41

Grouping of initial moves of a chess game are well known, though I've never seen analysis of the worst possible opening move, any thoughts? If so, please provide more than just a move, but the exact reason why such a move would be the worst assuming such a move was against a GM.

0

15 Answers 15

32

1. f3 is almost certainly the single worst move. It irrevocably weakens white's king position without doing anything useful.

Other moves like 1. b3 and 1. g3 prepare to develop a bishop, so they're not so bad. 1. c3 doesn't do too much for white, but it also doesn't hurt him.

1. a4 and 1. h4 both ignore the center, but they do help develop the rooks, and there are positions where 1. a4/h4 can be useful (especially when the kings have castled on opposite sides of the board.

1
  • 1. a4 can be met with e5, the bishop blocking development of the rook. It gives black a lead in tempo and center control with nothing gained immediately. It also stops you from controlling b4 where knights and bishops sometimes like to go. It's true a4 is sometimes useful whereas f3 is rarely played and that f3 weakens the king's safety, blocks development of the knight, and has little benefit. It's hard to say which one is worse. I'd guess f3 just because it weakens king safety. Although engines aren't the best at openings, Stockfish gives f3 a -0.9 and a4 a -0.2.
    – user904963
    May 13, 2022 at 7:32
29

I agree with Andrew on 1. f3 (Barnes Opening). No first move can be decisively wrong, but some first moves are questionable. In Chess Opening Theory (Wikibooks):

Quite a rarely played passive starting move that gives up White's first move advantage. White's position is slightly weakened by the White King exposure to a check on the h4 - e1 diagonal. The kingside Knight's favourite developing square f3, is also not currently possible. In fact, this is arguably the worst possible first move by White.

No stats as 1. f3 occurs rarely among serious chess players.

Next on the list would probably be 1. h4 (Desprez Opening):

1. h4 is a rather useless move that does nothing to assist the important central squares. No top masters have ever used it in professional play, although some, most notably Hikaru Nakumura, play it during blitz. It could be said that White has made his position worse, as castling kingside is now less attractive.

This move is rarely seen among serious chess players.

0
15

The other moves cited, especially 1.f3, are bad, but I would rather plead the case of the ugly Grob opening, 1.g4?.

Pawns do not go back, so this double step weakening is much more committing than a one step pawn move. After, say 1...e5, White has not only lost time, he has also:

  • Self-destroyed the most natural shelter for his own king, reached by short castle.
  • created weaknesses for himself, as Pg4 can and will be targeted by Bc8 or Ph5.
  • got himself a disadvantage in the fight for the center.

If, in some kind of handicapped game, I could choose my opponent's first move in order to maximize my chances, I would rather force him to play 1.g4 than anything else.

2
  • 1
    You could say that at least 1.g4 enables one development move (Bg2) instead of preventing one like 1.f3 (Nf3), but yeah. That's just a small comfort that still doesn't make the move "good".
    – Annatar
    Jun 19, 2020 at 7:25
  • 2
    Although engines aren't the best at opening moves, Stockfish 14 scores 1. g4 as -1.7 and 1. f3 as -0.9. It's the lowest scoring first move.
    – user904963
    May 13, 2022 at 8:35
12

For an objective take on this question, one might take a look at the Chessok Opening Tree:

  1. 1.f3 -0.33
  2. 1.g4 -0.29
  3. 1.Nh3 -0.22
  4. 1.b4 -0.18
  5. 1.h4 -0.18

Given that white should be able to achieve a plus of 0.11 (according to the Tree) these are the only moves that not only lose the first move advantage but more than reverse it.
Of course these computer evaluations have to be taken with a grain of salt, but they do bear out the intuition of experienced players and provide some food for thought (especially to adherents of the Sokolski …).

4
  • I don't think that the way the leaf evaluaton parameter have been historically optimized would carry much of the expressable intuition intented by experienced players, they would have to understanding globlac optimization of some function basis (I am not even talking about NNue), the global mathematical components mutual interactions or multidimensional independence have never been studied or likely on the mind of niether de developper at the time and how they would be precenting their whole algorithm in chess interpreatable ways.
    – dbdb
    Nov 15, 2023 at 23:16
  • The communication would (and likely has) have been on isolated chess theory concept (via their words, themselves not really meant as measurable truths, but calibrating points of raw dichotomies, meant to attract the rest of the brain, its intuition, to firt the future experience in that "bracket".
    – dbdb
    Nov 15, 2023 at 23:26
  • The leaf evaluation function evolution has been frozen one dimension at a time since the first shots in early AB engine types of which SF is the epitome. With NNue, it was officially frozen even more leaving all improvements to the evolution of NNue and its docking back into the SF search and classical leaf evaluation function (that last part frozen, but still what NNue is being evolved to universally approximate given its reduces feature set at input of the full position information). So by objective, you were right to add the grain of salt. Developed above.
    – dbdb
    Nov 15, 2023 at 23:28
  • In the first comment I should insist on correcting an important typo. I meant "the way that the leaf evaluation function parameters".. parameter should have been plural, with emphasis. But the optimization process in fish-test, AFAIK, has always been one parameter at a time/improving version. Hence the importance of the plural erratum. NNue within its output corset definition of training, is using global optimisation using its basis function of NN, to mimic that output on many positions. But, it is assuming that the target has been objective. Looping back to that, in this last post.
    – dbdb
    Nov 16, 2023 at 0:10
11

