Endgame Tesuji4
Once one has seen , it looks so obvious that one probably would not want to call it a tesuji anymore. But it is surprising when one encounters it for the first time. Since the sequence Black a through White d will almost certainly be Black's privilege, this is about 4 points better than the result of the monkey jump.
Comparison
This move can sometimes be seen in amateur games. It is almost always wrong. It can sometimes have shape problems when White plays at a. But even for pure territory, it is a loss:
While the shape built by the marked stone may look akward, there is nothing White can do.
One tricky question is how early one wants to play this sequence.
Since the white move at is either sente, or creates the nerai of White a, Black's sequence is very big. Often, Black can play it as soon as he is sure that he does not want to enter at b instead.
Timm(5k): Is correct ? Shouldn't it be at the 3,3 point like below ?
tapir: at 33 means was sente as in " is either sente, or creates the nerai of White a". (The whole reason this is a tesuji series is that you can teach correct sequences vs. sequences that lose point w/o going in full detail. There is even another endgame tesuji page for the move at 33.)
Timm: Thanks tapir. Actually I thought that Black's possible answers were 3,3 or tenuki, and that was aji keshi. So is the proper answer to keep sente, and tenuki is wrong ? Or maybe both tenuki and are possible depending on the situation ?
Uberdude: Both are possible depending on the situation. But if you do answer in gote please do not disregard the pole connection at 4-3 as an alternative to the 3-3. The 3-3 is better for endgame, but the pole connection is thicker with respect to the outside and can sometimes come in useful in later fighting there.
Value
, is middling out, because both hane plays are gote. and are two sente exchanges. There are probably some other possibilities for comparison too. If we assume all this, then this move is 9 points in double sente.
N. B.: This is not a double sente. See http://senseis.xmp.net/?DoubleSenteIsRelative
--dnerra