I don't understand how it's possible to have an opening repertoire...

I wonder if you are more confused on the math side or on the chess side - probably a 50%/50% I guess... 

If we cannot convince you, why don't you take a look at any large games database, take any opening repertoire of your choice, enter the first X moves and see how many times the same player shows up?

Evidence should convince you...

I can assure you I'm not lying. Now, some openings like the dutch, you see once in a while but not that often so you don't have to prepare for that opening as much but many main openings are played very often. I could try to find the flaws in your math but I know that's what happens at my level at least. You do realize that not every move is an opening, that you study, and that for example the first white move (most likely e4 or d4) draws a big line on what openings are possible. For example, with 1 d4 black could play 1...e6 but unless white plays 2 e4 he won't have to be in a french defense. Same goes for the sicilian and other e4 openings.

For evidence, I went here http://www.chessopeningsdatabase.com/11/e4,c5,Nf3,d6,d4,cxd4,Nxd4,Nf6,Nc3,a6,/Chess-Openings-Database.htm (the first Google result for chess opening databases) and went through the first five most popular moves. Out of the 3 million games in the database, the MOST POPULAR OPENING occurs in 0.6% of the games, or roughly 1 out of every 200 games. Every other sequence of opening moves occurs even less.

This means that the absolute best you could ever possibly hope for is for your favorite opening to be played 1 out of every 200 games. Not great odds, if you ask me.

Of course you're right. I've played many duplicate games over the years, especially lines that involve traps.

I must have had the smothered mate in the Budapest a dozen times over the years, Same for the Fritz Variation right up to mate.

I've followed a Mieses Chigorin OTB almost to the conclusion. (not because I'm as good as Mieses, but I knew the game :) We know things, therefore we play them.....

Remember that if you have an opening repertoire, you're not playing random moves. You're going try to direct the game towards your opening. So, say you want to play the Najdorf.

Using the Chess.com database:

e4 is played roughly 50% of the time.

Then you play c5 100% of the time.

Nf3 is then played roughly 75% of the time.

Then you play d6 100% of the time.

d4 is then played roughly 85% of the time.

You then play cxd4 100% of the time.

Nxd4 is then played roughly 97% of the time.

You then play Nf6 100% of the time.

Nc3 is then played roughly 98% of the time.

You then play a6 100% of the time.

.5 * 1.0 * .75 * 1.0 * .85 * 1.0 * .97 * 1.0 * .98 * 1.0 = ~.3

So you would end up in the Najdorf 30% of the time if your opponents played like the players in the database. Of course, you would have prepared something else for non-e5 (or tried to transpose those into the Najdorf) so you would rather say that 60% of the time, when your opponent plays e4, you can get into the Najdorf. Pretty good odds.

pbrocoum is amusing. Just doesn't get it, bless him. Nice post by Tenna.

Interesting.

Of course, there's this radical group of thinkers out there who  feel that leveraging off the collective wisdom of Master-level games that have been documented for over 300 years might be useful as well.  They might like a word with you.

ive played this five move sequence 1000's of times it seems 1.e4 e6 2.d4 d5 3.e5 c5 4.c3 qb6 5.nf3 nc6.  i  couldnt really estimate how many of my games "start" from thas position.  i cut my openings down to the french vs 1.e4 and QGD slav vs 1.d4 as white i always (90% of the time) start with 1.d4.  alot of my games as white or black i end up with pawns on d4(d5) c4(c5) and e3(e6) and knights on nf3(nf6) and nc3(nc6) thats my repertoire come get some if you dare. at this level we can "cut the fat" and stick with a few basic openings

I hate repeating myself, but you give me no choice.

YOU DON'T PLAY CHESS, DO YOU??

From the first year I played in tournaments (group b, below 1500 in national) I made myself a opening repertoar, thats was really easy! First wiht white:

-Hm.. e4 or d4? ok, I'll go for e4 since I like to attack.

Then you have to chose openings againts the most popular defences, the most popular is c5 and e5, but you'll face e6, c6 and some d5.