The worst move is 1. resigns. It goes from equal to lost in one move, much worse than moves like 1. f3 which go from equal to slightly worse.

1
  • 2
    "1. Resigns" might mean you don't lose rating points, though, since zero moves were played on the board. So maybe that alone makes it better than some other choices, depending.
    – D M
    Feb 24, 2019 at 18:17
8

1. h4 has been deemed the worst move by John Watson (author of the Mastering the Chess Openings series). Other people are saying 1. h3 and 1. a3. That is not true. Michael Basman is a noted unorthodox player that opens with 1. a3/h3, etc. and is rated >2400. 1. a3/h3 come out to be useful in a variety of openings, but it's certainly not what I would start with.

1
  • 4
    Not only is 1.a3 perfectly playable, but 1.a3 e5 2.c4! is a serious try for the advantage, with the additional appeal of offering familiar paths to a White player who plays the Sicilian Defence as Black. Of course after 1. a3 g6 White is struggling to prove that he's done anything useful on move 1, but he certainly need not come out of the opening worse. Jan 19, 2013 at 2:10
7

1.Nh3 and 1.Na3 (Ammonia attack, Sodium attack) are REALLY bad. It breaks the "move one piece one time" rule, and 1.Nh3 blocks h3, a good move to prevent back-rank mate.

0
7

I agree 1. f3 is quite bad, but 1. Na3 and 1. Nh3 aren't pretty either, especially as they have to move again to avoid a doubled pawn after ...d5 or ...e5.

0
6

The idea of the opening moves is to allow an early development of pieces. Therefore the worst moves would be the ones that allow you the least opportunity to do so. So I would say 1. a3, 1. h3, and 1. f3 would be the worst. You not only lose the race in developing your pieces quickly (the only piece you can develop after these moves is the knight), but you also give up control on the central squares.

0
6

I would say 1. Na3. The reason is that it is the most useless move as far as building a setup around it. Even moves mentioned like 1. h4 and 1. f3 can be useful in some contexts - 1. f3 can be useful if you play d4 and Nc3 and plan e4. 1. h4 can be useful because you discourage your opponent from fianchettoing his bishop, and you can push it again for cramping effect. It is difficult, however, to imagine any setup where 1. Na3 is a helpful move, and furthermore after ...e5 you will have to move the knight or play b3 to avoid getting doubled a-pawns, which are probably the worst doubled pawns since the open b-file is rarely useful and they are not supporting anything or guarding any important squares.

2
  • 2
    This is sometimes called the Sodium Opening: The Durkin or Sodium (Sodium due to the algebraic notation Na3, Na being the chemical symbol for sodium) is rarely played. The move does little to enhance the White position, although some sources consider it to be better than its reputation. Black's best response is 1...d5, although 1...e5 is sometimes seen. White may continue with 2.b3, preparing to fianchetto with 3.Bb2, or with 2.c4. Alternatively, if black plays 1...e5, white may choose 2.Nc4 to put pressure on the e5 pawn.
    – Daniel
    May 30, 2012 at 16:59
  • Incidentally, there is actually a system against the Sicilian which involves early Na3 (1. e4 c5 2.Na3). Iirc this is proposed by Nigel Davies in one of his opening books
    – prusswan
    Oct 22, 2012 at 13:36
6

1. g4 ... 2. f3??

This is a horrendous opening (especially in combination with the second move) because it severely undermines the ability to castle on the king's side, very aggressively throws a pawn into action where it has no defenders and isn't helping any cause in particular.

g4 allows to fianchetto the light-square bishop, as well as hide the dark square bishop behind it later, if needed, which is an odd, but somewhat redeeming quality, but as soon as it's followed by 2. f3, the light-square bishop is entombed and white have ZERO center control or active development for whole 3 tempi, allowing black to gain a MASSIVE center presence and advantage. And after 1. ...d5 followed by 2. ...h5 the g4 square now belongs to the black and sets a great entry point to a king-side attack. If white tried to take on g5, black now has a semi-open file without ever moving the rook. Thanks for that!

