What if there were no King’s Gambit?

by admin on September 17, 2013

At Mike Splane’s latest chess party, on Sunday, I saw Richard Koepcke lugging around a ginormous chess book that was about the size of a phone book, and I asked him what it was. To my surprise, it was a new book on the King’s Gambit! I refer to John Shaw’s The King’s Gambit, published this month by Quality Chess.

Anyone familiar with me knows that I am not a theory hound and I am not the sort of person who races out to buy the latest opening book. Quite the opposite. But where the King’s Gambit is concerned, I’m willing to make an exception. This is my main weapon against 1. e4 e5, and naturally I want to keep abreast of everything new. And fortunately we have had some spectacular books on the King’s Gambit in recent years, truly immense labors of love that have brought the theory into the twenty-first century. First Thomas Johansson’s book The Fascinating King’s Gambit (2004) and now Shaw’s.

What can make an opening book worthwhile in the era of chess computers and databases? In a word, viewpoint. (Well, in Russian that’s two words, tochka zreniya.) What you want from an opening book is a clear authorial voice that steers you through the maze of variations and says, here is what is worth paying attention to, here are the main ideas, etc. Johansson has that in abundance, and from my quick five-minute skim, Shaw does too.

It’s really way premature for me to review Shaw’s book because I have literally seen it for only those five minutes. But I can tell you the most interesting thing for me. I immediately looked up the chapter on the Bishop’s Gambit, 1. e4 e5 2. f4 ef 3. Bc4, and I was stunned to find a short chapter (compared to the rest of the book) called “The Refutation of the Bishop’s Gambit.”

How ironic! Johansson’s King’s Gambit book is a paean to the Bishop’s Gambit. He is a reformed King’s Knight’s Gambit (1. e4 e5 2. f4 ef 3. Nf3) player, who became disenchanted with that line because he couldn’t get enough compensation for the pawn in the old-fashioned … g5 variations. In the computer age, Black can grab the pawn and hold on. Now along comes Shaw with the opposite point of view: 3. Bc4 is busted and 3. Nf3 is in again.

To cut to the chase, Shaw’s claim is that after 1. e4 e5 2. f4 ef 3. Bc4 Nc6! White is at best fighting for equality. Obviously a careful evaluation of this claim will require me to buy the book and think long and hard about it, but here are some off-the-cuff thoughts.

First, I do have some prior experience against it, enough to be very wary. Back in the 2000-2005 era, when the Santa Cruz chess club was still alive, I played a lot of King’s Gambits, and Ilan Benjamin (who now seems regrettably out of the chess scene) went through a phase of playing 3. … Nc6. The main point is to invite 4. d4 Nf6 5. e5 d5! My first instinct was to play 6. Be2, but the game just went to a boring and lifeless equality. Then I looked it up in what books I had then (not even Johansson yet) and saw that 6. Bb3 was supposed to be the main line. The fun was back! After 6. Bb3 Ne4 7. Bxf4 Qh4+ 8. g3 Nxg3 9. Bxg3 Qe4+ 10. Kf2 Qxh1 (the old zigzag trick) White gets to sacrifice the exchange. In my opinion, exchange sacrifices are the life’s blood of the King’s Gambit. I looked at this line a lot (going out as far as move 30 sometimes!) and concluded it was fine for White.

But then, somewhere along the line, I put it on Fritz, and Fritz threw cold water on everything. Instead of 6. … Ne4, 6. … Bg4! throws a monkey wrench into White’s plans. After 7. Qd3 Nh5, the threat … Qh4+ is still in the air, Black’s pieces are well-developed, White’s are stepping on each other’s toes, and Black still has his extra pawn.

The funny thing is that Ilan never figured this out. I’m sure he didn’t use a chess program. Just about the time Fritz showed me how good 6. … Bg4 is, Ilan quite playing the 3. … Nc6 line and so I never had to worry about it again.

Now, apparently, I do. Of course, after seeing that Shaw considers it a “refutation,” I had to see what Johansson said. And his chapter on this is sketchy at best. He definitely considers 5. e5 to be no longer acceptable, and so he looks at two other ideas. He says that 5. Bd5 is “intriguing and fun to analyze. White might not have 100% compensation for the pawn, but it’s probably close enough for a decent surprise weapon.” Oh-oh. From a King’s Gambit partisan, this sounds like rather faint praise. The main line, he thinks, is 5. Nc3, and here he is a little happier about White’s chances. It’s still not hunky-dory, though, as he says things like this: “White’s pawn structure may not be the healthiest, but on the other hand White still has more influence in the center and a half-open g-file.” The big problem, though, is that there are no examples from master play. It’s all analysis.

