I prefer 0-0 instead.

Surprising that black can gain a slight edge after several iffy-looking moves. Is this because the computer feels White has castled into danger?

I would guess so!

I find the computer assessment odd, and all.

I don't agree with the analysis and think that white would have the edge here.

This move looks essential.

Black has a real problem on the K-side, which, owing to the King's location, extends to the centre. Could 14...Kf8 come into consideration? Perhaps with a view eventually to castle 'longhand'.

Yes - as I mentioned in a comment earlier...

I don't agree with Fritz. Instead, I like your variation the best, Charles.

Maybe 15...Nxd4 also came in for consideration? E.g. 15...Nxd4 16.Bxd4 Bxd5 17.cxd5 0-0 18.Ba6 or (probably my pick) 18.h4.

I reckon 16.h4 ain't too bad, neither.

Quite a clash of cavaliers on the Q-side there!

Not easy, but I remembered that nuntar once played Hope Chess. He was hoping I'd retreat, which I did. When, in reality, I could've sacced a piece to win material, and it works here. Having said that, after very careful consideration, I would play 17.Nxe7, after which, white would, at least, win the exchange (R for N.)

I'm pretty sure I would have played 17.h4, here, with a very promising strategic attack.

I would've played this. But, my mind was narrowminded here, so I forgot that lesson with Ken. Your variations are a better reason to do this. As I overlooked that he could play 17...Nxe7. I saw the second variation, and I, absolutely, disagree with Fritz here.

17...Nxe7?? would clearly have been losing. But this and the Q-recapture both look playable.

I don't agree about this being even since black can castle. Instead, I agree with what Charles posted earlier which is that white has an edge.

Maybe Black might have considered letting go the c8-rook in the interests of time to organise his position: e.g. 18...Qc7!? 19.Nxc8 Rxcc8, and Black has a considerable presence on the Q-side: 20.Bf4 Be5 21.Bxe5 Qxe5 - and Black doesn't look so badly placed at all. White would have a solid structure, but the 'wrong' bishop, and Black would have the virtues of flexibility and handiness. Very hard to assess: call it 'unclear'.

19.f4 would at least stop ...Ne5, but at the price of giving Black a move to come up with alternative countermeasures against 20.Qd5. So White seems to be 'sacrificing' a minor exchange for time.

I wonder what would have been the consequences of 20.Qf4, incorporating the same idea of the threat against f7. That move also has half an eye on the rook at b8.

I looked at that move too, but decided that Qd5 was better.

It's a hard move to go past: attacks the queen, attacks the bishop, attacks the knight. Has Black's Fabian strategy worked? Has White over-reached?