements up front. Overall, unquestionably, NCO presents many more moves and alternatives than MCO (remember that both MCOs have the same material). More precisely, the number of moves per page in NCO tends to be two to three times that in MCO, which is a very significant factor in NCO's favor. By contrast, MCO includes verbal comments in its notes, and NCO does not. MCO also cites all game references (players, sites, and dates), whereas NCO does so "only when they are of particular interest" (see below for some figures). MCO tends to have about one game per note, often with commentary in the remaining notes. Finally, MCO has longer prose introductions to sections. We will expand upon all of this next.

Let's examine the total of pages and their nature. To begin with, 734 pages (MMCO) and 708 pages (BMCO) are certainly more than NCO's 544 pages. But it's interesting to see how these are allocated. MMCO (I received this first, and have based quite a few numbers on it) devotes 14 pages of introductory prose to its broadest divisions, and about 127 pages (extrapolated from half the book) to its section introductions. NCO has only 22 pages of section introductions. Thus, about 120 of MCO's page lead is accounted for by introductory prose. Analytic- and information-oriented readers will consider this in NCO's favour: suddenly the number of NCO's information-packed pages approaches that of MCO's when it comes to games and analysis. Readers who like prose and descriptions, however, will consider this an advantage of MCO. Since NCO's introductions are done with smaller type (and smaller diagrams), it has somewhat more room per page to introduce variations; but MCO, even larger type, still offers much more prose material (and larger diagrams) in that area. The Najdorf Sicilian, for example, gets its own three pages of prose in MCO, apart from the general Sicilian introduction, and there is only a general "Sicilian: Main Lines" introduction in NCO. The French introduction gets 6 pages in MMCO, 5 in BMCO, and 1 in NCO. While MCO's introductions give historical background and reviews more general strategy as well as assessments, NCO devotes its space to key variations, some strategy, and assessments, the latter of which are concise but very informative.

So if the number of pages devoted to moves are about 40 more for MCO (discounting for the excess pages in MMCO), why is the analytical coverage in NCO (in the sense of moves) so much more detailed? For one thing, MCO uses more space to list the main moves, leaving less room for the notes. Also, NCO has few if any pages that contain only a few lines of notes, while MCO has quite a few more (especially MMCO). Probably the most important difference consists of the space between lines. Consider these figures for a few (randomly chosen) full pages of notes, each covering the same general subject as their counterpart. For the first page of notes I looked at I got these results: MCO: 147 full moves, 20 game references (in the notes), and 59 words of prose (ranging from things like "with a good game" to "taking the d-pawn allows Back good play on the dark squares"); NCO: 300 full moves, 25 game references, and no prose (always the case with NCO). In the next comparison, I got: MCO: 118 full moves, 18 game references, and 81 words; NCO: 344 full moves and 1 game reference. The third example: MCO: 115 full moves, 17 game references, and 99 words; NCO: 295 full moves and 15 game references. Finally, from a page of popular main lines: MCO: 109 full moves, 13 game references, and 164 words; NCO: 282 full moves and 10 game references. Looking through NCO, I found quite a few pages with 0-2 game references, but also ones with 5, 8, 12, 13, 21, and even one with 34! My guess would be an average of 8-10 game references per page of notes. Looking through MCO, I found it more consistent, with 15-20 references on most pages, and between 60 and 110 words of prose. (Full pages were created from adding fractions of pages on adjoining pages when necessary).

The fairly obvious conclusion is that NCO has a two-to-three times advantage in moves and analysis, but no words; whereas MCO has more words and game references, but still has considerably less total material per page (i.e., total characters, including moves, prose and references).

The specific nature of the material and its distribution is very interesting. Although I inevitably disagreed with some assessments in both books, I think they deserve high marks for their identification of the most important lines. The NCO authors do have an edge in playing strength and published theory, but deFirmian's crew is well chosen for their areas of expertise. I went over two major openings in detail, ones in which I have some experience, and snippets of two others. As expected, NCO tended to have more options to the main moves; a typical situation would arise with MCO giving equality in a main line, and NCO giving equality to the same line, but also showing one or two other paths to equality or inferior continuations that MCO ignored. An interesting aspect of this comparison, which reflects well on the deFirmian team's judgment, is that for most lines, despite the disparity in total moves, MCO's ultimate assessment tended to match NCO's. MCO neglected lines which should have been included more often than NCO, but not dramatically so (not including "normal" options in the notes). And although both books had surprising amounts of original analysis, this seemed to be lengthier more often in NCO. When the two sides disagreed, it was often because one book has extended the key example a few moves further; I felt that in these cases, MCO had the better assessment just as often as NCO. See other figures below.

I should mention some things related to convenience, which may be relevant to some but not to others. NCO's treatment of transpositions is much more thorough than MCO's, which is itself a virtue and also helps one to negotiate NCO's denser presentation. And while NCO's smaller type is more difficult to read than BMCO, as described above, I was reminded of and pleased by NCO's policy regarding the placement of notes as I was making these comparisons. NCO tries to place all such notes on the same page or the page across from the main moves, requiring no turning of pages to see the notes. Only rarely do they fail to achieve this goal. MCO is not at all as bad as the Yugoslav Encyclopedias in this respect, but one still has to flip pages much more frequently than with NCO, particularly in theoretically dense variations. This is in part due to the space consuming listing of the main moves. MCO's back Index lists openings according to their names, while NCO has a small type move-tree Index with footnotes. I suspect that most average players (certainly not bookish players or theoreticians) will probably prefer MCO in this respect. Both have less detailed Tables of Contents with named variations. There are some mistakes with the BMCO Index which probably arose from translating the MMCO Index, e.g., BMCO lists the Reversed Queen's Indian as being on pages 721-23 and the Reversed King's Indian on pages 710-14, although the book's variations end on page 703. That only applies to a few lines; but more significantly, the Reversed Queen's Indian doesn't appear at all.

What about the distribution of material? On the whole, NCO pays relatively more attention to contemporary variations played by leading players, as well as to theoretically controversial lines. When you exclude section introductions, the two books (using NCO and MMCO) devote almost exactly the same number of pages to the Sicilian Defense, NCO devotes about 50% more pages to the King's Indian, and 50% more to the Modern Benoni. In contrast, MCO tends to stress more traditional openings. It has, e.g., almost twice as many pages on the Queen's Gambit Declined and 50% more pages for older 1.e4 e5 lines like the King's Gambit, 4 Knights, Latvian Gambit, Ponziani, Danish Gambit, etc. There are some exceptions, with MCO being just as "cutting edge" in some sections, but even when you break down specific openings, NCO tends to have a more contemporary feel while MCO devotes relatively more space to traditional lines. Naturally, a given reader may prefer either approach.

Another distinction that might interest some readers is MCO's greater emphasis on most eccentric openings and certain little-played ones (with some qualifications noted below). Some of these openings even get separate introductions. Examples: Grob's (1.g4), with half a page and 8 notes in MCO vs. three lines of a single note in NCO. MCO gives a page and a half combined to 1...b6, 1...a6, and 1...g5 in response to 1.e4 vs. 1 row for 1...b6, a 3-line note for 1...a6, and no mention of 1...g5 in NCO. Somewhat more respectable openings like Latvian Gambit, 1.b4, and 1.b3 similarly receive much more attention in MCO, and the Blackmar-Diemer Gambit gets a half page and 10 full notes in MCO vs. 8 lines of a note in NCO. The Danish Gambit coverage in MCO, for a further example, swamps that in NCO.

But there are some exceptions and caveats here. For one thing, one has to remember that the same amount of space in both books contains more material in NCO. Then there are the exceptions. The Snake Benoni (5...Bd6 in a Modern Benoni) is given a full row in NCO and is not mentioned in MCO. Some surprising examples in frequently played openings come in the early Queen's Pawn Games. 1.d4 d5 2.Nf3 Nc6, of interest to Chigorin Defense players, has 3 rows and 8 notes in NCO and no mention in MCO. More popular are Bf4 lines: In NCO, 1.d4 d5 2.Nf3 Nf6 3.Bf4 gets a row and 3 notes, while 1.d4 Nf6 2.Nf3 e6 3.Bf4 gets a row and two notes. But neither line is even mentioned in MCO, as far as I can see. At least 1.d4 Nf6 2.Nf3 g6 3.Bf4 is present in MCO, but gets only a column and 3 notes, whereas NCO has 4 rows and 18 notes. NCO's coverage of 3.Bg5 swamps MCO's as well. Perhaps the best conclusion is that MCO is superior in truly eccentric openings; but that in little-played openings, it depends upon the territory, with MCO doing a better job with 1.e4 openings and NCO doing a better job with 1.d4 openings.

Who are these books for? Both MCO and ECO have claimed to be useful for players of all strengths. That may be true, but their usefulness for players above 2200-2300 drops off rapidly, in my opinion. With it's denser analysis, NCO may have an edge with stronger players, who are also seeking help with the very latest lines; those players will also be drawn to the NCO conveniences mentioned above. MCO's user-friendly prose, its appearance, and its coverage of older 1.e4 e5 lines will probably appeal more to traditionally-oriented and non-theoretical players. If one wants to know the "best" line in a variation without distracting details, you might prefer MCO. Those who wish to investigate options, which can be as good or worse than that "best" line, may lean towards NCO. But these qualities are relative, and subject to argument.

I hope that this review both gives the reader an introduction to MCO, and fairly represents the strengths, weaknesses, and options of both MCO and NCO. Both books have good qualities, and I don't think that you can go wrong with either work, or with both.

ADDENDUM:

My MCO/NCO comparison attracted more responses than usual. One reader noticed the large difference of material devoted to the popular 4...a6 variation of the Slav (1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3 a6). Indeed, NCO devotes a whole page exclusively to this move, considering 9 responses and 4 in detail. MCO has only a 4-line footnote that considers only one response, 5.Bg5. And although MCO often reaches conclusions similar to NCO's despite less material, that is not true in this case. MCO claims a small advantage to White after 5.Bg5; but both NCO and current theory disagree, NCO offering 2 paths to equality.

This is an extreme example of my point that NCO tends to be more detailed in contemporary openings, especially those arising from 1.d4, whereas MCO tends to have more coverage of historically important 1.e4 e5 lines. While 4...a6's popularity is fairly recent, it has still been important for at least 5 years. A possible explanation for the discrepancy in these contemporary 1.d4 openings would be that, aside from DeFirmian and Fedorowicz, the MCO authors are less involved with active tournament play and opening trends.

A chess teacher wrote to say that NCO was useless for students, due to its lack of explanations and readability. I would say instead that there are a least two types of students in this regard: those who are information-oriented and want to know as much as they can about the lines they play or intend to play, and those who blanch at crowded notes and trees of variations without words. Among very young students or adults who play fairly casually, it's likely that the second group is more numerous; as players develop more, they probably gravitate more to the first category.

Finally, I have been using both books recently to see what "general" theory thinks about a variation I'm addressing. In doing that, I realized that I had neglected to mention that NCO lists its ultimate assessment for primary lines at the end of each row, a handy feature shared by the Encyclopedias and Informant pamphlets. Thus, for an overview, one needn't read the crowded footnotes to see how a top-level line is assessed. In MCO, this is really not an issue, since its footnotes tend to be about one line, with large type. In general, no one wrote anything that strongly contradicted the gist of the review.

YOU CAN FIND THIS BOOK AT