Nine years later, Shaw has presumably gone over Johansson’s analysis with a fine-toothed comb and found problems. I think that line with the messed-up pawns was an important one. I still don’t know if there are any examples from practical play — for that I’d have to buy the book.

But let’s suppose he’s right. Suppose that 1. e4 e5 2. f4 ef 3. Bc4 Nc6 4. d4 Nf6 5. Nc3 Bb4 6. Ne2 is just good for Black. What then?

Well, first of all, here’s a puzzling point. White can also play 4. Nf3, offering a transposition back to the King’s Knight’s Gambit. Johansson is not keen on that because he has given up on the King’s Knight’s Gambit. But Shaw’s book, on the other hand, seems to be much more positive about it. So what does Shaw have to say about this simple transposition? I’m not sure! I handed the book back to Richard before I thought of that question.

But let’s look at a worst-case scenario. Suppose that even transposing to the King’s Knight’s Gambit is no good. Suppose I can’t play the King’s Gambit any more. What am I to do?

Actually it’s not a bad question to ask, because I ought to have a lower-risk backup plan anyway. Currently, my backup plan is simply to play 1. d4 in case I don’t feel like having a life-or-death battle starting on move two. I’m moving more and more toward d-pawn openings anyway, because I seem to have better results with them.

But staying in the e-pawn spectrum, I think that one thing I would want to do is check out Tiger Lilov’s current series of lectures at ChessLecture.com on a “Four Knights Repertoire.” The Four Knights Opening (1. e4 e5 2. Nf3 Nc6 3. Nc3 Nf6) is traditionally seen as boring, but it can’t be any more boring than the current fad of playing the Ruy Lopez with d3 for White, which seems to me like trying to kill your opponent with a marshmallow. The other thing I’d look at, if I were forced to give up the King’s Gambit but still wanted something sharp, would be the Evans Gambit. (1. e4 e5 2. Nf3 Nc6 3. Bc4 Bc5 4. b4.)

P.S.: As I mentioned before, I am about to go to Germany to attend the Heidelberg Laureate Forum. Most of the readers who responded to my post seemed okay with the idea of having some math posts here. You can always just choose not to read them. So for about the next week and half or so, this blog will turn into “dana blogs math”. If you’re not interested in math, come back in two weeks!


Print Friendly, PDF & Email

{ 13 comments… read them below or add one }

Ashish September 17, 2013 at 10:04 am

In the Four Knights, there’s always the Halloween Gambit [4. Nxe5!], at least for blitz games.


admin September 17, 2013 at 10:12 am

Yeah, I don’t quite believe in it for a tournament game. However, with colors reversed, if White plays 1. e4 e5 2. Nf3 Nc6 3. Nc3 Nf6 4. g3?!, then I absolutely would play 4. … Nxe4, because White’s 4th move took away his knight’s best flight square and weakened the light squares.


Matt September 17, 2013 at 12:34 pm

The Halloween Gambit is good for blitz but I would never play it in a tournament game. I think it’s just close to losing for white if black knows what he is doing.


ben March 4, 2014 at 11:28 am

I analyses it a few times with rybka on varying depths, white won a few of those a few draws, but no loss


weng siow September 17, 2013 at 2:51 pm

I am looking forward to your postings from Heidelberg. I have just noticed that blogging has just started from Heidelberg.


Ashish September 17, 2013 at 8:28 pm

And for those who are not averse to using the fruits of plagiarized software, there’s of course:


admin September 19, 2013 at 3:32 pm

I did not catch on to this, but Wikipedia (http://en.wikipedia.org/wiki/Vasik_Rajlich) says that this article was an April Fool’s joke. And indeed, the article says (if you work out the dates) that the interview took place on April 1.

