Game I lost defending against early kingside attack.

I often have trouble dealing with pawn storms on my kingside like this. Any advice on what I could've done better? I felt like I was already losing at move 13.

This is a good topic to ask for strategic advice, because often the computer will show you a ?? at some random place, because some impossible line was holding still, but actually your step towards trouble was earlier.

I'll give my strategic pondering first and maybe later we can analyze concretely any position?..

So

Thanks for the tips and analysis. I'll try to explain my thinking behind some of these moves.

>4. Bc4
My goal was this move and most of my early-game was just to develop normally, castle and claim advantage because of my opponent's lack of development and uncastled king. I didn't see any clear weaknesses Black had (my knowledge of this opening ends at the idea of Nd5 kicking the queen back). Concretely I am taking more control of d5 (no Ne7 d5's) and maybe in the future my bishop eyeing f7 will become important. I also have an aversion to moves like d3 before I develop my bishop out of the pawn chain, but perhaps the bishop is better placed stopping a potential Bg4 pin?

>a3 b4, h3
I see what you mean on this one. My opponent's bishop is better on the diagonal I kicked it to and while b2 is nice square for my bishop I'm contributing to my opponent's future kingside attack (in fact that very bishop's pin is what I blundered checkmate to.)
I agree regarding h3, but my question is whether I'd have to end up playing h3 anyways after something like d6 threatening Bg4. Or whether I can claim Bg4 isn't even a threat if delay d3 so I can play Be2.

> O-O??
This one is a bit difficult to grapple with because I've always thought castling should be done as early as possible. But I definitely see where you're coming from: often my opponents will wait for me to castle so they can start attacking my king or so they can castle the opposite way then do so.

As for the general principle of castling as early as possible - you are above that level at this point. You have reached the level where dubious players keep their king in the center and attack premature castling with a cave man attack.

GM David Howell said he used to like to see how late he can delay castling to drive his opponents crazy and have them guessing where the King will go, if at all. Often when your opponent hasn't committed the safest thing is to not commit in turn.

Imagine the satisfaction of developing all of your pieces with tempo then as your opponent really commits to some pawn storm & piece placement on the Kingside- you suddenly surprise him with O-O-O!

I castle "early" only when I need my Rook in the center for a concrete reason, for example; my opponent's king is stuck in the center, or I have something I can pressure in the center or a file to take over and penetrate through.

That-- or if it's just theory, like in the Sicilian, Italian, or Petroff, you just castle early because it's what you do. Also both these openings can almost never lead to a closed center, so the Rooks see daylight immediately.

With time you will know when it's important to not "give your address", when it's OK but a bit dangerous, and to do it happily.

Castling is not a rule of thumb thing, but ideally you want to castle in a moment where it's both attacking and giving safety to your king at the same time. If it's just one or the other, it's not the most time efficient (e.g; I just wanna move my king out of the center cuz I don't know what else to do)

If your opponent is completely underdeveloped, thus breaking the basic principle of developing fast, by not castling fast, you will be confusing him and showing you have nothing to fear until he commits to something if he can even get a piece out. So flexibility is a weapon. And then of course, you will be the one launching pawn storms but good ones not cave man ones...

Now I did look with the engine and of course there's this super basic line against this specific opening, So even though I look like a moron after checking the refutation, I hope my analysis gave you other important concepts to consider... Every position is different ofc...

Same logic applies, but here we actually use are 2 piece lead in development immediately with a tactical shot against the Queen and developing our Queen. Imagine that this position is OK for White:

In both later games, White gets substantial leads in development and attacks and in the second game, you have the pin breaker in Bd2 and now the Knight runs to g5 to put pressure on the f7 pawn.

15. exd6 was a very careless move that loses immediately if black is alert. Black should play hxg3 and the attack crashes through. It's interesting that you did not mention it in the analysis.

Surviving that, white is probably winning unless you miss the Qxg3 trick which you did. After any move that stops it (like Kh2) white is up 2 pawns with a safer king and better pieces.

And here come the stronger players

This is great analysis..

Mine was more about general strategy regarding dubious openings and pawn storms... but OP, listen to this guy ^

(just my point was Black shouldn't have even had any type of initiative, regardless that White could have survived the attack and ended up better)