I have been looking at several books on how to assess positions. While trying to absorb the information I am also transcribing old games. Naturally I filter the games through the (very open) mesh I am slowly constructing for use in the assessment. The following no-komi game threw up some surprises for me, though.

Take the position after move 25 (triangled).




The commentary criticised Black's move. He had mistakenly assumed White would answer it. He ought to have played A instead. That is perhaps not the exact move I would have chosen, but playing in that area does not faze me. White actually did not grab the opportunity to play in this area himself but switched to B. No surprises for me there, either: it's hard for White to find a good move in the lower left so mowing the lawn while putting off fixing the tile on the roof is a strategy I can fully emphasise with!

But what I think we can draw from these options and this slack move by Black is that this position can be regarded as quiescent enough to be "countable."

I think many amateurs would count the top area as a Black territory of about 25 points. Those at the stronger end of the scale would be chary about making it that high, or definite, and some pros would say the area is not countable at all. The area in fact is what the Japanese call a jimoyo. This word is made up of ji = territory and moyo = moyo/framework, and it has almost always been translated as either 'territorial framework' or just 'framework'. I've done that myself, but have always felt a bit uncomfortable about it. Moyo actually has two uses. One is the familiar one of 'framework' (more accurately 'pattern' or 'design'). The other is as a bound morpheme meaning 'appearance of ~'. That is really the meaning in jimoyo, and so a better translation would be something like 'quasi-territory', or perhaps Matthew Macfadyen's 'virtual territory'.

Wherever you are on the spectrum, though, it is safe to assume that Black has rather good propsects in this area. If we say, therefore, that he has potential for around 25 points, we can probably keep everyone in the same tent.

Now as you already know White pissed in that tent. What happened next led to the next diagram (last move again triangled).



I am still feeling no sense of surprise. White's invasion has robbed Black of some potential territory, perhaps of the order of 10+ points. However, White has created for himself a weak group, so far eyeless but about to get an eye easily so we count that as -10, following the QARTS principle.

There has thus been no loss or gain for either side. It's a sort of commutative process. Of course the nature of Black's play changes. Rather than surrounding a large territory at the top he has to convert the potential lost there by gaining it elsewhere, that is by attacking White's weak group and so making -10 a reality. The first step in attacking is separation and so Black's last move is no surprise. It will be no surprise either that White next formally converts his two stones into at least a one-eyed -10 group (without an eye it would count as -20). That happened as follows.



This is where it began to surprise me.

What I expect to see happening now is a process whereby Black attacks the upper White group and in the process builds strength and causes collateral damage to White. In this case, I'd expect him to be looking at moves inside the White area on the right side and also activating the aji of the Black stone in the upper left. What I actually saw was almost the opposite of that.



I can see a justification for Black's exchange on the left: his own upper group is not entirely safe and the aji in the upper left is perhaps too fanciful even for a pro. But the Black moves on the right puzzled me. He is strengthening White in an area where he looked decidedly thin (and making that slack move in the lower right look even worse in the process). I am not convinced that it's the sort of White area where we would normally talk about reduction, and so the only rationale for Black's play, it seems to me, is to continue attacking the upper group. But it looks defensible and Black does not so far seem to have genuine prospects of territorial gains in the centre - and he has himself killed off the possibility of collateral gains on the left and right.

Remember I was transcribing this game rather than studying it. My surprise at the way the game was progressing manifested itself in having enormous difficulty locating Black's next move (lousy printing, mass of moves on one diagram, many-stroked Chinese numerals). In all the obvious next-move places I looked I was clearly not entirely off-track because the moves I looked at were all played soon after - just not now!

In the end I had to resort to scanning the diagram line by line and stone by stone from top to bottom. Of course Murphy had been there before me and ensured the move was near the bottom of the diagram.

And that move was a double surprise. See below.



This surprised me for a couple of reasons. One was that the move recommended in this area before was A. The different location was not criticised in the commentary, and although I say I was surprised, the surprise was mild. I can easily accept that the situation has changed so that now Black sees a more specific way to attack the White group on the lower side - specifically in a splitting attack between that and the group above. I have also learnt enough to know that such an attack is not as airy-fairy as it seems to most of us amateurs - I have just not learnt the patience and long-sightedness to try such vague attacks myself.

The real surprise for me was in the commentary. It said that after the triangled move "the position is now favourable for Black."

My surprise was not that Black was ahead despite making a slack move in the lower right: White had earlier made a slack move himself (the square-marked stone), so they cancelled out. In fact, although I know I felt a strong feeling of surprise I am still not entirely sure why.

But here's a guess. Like many amateurs I had been hung upon on evaluating a position by using counting. Even though I knew enough to be chary about precision and I was still assuming Black only had a certain potential at the top, when that was changed into a different kind of potential I was expecting to see immediate, or at least early, gains in recompense for Black elsewhere. Instead Black took the scenic route. The nice thing is, though, that when I look at the last position and try to count it (a very rough count is what the pros recommend, but either way I come to the same conclusion), I can suddenly "see" why Back is ahead. I have achieved the instant "recognition" rather than slow recourse to general principles that is being discussed in another thread.

A large element of surprise can be instant recognition of something, and what I now recognise instantly is that known territorial prospects are roughly equal but Black has no real defensive worries at the moment, whereas White has two or even three weak groups. Ergo, Black has the advantage.

One thing I can add, though, showing that I still have a long way to go, is that while I can see that White is at a disadvantage through having unilaterally weak groups, I assess that disadvantage as being far less than pros seem to. They can see the longer-term playouts.

As you will see from the full record below, the three-pronged attack by Black dominated the rest of the game and he won without counting. It was almost an exemplary example of splitting attacks, especially the simple one-space jumps. It was marred only by an overplay on 89 (should have been C3), and his attack was so strong he didn't need to kill anything. The balance of (real) territories did not shift in his favour until move 181, but Black (Murashima Yoshinori) held such a string initiative that this eventual shift was never in doubt.