Test Your Go Strength (book)

[YVW]

Yesterday my friend brought me a problem book called "Test Your Go Strength" which includes 50 problems, divided to early, middle and endgame (20, 20, 10 respectively).

Each problem consists of a whole board situation with 5 options marked, you should chose the one you think the best and you get a score from 2 to 10 for it (2, 4, 6, 8, 10).

I did this entire book yesterday and it took me around 4 hours. I got 1D in the opening, 3D in the middle game and 6D in endgame (first 8 problems got the right move and last 2 got second best). 3D overall. And this is exactly my problem with the book, as I'm currently a 4K player, so the scores seem to be VERY inaccurate.

By guess work you should get to around 1K with this book (6 avg per question, 20 questions. 120 is 1K), and assuming that from there you can only improve it's possible to see why the scores are so inaccurate.

overall I think it's still a good book since doing 50 whole board problems like this can really help and it's easy to learn new ideas with this book.

Have you tried the book? How much did you get in each section?

[Hushfield]

I currently seem to have a similar experience. The play.baduk.org strength test has returned a result of 2d the last few times I went through it, as did Miyamoto' Naoki's "What's your Rating?" (1975). As I'm currently ranked around KGS 7k, this seems really weird. So I here's my thoughts on these strength tests.
1) They test the upper limit of your current go strength, allowing you as much time as you need to read out certain positions. In actual games one often plays too fast, on instinct, without reading, could be having a bad day,... which seem to lower your rank (considerably). I think one of the differences between actual dan players and kyu players is that dan players are more experienced at maintaining this concentrated attitude of looking for the absolute best move during their own games.
2) Those tests often involve choosing between 5 possible moves. Since you can easily eliminate two moves which are clearly very small, or just plain bad, the problems boil down to pick one of the three. With some reading, this shouldn't be too difficult, even for mid-range kyu players. In your own games, you can't choose between options a, b and c. There's literally around a hundred different points you could play. Admittedly, some are much more attractive than others, but still. I guess another difference between kyu and dan players, is that dan players actually consider less moves, all of them better candidates.
3) Apart from the play.baduk.org one, the tests seldom specifiy the rank. EGF 2-dan? Japanese 2-dan? KGS 2-dan? They might very well be overrated.

Nonetheless, congrats on getting such a good score. It means your go knowledge and concentration during your solving of these the problems really was that good. Now for the difficult part: porting that concentration over to your own games.

[Csama]

haha, i actually had similar experience
one reason I can think of is doing multiple choice questions is different from doing short answers or essay questions
in a real game we don't have the potential answers provided to us
just take them as good exercises

[Marcus]

LOL, I consistently score as European 12k on the linked strength test ... I'm apparently REALLY bad at multiple choice.

I wonder what that says about my playing, as I'm currently 3k on KGS? Perhaps I just don't understand Go theory at all ...

[snorri]

I don't know about the book, but if it's similar to the linked to test (or contains too many opening or joseki problems), it's not going to be a very accurate measure of your playing strength. It's more a measure of scholarship, or, more likely, trivia. I wouldn't say understanding. So if your score is much higher than your rank, than it probably means you should work much more on fundamental skills like reading and less on adding yet more position-specific knowledge. If your score is lower than your rank, maybe you're not trying very hard or maybe you're closer to one of those unschooled street fighter types, in which case some proper study might help you a little.

Or better, just ignore tests and continue to do what works for you. One thing about go is that the only tests that matter are the ones your opponents give you and the ones you give to your opponents

[blindgod]

I currently seem to have a similar experience. The play.baduk.org strength test has returned a result of 2d the last few times I went through it, as did Miyamoto' Naoki's "What's your Rating?" (1975). As I'm currently ranked around KGS 7k, this seems really weird. So I here's my thoughts on these strength tests.

According to that test, I'm a KGS 6 kyu. However, whenever I play against anyone higher than about 18 or 19, I get thrashed. The only reasonable conclusion is that the people I'm playing are sandbaggers.