Such a horrible game would typically go like this: 1. g4 d5 2. f3 e5 3. Bg2 h5 - and black is 2+ points in the lead without having taken a single piece yet, according to any decent engine. White would need a major miracle and a blunder from black to have any chance to recover from that.

Example: Hou Yifan throws game

0
6

I'm just trying to protect a move that's mentioned above as being perhaps one of the worst. Grandmaster Mednis years ago wrote a little opening theory book ("How to play good opening moves," Edmar Mednis, ISBN-13: 978-0679141099) in which he states that

1.g3

is one of White's perfect opening move choices, when followed by

2.Bg2.

He does mention some of White's awful first move choices, but doesn't rank their badness.

0
4

1.f3

PROS The only pros of the opening is some small center support.

CONS

  1. This opening weakens the king's safety. The f pawn is an important shelter of the e1-h4 diagonal and the g1-b7 diagonal so moving it makes the kingside pawn structure destroyed.
  2. This opening don't open any lines fo development, and it blocked the knight!
  3. Wasted tempo.

SUGGESTION FOR PLAYING I do not suggest it for anyone

1.h4 and 1.a4

PROS None

CONS

  1. Doesn't open any lines nor development.
  2. Weakens pawn structure.
  3. Wasted tempo.

SUGGESTION FOR PLAYING I do not suggest it for anyone

1.h3 and 1.a3

PROS Does not weaken anything.

CONS

  1. Wasted tempo.
  2. No development lines.

SUGGESTION FOR PLAYING For white players who want to play black.

1.Na3

PROS

  1. It can develop to c4 or c2 but that would waste some tempos.
  2. Support c pawn, but can't the c pawn move already?

CONS

  1. After 1...d5 and an exchange the pawn structure is lost.
  2. Wasted tempo.

SUGGESTION FOR PLAYING Only in blitz.

1.Nh3

PROS

  1. It can use another tempo to go to f4.
  2. I'm surprised at nobody mentioning it, but it protects f2!

CONS

  1. Wasted tempo
  2. After 1...e5 and exchange the pawn structure is lost... and the f2 protection is gone!

SUGGESTIONS FOR PLAYING Only in blitz.

1.c3

PROS Support d4

CONS

  1. No development and blocks knight
  2. Wasted tempo.

SUGGESTION FOR PLAYING For Caro Kann players.

1.g4

PROS Open lines for bishop development

CONS

  1. The g pawn is undefended.
  2. Weakens kingside.

SUGGESTION FOR PLAYING Only in blitz.

The third most bad is 1.Na3?! and 1.Nh3?! The second most bad is 1.h4?! and 1.a4?! The most bad is 1.f3?!

1
  • I like a pro and con structure when a question is not yet solved or formalized so that answers from above might be a temptation of objective looking answer. As long as the premises and meta label on the statements/arguments are explicit. We can progress. Reason for my voting up. Even If I can't concur or incur based on personal expertise in chess (I am a dedicated learner though, but not from above).
    – dbdb
    Nov 16, 2023 at 0:16
3

I would say 1. Nh3.

1.f3 is bad however after g3 white is only down a couple of tempos and has a somewhat reasonable position.

1.Nh3 allows black the potential to to wreck white's pawn structure and even threaten a very quick mate. Yes, white can avoid that with Ng4->Nf3 but you've wasted two tempos when you could have done the same thing with 1.Nf3

White is in a struggle for equality after 1.Nh3 and is one slip away from losing the game for the next several moves. Even if white does play perfectly, he has thrown away several tempi for no reason and allowed an equal game.

1
  • Would be interested to differentiate the arguments with the other answer on the other KQ side. What does the KQ asymmetry would change. Or is that implied already. I may not be able to read that yet. Maybe making that explicit would make for some discussion, so we know how much debate room is still left in such questions.
    – dbdb
    Nov 16, 2023 at 0:22
3

Just tell you one thing, 1.f3 is bad, it does not control the center actually, it stopped the knight, and it place the king in danger. Sometimes even followed by 2.Kf2, I think this is the hammerschlag variation, further weakening the king, this is to laugh at beginners, but no one have win me with it yet.

1
  • Assumed subjectivity, which is fine, as it is clear, or explicit. I would guess that player pair variables in experience might alter the answer. It might already be present in all answers. One could, as Carlsen did in one game I read about, in bullet, play uncommitting moves for long time and find something to target in the opponent development. One can blame bullet, but within that, one can still find chess reasonable arguments about done development weakness, and yet to be developed other side.
    – dbdb
    Nov 16, 2023 at 0:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.