h3 question

Sort:
tigerprowl9

This seems to be coming up a lot.  I will play h3 to prevent the pin Bg4, and then when I put it through an engine I see it wasn't the best move.

 

Can someone explain why h3 is bad and when it should be played instead because I see it listed in other positions.  a3 is also in question but it seems like it can be played more frequently than h3.

 

ewq85

In positions where it's important to not let your knight get pinned like your example the ruy lopez h3 is absolutely necessary in order to play d4. Positions that the pin is not Immediately relevant h3 is a loss of tempo in fact I often welcome the pin then play h3 and many players just take my knight and give me the bishop pair for nothing

leiph18

Loss of tempo is the easiest reason IMO. The diagrammed positions is not a bad example. When h3/a3 doesn't threaten a piece, it's more usually seen after other useful developing moves have been played... or if it's very important to take away b4/g4.

In any position notice after Bg4 h3 Bh5 (maintaining the pin) puts the bishop on a less active diagonal with less chances to get back to the center. It can even be useful to throw in g4 sometimes.

In that position, yes, you're usually aiming for the d4 break, so keeping the knight unpined makes sense, but (and this is useful to remember in any position) what if you can play h3 with tempo, then it's like a free move. If you follow up with g4 (at some point later, probably not right away) then in a sense you get the best of both worlds. You get to play d4, and he lost time putting the bishop on a worse square.

So play around with it in the engine anytime you come upon a position like this. And don't be too afraid to play the 2nd or 3rd choice when it's just a few tenths of a pawn difference from the best choice. It's (much) more important to get a middlegame position where you're comfortable with the ideas and understand how you should generate play.

Uhohspaghettio1

I think it's in The Right Way to Play Chess that the author advises to avoid prophylactic h3 if the opposing bishop has another good place to develop to, for example f5 or in this case b7. Everyone hates a nasty pin, but after Bg4 h3 black will have to choose whether to keep to either relinquish the pin be and be down a tempo, or keep up the pin and his bishop will lose some of its power on the light squares, and now the pawn movements seem to have weakened his position. White might stick his light-squared bishop on d5. 

I_Am_Second

http://www.uschess.org/content/view/7907/341/

Uhohspaghettio1

OP that's ridiculous if you're taking an engine evaluation over established opening theory. You're not going to prove opening theory to be a "mistake" with an engine, that's not how chess works.   

TitanCG

You're allowing 7...Na5 here. 

You usually only play h3 in the Ruy Lopez to prevent Bg4 after you've played c3 to give your bishop on b3 an escape square and to prepare d4. When you play c3 and d4 without the prophylactic move h3 then a move like ...Bg4 puts pressure on the d4 square and could force White to make a decision in the center earlier than they want to. Since White hasn't even attempted to play d4 here the move ...Bg4 doesn't really do much. 7...Na5 on the other hand just nabs the bishop pair and gets the knight off of c6 at the same time.

ThrillerFan

There is more to it than just loss of tempo.  a3, h3, a6, and h6 should only be played when absolutely necessary.

Examples include:

Main Line Closed Ruy - 9.h3 is necessary in order to not weaken d4.

Sicilian Najdorf - Before you can play ...e5, which weakens d5, you need to play ...a6 first because leaving both b5 and d5 available to White is fatal (i.e. N on b5, N on d5, with an unstoppable fork upcoming on c7, just as an example).

When playing those moves are not 100% necessary, they are not only a waste of time, but also severely weakening moves.  What most fail to understand is that playing h3 severely weakens f4.  Playing a3 severely weakens c4.  Playing a6 severely weakens c5.  Playing h6 severely weakens f5.

Here's why.  Let's say you play ...h6 as Black.  White plops a Knight on f5, protected by a pawn on e4.  That Knight can be very annoying.  While there may be the possibility of surrendering the Bishop Pair, trying to knock the knight out of there with the pawn move ...g6 can be difficult because it weakens the h6 pawn, and in some cases, that h6 pawn just hangs.  So if Black can't play ...g6, the Knight becomes a major thorn.  Had the h6-pawn been on h7, then ...g6 is a lot easier to play as it doesn't leave a pawn hanging.  You would need to watch out for the dark squares, but at least there is no pawn hanging.

leiph18

That's an interesting detail.

It seems it would also be true because, e.g. an a2, b3, c4 pawn chain has its base at a2. So after a3, the pawn chain with its head on c4 is weaker at every point.

I suppose this logic can be extended to all pawn islands when choosing to move the base (or potential base).