Learning from My OTB Game – Need Help with Strategic Thinking

I played an OTB game today as black, and after analyzing it, I’m still unsure how I would have played differently in certain situations. I made a few bad moves, but I don’t fully understand these situations.

Here’s one position where I spent a lot of time thinking:

• I noticed that I was just one move away from castling, but my opponent had started pushing pawns on the kingside, so I wasn’t sure if I should castle yet.
• I needed to develop my light-squared bishop.
• My opponent was also one move away from castling but on the opposite side. I figured that if they castled long, I wouldn’t have enough time to do the same, meaning I would likely have to castle short—if at all.

Based on these observations, I decided to gain space on the queenside and signal my intention to launch a pawn storm if my opponent castled long. I played h5, but the engine considers it inaccurate, and I don’t understand why.

This is a similar situation with the same thought process, but here I had already developed my light-squared bishop and my rook. So my final plan was to start pushing on the queenside and prevent my opponent from castling long.  

However, after my opponent played Qg3—which the engine considers a mistake (though I don’t fully understand why)—I played b5, but the engine says this was a blunder.

Can someone help me understand if my thought process is correct, or if I’m missing any general concepts? I want to improve and learn as much as possible from every OTB game I play.

In your first diagram here... 
- black might be able to counterattack in the center with a move like c5. Admittedly his DSB is in an odd place, blocking the d file. But when you make a non-developing move on the flank here, like h5, a center attack is back on the table. Black also doesn't have to castle queenside, especially if he's attacking in the center.
- your LSB is exerting some control over f4 for now. As pawns get traded off diagonals can open up and that piece may become active that way.
- A move like Qc7 is or O-O is more non-committal. For now there's no scenario I can see where you castle queenside, right? So O-O or Qc7 will be useful regardless of what black does, but when you play h5... it's a move which could end up not being useful.

That's my take on it, I didn't look at an engine so take it with a grain of salt.

In your second diagram... you're trying to push your pawns on the kingside, but play is going to happen in the center before that. You don't have the advantage with white, or the time, to execute the plan you're trying to go for. You need to be more flexible in your planning. You have to think.... "do I have the time to pull this off?".

The knight needed to jump in and plug the hole on d4. You have no other way of ever opposing that DSB. And until you do, you can't castle kingside... only queenside. So when you move your queen away... well, your queen will no longer be capable of defending the queenside, she'll be over there while you slowly push pawns and your king is coming under attack.

I think you're rushing to push your pawns too quickly, you're not adapting your plan to the situation, and so you give black more opportunity to counter your plan. Try to be less committal.

Thanks for your input! However, I don't quite understand how castling short is safe in these positions. Most of his pawns are deep in my territory, and his bishops, knight, and queen are just a few moves away. I didn't feel comfortable castling short, but maybe I'm overestimating the threat.

solve this puzzle

the first picture and concerns are good reasons why NOBODY should be playing the Duras/Bahr gambit (not saying you did)