Home » Forum Home » Analysis

Topic: The status of K9
Replies: 2   Views: 28,152   Pages: 1   Last Post: Aug 25, 2019, 7:21 PM by: karlw

Search Forum

Back to Topic List Topics: [ Previous | Next ]
Replies: 2   Views: 28,152   Pages: 1  
karlw

Posts: 968
Registered: Mar 7, 2006
From: Eugene, Oregon
Age: 32
The status of K9
Posted: Aug 17, 2019, 10:54 PM

As a self-described leading authority on K9, this is how I currently believe White's responses to K9 perform:

Almost certainly win: N10, O10

Probably win: K7, N9, N11

Unclear: K6, N8, O8, L7

Probably lose: the rest

Would be interested to hear other K9 authorities' opinions, and similar write-ups done for other common Black 1st (and maybe 2nd) moves.


tin_can

Posts: 26
Registered: Jun 1, 2019
From: Richardson, TX
Re: The status of K9
Posted: Aug 25, 2019, 2:15 AM

Please pardon my ignorance, this is my first time into this type of discussion.

Looking at the database, your two options of certain win are quite popular. But sorting based on percentage win there are some clever, but maybe hard to wield options.

1	 K6	8,221	67.6%
2 M6 1,351 67.6%
3 O8 1,522 61.9%
4 L7 2,633 60.2%
5 O10 23,437 59.1%
6 N8 1,803 58.5%
7 N10 43,162 57%
8 O12 1,375 55.5%
9 K7 5,885 53.7%
10 N11 2,275 52.9%
11 N9 2,766 52.6%
12 L13 1,495 51.5%
13 K14 8,724 51.1%


As a novice, N10 (for the V) and O10 (for the L) are appealing to me for the ease of use, so I can see their popularity. But is this the "wisdom of crowds" honing in on the best approach or is this group think and have found a "local min/max" to the game?

K6 looks like a very interesting exploration from this angle.

Edit: K6 is another potential for a V, but maybe it can be used to reign in black?

karlw

Posts: 968
Registered: Mar 7, 2006
From: Eugene, Oregon
Age: 32
Re: The status of K9
Posted: Aug 25, 2019, 7:21 PM

Two quick tips on using the database:

* Going one move further on an opening line can completely change your opinion of a move. Often times a move has the highest win rate not because it is strongest, but because the best response is non-obvious. 2. K6 is a good example of this, with 2...L11 the non-obvious strong reply.

* It's usually a good idea to use the ratings filter (at the bottom right). I typically use P1 and P2 ratings above 1800 only, sometimes higher. Setting it to >1800 gives you the following:

1 M6 421 77%
2 N10 3,592 64%
3 K7 492 58.9%
4 O10 1,734 57.3%
5 K6 640 57.2%
6 N9 171 55%
7 L7 228 53.9%
8 N8 130 52.3%
9 O8 112 51.8%
10 O12 59 50.8%
11 N12 58 50%
12 L13 42 50%
13 N11 120 47.5%
14 N13 70 47.1%
15 K13 96 46.9%

Now unfortunately the M6 data is quite biased by over 200 games of xtraclassy vs. data22 (this can happen frequently, as the sample size of top pente players is pretty small). It's a fine move, but I believe it is quite strongly contested by H11 (70.4% in the db with my filter). Excluding that, N10 is the clear winner with thousands of games played and the highest % overall.

K7 is the second strongest, but I'm working on some lines right now that give it a good challenge. After that it's O10, and then K6. K6 looks pretty good, but as I said 2...L11 gives it a run for the money.

Finally, we get to rarer moves like N9, L7, N8 and O8. I generally have success against these off-axis moves, but I think whether they win or lose is still pretty unclear.

Replies: 2   Views: 28,152   Pages: 1  
Back to Topic List
Topics: [ Previous | Next ]


Powered by Jive Software