Of course it should have been obvious. The idea that there is something special about the King’s Gambit that makes it easy for the computer to solve is rubbish (or a joke). If a computer could solve the position 1. e4 e5 2. f4, it would I’m sure be able to solve the position 1. e4 within a year’s time. The more fundamental point is that no amount increase in computer speed or use of clusters of computers will get you from looking ahead 14 ply (or 16 ply, or wherever computers are today) to looking ahead 60 or 100 ply (which is what it would take to solve the King’s Gambit). I don’t care how aggressively you prune the tree. The exponentials are too exponential.

And yet the article still fooled me until I looked at Wikipedia. Shame on me!


Phille September 22, 2013 at 12:55 am

I faced the same choice one or two years ago, when I got scared with the KG’s bad reputation and my own lack of theoretical knowledge. First I looked into the Bishop’s Game, which is just a cop-out, really. Then I prepared the 3-Knights Bd3 variation and the Moeller Gambit. I won two games with the Moeller, but finally convinced myself that it is crap.
I also looked into some offbeat Ruy Lopez lines. Like the Worrall Attack and some early d4 stuff. And began to prepare the Evan’s Gambit to replace the Moeller.
But now that the Shaw’s King’s Gambit is finally out, I might just return to 2.f4 …


tom bombadil September 25, 2013 at 4:34 am

Hello everybody,

I’ve been experimenting with 1.e4 c5 2.b4
White’s development on Q side is very fast. Has anybody here tried it?


Michael hancock January 29, 2014 at 9:00 am

First of all I truly believe shaw is wrong after buying the book. It very much seems to be equal in some positions and he just stretched it out to be better for black . I E Tomas J,’s line in fascinating kings gambit seems to with compensation BUT nothing more . My comp agrees with me and the few games from master practice from the ending position correlate with that assessment. ( in regards to john shaw’s opinion on the knight e2 line in the duras defence) and To be honest I wouldn’t worry about it. Theres positions that are spouse to good for white like the line in the petrov 1.e4, e5 /2 nf3 nf6 3 knightx p ,queen e7 ?! Look it up in the database a guy Afromeev v. Just kills it with this line for black lots of draws. In fact the canal attack in the sicilian 1e4,c5, 2 knight f3,d6.3 bishop b5+ bishop d7 4BXB Q X BISHOP. Is probably just equal ( shirov thinks so and rublevsky the main exponent stopped playing it because he thinks so) and look up the petrov and marshall gambit in the database because sadly it looks like they just equalize against e4 and the its main moves I E the ruy lopez . I just went through the chessbase database and theres like 1000 draws in the petrov and seldom any wins or losses compare, and the end of the day , playing from equal positions is fine because were human, we outplay each other and that with perfect play its draw stuff will be left for the comps ! ps great blog first time here and to be honest i was outraged by shaws claim! he and the people who don’t like the kings indian can have a were better party before we crush them with are “ramshackle” openings


Sherman May 31, 2014 at 1:40 pm

Well, fancy that! I was directed to Shaw’s book yesterday, and being myself a “reformed 3.Nf3” chessplayer, the chapter pretentiously called “Refutation of 3.Bc4” immediately drew my attention. Now I must say I’m very disappointed with your post’s un-exciting finish – fancy having to stop playing 2.f4 altogether, or even switching to 1.d4! I think we must take the braver approach and tackle 3…Nc6 head on.
I have spent part of my day checking a few lines – first by myself, then, alas, unavoidably with the sillicone brain’s aid, and I would cheer you up that the prospects for the first player aren’t nearly as dire as our dear Mr. Shaw would have us believe. There’s still hope, and very well-grounded hope, if you ask me. Well, since I’ve just noticed you’ve written your post back in September last year, I’m wondering if you would care to share your conclusions on this line, which you certainly must have ripened by now. I’d be happy to read any fresh insight you might have, and then maybe we might join forces on taking the battle to 3…Nc6.


Marc Z December 18, 2015 at 9:25 am


I’m not a native English speaker, and there is an expression I’m not sure I understand correctly, that is: “After 6. Bb3 etc. White gets to sacrifice the exchange”. What I don’t understand is the exact meaning of “get to” in the context of this article. Could you please help?



admin December 18, 2015 at 10:24 am

Here “get to” means “have the opportunity to.” My opinion is that exchange sacrifices in the King’s Gambit are usually quite pleasant for White.


Leave a Comment

{ 1 trackback }

Previous post:

Next post: