Why is 4...b5 an Error in Ruy Lopez?

it isn't a terrible move for Black.  it has been played hundreds of thousands of times...it can either tranpose into the mainlines or can be followed up by ...Bb7, ...Bc5

How come do you choose your openings based on percentages?? If I am right, this variation of the Spanish game has been played many times by Morozevich. So, if a 2700+ player chooses this line, why should you ever worry about some useless statistics? In this line, Black goes for White's light-square bishop while neglecting his development. Well, given how powerful this bishop is in the Spanish, it might be worth the price and certainly is a playable line!

It is a playable line, but there are certainly problems with 4...b5 so soon. It gets the bishop to b3 right away before black is close to castling and so sometimes Ng5 can be strong. Otherwise (I'm not sure myself if black tries to transpose whether it actually will or not) it would transpose. 5...Na5 on the other hand, isn't so great. GM's have played it sometimes, but according to books I've read they weren't having alot of fun with it because it's not worth opening up the queenside and tempi to hunt down a bishop that is often closed in on c2 (at least for awhile). My guess is that white can get some early play in the center with a quick d4 (the knight on c6, a big controller of d4, won't exist) when black's queenside pawns are weak (this is not nearly as bad if the center is closed like it is most of the time) in addition to having a centralized game with more space there.

1 a3 is andersson's opening. He was a great player, but it's not a great opening. And I always thought this was called norwegian opening anyway, unless Taimanov has more than one variaton named after him (besides in the sicilian).

After black plays Bc5 and/or Bb7 the game will transpose to the Arkangelski variation which is a respectable line for black.

I found it listed in Game Explorer as Ruy Lopez, Morphy Defense, Caro Variation:

http://www.chess.com/opening/eco/C70_Spanish_Game_Morphy_Defense_Caro_Variation

http://www.chess.com/explorer/index.html?id=3391&ply=8&black=0

However, all the following lines (e.g.: 5...Na5[Norwegian Variation], 5...Bc5[Graz Variation], 5...Nf6, 5...Bb7, 5...d6, 5...Be7) are playable by black - even Ng5 does not create any problem as in the Two Knight's Defense.

A fun line to try for White:

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 b5 5. Bb3 Na5 6. B:f7 + (!?)

This is not completely sound but, with his only developed piece being the Knight on a5, and the e5 pawn hanging, the Black might find it difficult to defend in blitz at least.

Don't rely so heavily on DB percentages... this is the WRONG way to use a DB. There are so many things that can influence that number that it really should not be relied on.

maybe because he moved the bishop twice & he gonna move it for third 3th, his OPP. opening lines for his game & he took the center but Ruy he is non

(try as much as u can to move the piece not more once)

It's not.  Try following it with a quick Bb7.

Check out Andrew Greet's Play the Ruy Lopez where he states its the Norwegian opening and not a bad move. The black player sacrifices time but gains the powerful lopez bishop from white.

"Useless statistics" like....winning?

Drastic shifts in DB percentages are going to tend to indicate that an early move in an opening (like this) creates some kind of weakness. It's clearly not unplayable for Black (since people still win) but that when strong players encounter it that they found something to do well. Black chose this variation, but White likes it. 

My question came from looking at the game and not seeing what good things White found. 

Also, from playing my game I found that one major problem Black creates for himself is that 4...b5 forces the bishop back to 5.Bb3. The white bishop is now threatening Black's weak f-pawn from a position that's hard to attack. It's clearly not impossible to defend against, but it's tricky. 

I dont understand why it is considered worse by database as well. Look at this:

It just transpose to mainline, either archangel or close spanish, white can try to attack f7 right away, but after d5 black has good version of 2knight defence, so it is probably not a good idea.

I think this can be valid move order for black cause it avoids delayed exchange variation without any disadvantages as far as I can see, If white has additional option, let sb explain it for us:)

3...a6 4.Ba4 b5 5.Bb3 Na5 IS PERFECTLY PLAYABLE; besides Taimanov there have over a dozen other Grandmasters who've played it:

1. Furman
2. Barcza
3. Fischer (yes, THAT Fixcher)
4. Robatsch
5. Evans
6. Donner
7. Kupreichik
8. Agdestein
9. Timman
10. Short
11. Morozevich
12. Mamedyarov
13. Vallejo Pons
14. Alekseev
15. Moiseenko
16. Firouzja

Not to mention a whole raft of IM's

I remember being very puzzled by this when I first started studying chess in my early teens. None of the books recommended it but none would say why (except for one that gave the speculative Bxf7+ sac mentioned above). Later, I found out that, as others have said, it's perfectly playable.

I think it's a fine move, but in terms of flexibility, it's not "optimal". For one thing, you fix the pawn structure right away.

Generally black would do better to play the pawn push later and immediately develop one of his minor pieces instead.

i love men