Then you pick some lines you like, I went for the gambits.

Smith morra against the sicilian, kings gambit against e5, french gambit, mainline in the rest.

As a beginner you don't need to know much, just the mainlines, you'll learn much from your games!

So after some weeks with a book called "winning with the smith morra" I won many games against the Sicilan. I still know those lines by heart, that was in 1994.

Don't forget that you learn openings with both colours, you use what you learn from your white games, when you play on the other side.

And if you don't want to learn many openings, you could just chose a system, KIA is popular, c6-d5 and c3-d4 is allso playable in all your games!

100 games, and all wiht the same system, that's 100% with your favorite opening!

-Actually, most chessplayers from time to time, looks for other openings than the one they uses all the time! because they wants to learn something new!

If the chance for using your favorite was so low (0,6%) no one would bother to read opening books. And Bobby would not have bothered to go crazy, and invent chess 960!

So, play more chess, and write less stupid topics. I'll try it myself  (currently 415 stupid comments and 583 finnishd games)

this is why your rating is 1000

You have a small opening repertoire, by definition.  Not none.

I think we might have this all backwards.

The point of an opening repertoire is too limit the amount of lines you have to study by organizing your responses against your opponents responses. 

It is redundant to know theory on all twenty possible starting moves because you will not play all twenty, probably just two.  The same can be said about black's responses.  You should only need to know a plan against all the major responses.  If you meet another move, you can be sure of a good position by following opening principles.

If you want a really simple repertoire, you could just play 1.e3 as white and 1...e6 against any white response and you should be solid.

Did your equation factor in that sometimes they transpose into there favorite lines?

From my personal experience i find your maths inaccurate to say the least. I have played plenty of tournament chess and i found without a doubt the stronger plays ALWAYS had favorites. I know i do, even though i try to be more random here for the most part. I still have the major lines of even the other openings i have gone through.

By the way can you explain why we have so many opening books which mention certain specific lines if the chances are so bad?

Well, I am a stronger player and I never play favorite openings.  In fact, all of my openings are uncommon or my responses to them.

I thought everyone played the Slav.

Or does it just seem like that to me

The opening in chess is a curious phenomenon. If playing White, Black's immediate response largely determines how White will continue. If playing Black, your first move in response to White's opening will largely determine the future course the opening takes. It's like a mutually dependent relationship, mainly with threat and counter-threat deciding the moves (whether it be immediate tactical threats or traps, or strategic threats where the opponent threatens to get the best position).

I've seen the following position illustrated in at least 2 chess books:

This is claimed to be a more-or-less ideal opening setup for White. I've actually achieved this position in a couple of my games. (IMO the position is even better when White's rooks are centralized - anyone agree?). But of course, very few opponents are going to let you achieve this, and certainly none who know what they're doing.

As a novice, I can say that IMO it's much more important at our level to follow sound opening principles rather than specific openings in great depth.

When starting out, I thought knowing some specific openings in detail was the quickest way to get ahead. I spent a fortune on ChessBase DVDs and opening monographs on openings better players said were good (but good for them, not really for me!). That money was 99% wasted. Most of these resources just confused me and did not improve my play or results.

Much later on, I was advised to play a few specific openings which are mainly concerned with simple but sound opening play. Some principles of sound opening play are:

- develop your pieces quickly

- try to fight for a bigger share of the centre

- move only 1 or 2 pawns in the opening (focus on mobilizing pieces) and preferably your two centre pawns

- knight first then bishop (the best squares for Knights are almost always f3/c3 and f6/c6; Chernev said that a Knight at f3/f6 is the best defender of the castled King!)

- don't block your bishops with pawns

- put your bishops on active posts where they command as many squares as possible ("the best attacking piece is the king's bishop" - Tarrasch)

- castle early and connect your rooks

- put the Queen on a safe square off the bank rank

- centralize your rooks as soon as you have finished the other development and it is safe to do so.

