# The best first move for White

The most common first move for White is 1. e4. But is it the best? No, it's not. And this is not a matter of opinion, it's a statistical fact (using data in the Game Explorer here on Chess.com).

1. e4 is played in no less than 46 % of all master games in the database. But there are actually 8 moves that are more successful. Here is a list of these 8 moves, as well as 1. e4, ordered by success rate.

• 1. Na3 (4.00)
• 1. h4 (1.50)
• 1. a4 (1.50)
• 1. c4 (1.29)
• 1. d4 (1.27)
• 1. Nf3 (1.27)
• 1. g3 (1.25)
• 1. f3 (1.20)
• 1. e4 (1.19)

The number written after the move is the success rate, calculated like this: (White wins + ½ draws) / (Black wins + ½ draws).

So, what can we learn from this? That we should play 1. Na3 all the time? No, not really. But why not? The success rate for 1. Na3 is more than 3 times the rate for 1. e4, so it seems to be a much better opening move. Well, the thing about 1. Na3 is that it's only been played in 5 of the 1.3 million games in the database, so although it won for White in 80 % of all the games, it's not really statistically significant. The same goes for 1. h4, 1. a4 and 1. f3, played in just 20, 10 and 11 of the games in the database, respectively.

That leaves us with 5 moves worth considering: 1. c4, 1. d4, 1. Nf3, 1. g3 and 1. e4. Except for 1. e4 (King's Pawn), 1. d4 (Queen Pawn Opening) is the most common opening move, played in 35 % of the games in the database. And, as can be seen in the list above, it's a great opening move, with a success rate 7 % (1.27 / 1.19) higher than 1. e4. The least common of these 5 moves is 1. g3 (Hungarian Opening), which is only played in 1 % of the games in the database. It too is showing a higher success rate than 1. e4. In fact, all these moves scores better than 1. e4. The best being (according to the data, not opinion): 1. c4 (English Opening), scoring slightly better than 1. d4, 1. Nf3 (Zukertort Opening) and 1. g3.

Does this mean we should never play 1. e4? No, it doesn't. The reason why 1. e4 doesn't have the same success rate as the other four first moves mentioned is primarily due to the strong 1...c5 (Sicilian Defense) reply from Black. Most other replies to 1. e4, and White will be fine (more on that in a later article). So, by all means, do play 1. e4, just remember to discretely cross your fingers under the table while doing so, and hope Black will reply with 1. e5, or something even worse.

So, Fischer was wrong when he called 1. e4 "best by test". It really is 1. c4.

• 8 months ago

@Fred-Splott:

"I don't believe that numbers definitively describe how strong an opening is".

I think they do.

"I believe you are using subjective criteria."

If you use numbers, than that is objective, not subjective. But of course, conclusions can be wrong.

I think that, if both players play perfect, it will always be a draw.

If you say "I don't believe": that sounds more subjective to me than using numbers.

• 3 years ago

@fred

I was quoting Cigol. I haven't stated any opinion on the matter at hand. If I had to, I wouldn't say 1...e5 is any weaker than 1...c5/

• 3 years ago

hicetnunc

I don't believe that numbers definitively describe "how strong an opening is". An opening like the Sicilian can seem stronger than, say, ....e5 simply because it's fashionable and so people use it and specialise in it. I don't think the masters of old can have been that far wrong when they proposed that the two best, winning attempts for black against 1. e4 were ....c5 and ....e5 and that they are roughly similar in strength. I believe you are using subjective criteria.

• 4 years ago

Yes, 1. h4 and 1. a4 scored higher than all other moves, except 1. Na3. But these figures are not statistically significant, and does not tell anything about the merit of these moves.

• 4 years ago

a4 and h4 had a higher rate than others?

• 4 years ago

hicetnunc, I agree, that would be interesting. I don't have the data. If you do, I'll gladly perform the calculations.

• 4 years ago

Fred-Splott, you can't really argue with the numbers. The reply 1...e5 - after 1. e4 - has a success rate 11 % below 1...c5. So saying that it's "as strong as or slightly stronger" simply isn't true. An 11 % increased chance of winning, in a drawish game like chess, is significant. I don't see how anyone can disagree, the numbers speak for themselves.

What if you perform the same calculations on another sample :

- games where both players are rated 2700+

- year 2010 and 2011

Do you still get the same results ? (I don't know the answer, just curious)

• 4 years ago

"Forget about what is best, its a myth, just enjoy the games you play.

If that were true, then 3...a4 would be as good as any move here.

• 4 years ago

All these statistics are fine but do we really know what percentage of wins can definitely be put down to the opening. For example many of the losses documented after playing e4 may be due to middle game errors or endgame errors.

This means that we cannot talk about the best move in any way statistically or otherwise. All the statistics show is what moves are prefered but the players listed. As you correctly point out some moves are rarely played and one game out of two is fifty percent but so is five hundred games out of one thousand.

Forget about what is best, its a myth, just enjoy the games you play.

• 4 years ago

I play the English 99% of the time as white, and if I'm black and they play e4, I play the Sicilian. I like to convert the English over to a Botvinnik Formation. I typically have pretty good results, but that's not saying much at the amateur level.

• 4 years ago

Fred-Splott, you can't really argue with the numbers. The reply 1...e5 - after 1. e4 - has a success rate 11 % below 1...c5. So saying that it's "as strong as or slightly stronger" simply isn't true. An 11 % increased chance of winning, in a drawish game like chess, is significant. I don't see how anyone can disagree, the numbers speak for themselves.

• 4 years ago

Well, the most combative winning attempts by black against 1.e4 are generally considered to be e5 and c5 and they're playable according to taste, so the writer is not necessarily correct to categorise 1 .... e5 as weak for black. It's about as strong as c5 or slightly stronger. Other defences may increase the chances of a draw at the expense of a black win. Incidentally I've been playing e6 based Sicilians for years and for me they're the best ground on which to challenge 1 e4.

1 c4 might have a statistically significant best success rate for white but I think this is because even at the highest level, defences to 1 c4 aren't taken quite so seriously as those to 1 d4 which I think is objectively white's strongest move. Anyway, I almost always play d4. I spent a few years with c4 but found it rather drawish against players booked up on either symmetrical defences or the natural 1 .... e5.