The best first move for White

Hi fellow chess'ers, 

I've been looking at opening statistics to find out which move is the best first move for White. And no, it's not 1. e4. I've written a little article summing up my findings. 

Here it is: The best first move for White. 

Enjoy! And let me know what you think. 

I play c4 and I also play the sicilian.

Nice to know the statistics back my choices.

Nice. And what do you play against 1. d4?

Either 1...Nf6 2...g6 or 1...g6 2...Bg7

I also sometimes play 1...g6 2...Bg7 against 1. e4 as well.

As reply to 1. d4, both 1...Nf6 and 1...g6 are good (statistically). However, one reply is better than both of them. Guess which one. 

Meh, it's not going to be 1...d5.  If you don't know what to do against 1...d5 you aint going to be playing 1. d4.

Meh, 1...e6?

Good guess, Mr. Horwitz. But not quite. It was surprising to me. It's the Rat Defense, 1...d6. Most often followed by 2. e4 Nf6, 3. Nc3 g6. I will certainly toy with this from now on.

Pretty much all part of the same 1...g6 1...Nf6 family.

Mr Horwitz?

Kind of, yes.

Ah, Mr. Horwitz was just a joke. It's the name of the opening, 1. d4 e6. 

Well the most major problem about looking at this statistically is that the difference in variables are huge. You have to consider even among masters there are large gaps in ability making it so the results of many of the games insignificant to the opening move. Besides that the best openings should depend on the player play style anyways. 

Surely, these "bare numbers" doesn't tell it all. However, seeing that e.g. 1. f4's success rate is 0.87 in more than 4,000 games does tell something. A success rate below 1 for White means that Black wins in the majority of the games. Why would anyone spend time studying an opening unlikely to win? Insisting on playing 1. f4 is kind of like only playing garbage hands in poker, folding all big pairs before the flop. This is simply not a good idea.

I do not think 1. f4's low success rate based on database numbers is the reason less time is dedicated in the opening. It's probably because the studies of the opening give dubious results for 1. f4. 

It's funny how you can conclude playing 1. f4 as being almost as garbage, yet at the same time treat these results as if it were almost guaranteed. In other words you just said there are 4,300 games with masters who have played an inaccurate move, making the source itself inaccurate. 

Anyways again, we are not computers so basing moves on these statistics really only tell you so much. Players are not only effected by play style but also the psychological effect of an opening. 

There are so many transpositions that the statistics are meaningless. You have to dig deeper and figure out what the statistics are really saying.  For example, in correspondence games played over the last few years the Dutch is statistically blacks best response to 1...d4. Nearly all of blacks success is due to one particular branch - the Leningrad Dutch.

The best opening is always the one you're most familiar with. If you enjoy playing it, even better. Statistics don't mean shit when you're hanging pieces, getting killed by tactics or losing endgames that are won or drawn.

I agree statistics have limited value. I have found that many players are not able to deal with the Dutch Leningrad.

If I always play garbage hands in poker, yes, then I'll do better playing these hands, than someone with no experience playing them. That doesn't mean a pair of pocket Aces isn't better than 2-7. It is. Likewise, in chess, if you play 1. f4 all the time, you'll do better playing 1. f4, than someone at your level, who try to play it for the first time. So, you are right, experience counts, but so does statistics. Saying that statistics mean nothing if you are hanging pieces, and blunders your game away, is stating the obvious. It's however not an argument for the idea that no opening is better than any other. 

Let me put it this way. Imagine Kasparov being forced to play 1. f4 in 1,000 games. And then being forced to play 1. c4 in 1,000 games - against the same opponents. Do you think he would score better with one opening than the other? I'm quite certain he would, and I'm quite certain it would be 1. f4 he would score the most with.

"This  early" - meaning at beginner level, or meaning early in the game (after few moves)? 

The idea that "statistics doesn't matter" is like saying that we might as well roll a dice to "choose" our moves". Obviously, this wouldn't work.

I don't know how to put this gently, but your understanding of statistics is mistaken. But honestly, I don't want to get into an argument about this. If you don't like my article, then just ignore it. ;)

Let me elaborate. Saying that statistics for the first move doesn't really matter, is essentially saying that any of the 20 opening moves are equally good. They are not. Chess is a game of gradually gaining a slight advantage, and every move counts - either in your favor, or the your opponents favor.

Well, well.

Any statistician can perform these calculations, even if they never played a single game of chess, and are 2,000 points from GM level. You are missing the point, amigo.

I'll try to put the question another way :

- Let's say after 1.d4 g6 2.e4 white scores above average (from white's point of view) and that after 1.d4 g6 2.c4, he scores below average. But overall, 1.d4 g6 gives black a better score than other black 1st move replies to 1.d4. Should you still consider 1...g6 to be the best answer to 1.d4 ?

- Besides, you have to take another element into account : the elo difference between white and black in those games - 1...g6 is a way for stronger players to increase their winning chances against weaker players. So what makes the result : the opening move or the ability of the player behind the black pieces ?