[Hushfield]

I currently seem to have a similar experience. The play.baduk.org strength test has returned a result of 2d the last few times I went through it, as did Miyamoto' Naoki's "What's your Rating?" (1975). As I'm currently ranked around KGS 7k, this seems really weird. So I here's my thoughts on these strength tests.

According to that test, I'm a KGS 6 kyu. However, whenever I play against anyone higher than about 18 or 19, I get thrashed. The only reasonable conclusion is that the people I'm playing are sandbaggers.

Actually, I don't think that's the right conlusion at all. What I'd take from that is that the test really does overestimate one's rank, because it seems to neglect one's actual fighting strength and reading abilities under time constraints. Perhaps we should play some KGS games and see whether the estimates the test gives us are anywhere near correct. Personally, I think they aren't, for the reasons stated in my previous post.

[Marcus]

I currently seem to have a similar experience. The play.baduk.org strength test has returned a result of 2d the last few times I went through it, as did Miyamoto' Naoki's "What's your Rating?" (1975). As I'm currently ranked around KGS 7k, this seems really weird. So I here's my thoughts on these strength tests.

According to that test, I'm a KGS 6 kyu. However, whenever I play against anyone higher than about 18 or 19, I get thrashed. The only reasonable conclusion is that the people I'm playing are sandbaggers.

Actually, I don't think that's the right conlusion at all. What I'd take from that is that the test really does overestimate one's rank, because it seems to neglect one's actual fighting strength and reading abilities under time constraints. Perhaps we should play some KGS games and see whether the estimates the test gives us are anywhere near correct. Personally, I think they aren't, for the reasons stated in my previous post.

Given my own results on the test, I suspect it relies heavily on theoretically "correct" plays, ones that can be demonstrated to be superior in a teaching environment. However, as is often the case, knowing the "correct" move to make doesn't really translate into understanding how to handle an opponent (like me) who makes theoretically sub-par moves.

It might be nice if I understood why certain moves were the correct moves in these positions, but not knowing hasn't really been holding me back from advancing in rank, at least on KGS. Once I stop progressing, maybe I'll take another stab at learning all this Go Theory ... for now I'll keep playing and get better that way.

[jts]

I consider myself to be much more of a bookworm player than a street-fighter player, and I got ten stones below my kgs rank. So don't think the test sucks in any systematic way, it just sucks.

[mmigo]

Is there a place to test my strength on the web? I have tested in play.baduk.org. Especially I every day am finding the book of Naoki Miyamoto. But no pace is found. And no version of PDF and of online. Someone can help me?

[CnP]

um KGS, IGS etc. Go strength estimate quizes are fun and the questions can be thought provoking but taking the ranks they produce seriously is a mistake. Your playing strength is the strength you play at.

[Mef]

One problem with tests like these, it's entirely possible to get the correct move for completely incorrect reasons. Finding a specific move to play in a specific position is not the same as developing and applying a coherent plan making that move meaningful in the full context of the game. I'm not familiar with the book in question, but I would ask this -- Did it provide 5-10 move continuations for each move? How many of those continuations did you foresee in context?

Also worth asking, in how many of your games do you spend 4 hours on 50 moves? (=

[Boidhre]

One problem with tests like these, it's entirely possible to get the correct move for completely incorrect reasons. Finding a specific move to play in a specific position is not the same as developing and applying a coherent plan making that move meaningful in the full context of the game. I'm not familiar with the book in question, but I would ask this -- Did it provide 5-10 move continuations for each move? How many of those continuations did you foresee in context?

This. As was pointed out to me in a review once: "That was a dan level move, the continuation though..."

[Uberdude]

Also worth asking, in how many of your games do you spend 4 hours on 50 moves? (=

I have played 294 games on OGS

[speedchase]

I play at a pro level for the first five moves...