- try to combine a developing move with a threat

- don't make a full assault on your opponent's position until you have first developed all your pieces (but if you only lead in development, it can sometimes pay to attack early)

Of course, you may need to deviate from a few of these principles to react to your opponent's threats, but they are a good guideline.

The openings that were recommended are:

White: Giuoco Piano, Four Knights' Game, Scotch Game

Black: Caro-Kann Defense, Two Knights' Game, Queen's Gambit Declined

These openings are simple to learn and are great for the beginner. For both sides, these openings are concerned with rapid development of the pieces to good squares, early castling and the fight for the centre. (Only the QGD is not as good for developing your kingside quickly). You can't ask much more from an opening in my view.

These openings give me 3 ways to play 1.e4 (or alternatively, 1.Nf3, as a different way of starting the Four Knights'), as my first move for White, which is arguably the best first move for White (at least if you like tactical games, and those are considered better for novices). For Black, they give me a solid response to 1.e4 and 1.d4, which are the most likely moves White will play.

I didn't learn these openings deeply, just the first few moves. That's usually enough to offer me an at least respectable position. After the first few "book" moves, I try to follow sound opening principles to complete my development, while reacting to my opponent's threats. In many games your opponent will "take you out of book" after a few moves anyway, so it seems pointless, at least at this stage, to learn specific lines many moves deep, not to mention the many variations that often accompany them.

Learn the sound principles of opening play and then focus on tactics and endgames. You will be spending your time most profitably that way.

An excellent book for teaching good opening principles is Chernev's Logical Chess: Move by Move.

Uh, for a 1 e4 player, it would be substantially higher than 1 out of 200. Your statement only makes sense if the guy could play any first move.

Great discussion topic.  I am just coming to terms with the enormity of the task in improving my chess.  I once thought learning the openings would lead to boring chess.  Losing to better players in the open only to regain material in the mid and end games is whats boring!  I play some very flawed openings because I like to play them and beat most players up to 1300 with them.  I know the moves, like the positional aspects and have some very fun games.  So my chess life is good! 

It was a hard lesson to learn that there are learned defenses and offensess within the openings but with a bit of practice and a bit more reading you can gain some very good mid game chess.

The Art of Attack in Chess for me is like chess 101 but would apprecaite any suggestions on a starting point in opening theory!!

This is possibly an odd case of anecdotes trumping mathematical calculation. However, the simplistic calculation you used is rather naive and does not lead to an accurate result. Tenna's calculation is probably the most convincing argument I've seen in the thread. The fact that you are actively attempting to force your opponents into certain lines greatly decreases the number of permutations as every one of your moves can "eliminated" from the calculation. In addition, as you can plainly see from the databases and the figures Tenna gave, the idea of there being about 3 reasonable moves per half-turn is not really accurate. Sometimes the moves are more or less forced (e.g. using the example of the Najdorf, Black is almost compelled to play 3. ...cxd4 to prevent 4.dxc5 opening up the d-file), and in other cases certain moves are much more popular than others, even though other theoretically sound moves are possible (e.g. again in the Najdorf, 4.Nxd4 is overwhelmingly popular at a 97% rate, despite the fact that the Chekhover Variation 4.Qxd4?! is also possible - although it is admittedly probably not the best move, it doesn't confer too much advantage to Black either). The "effective number of likely moves" is a lot lower than 3 per half-turn.

In fact, even excluding Tenna's allowance of trying to transpose from other lines, I'd say that being able to play an opening 30% of the time is certainly worth your time studying it. So that's how you have an opening repertoire.

If you need hard evidence of multiple games being played with the same opening, ask for the game records of any active live chess player or correspondence chess players with many games going on simultaneously. You'll see the same openings repeated dozens of times in their records. Speaking for myself, I play the Queen's Gambit as white on live chess and off the top of my head, I have seen about a dozen accept the gambit (out of which about half a dozen were newbies who fell for the 3.e3 b4 trap), half a dozen play the QGD, and half a dozen responded with the Slave Defence. Also I recall there were three who played the Albin Countergambit and five who responded with the Marshall Defence (although I remember those clearly because they're supposed to be uncommon).