How to tell whether a long winning streak represents real improvement or luck/other reasons?

In this book each kind of tesuji is introduced in a so simple way that you can search it too. So reinventing the wheel can become a wonderful experience. But if you can’t well you can decide to read it like an introduction.
The very first chapter (introduction with a complex problem) is the only chapter that is designed more like something to read as to search, it’s just pushing you to process later with organization.

3 Likes

I actually did forget the way it’s introduced, and went to check it again, it’s a pretty enjoyable book. I think though in some sense just seeing one example isn’t always enough to understand the moves and when they’re likely to work. Hence I still think you can probably just try to absorb the techniques quickly if you don’t have large amounts of time to stare at a diagram or set it up on a board :slight_smile: I agree though that for instance one would probably have to put in a bit more work, imagining which are key stones making a tesuji work or not, numbers of liberties at play etc (I don’t see why these things couldn’t be pointed out though in a book :slight_smile: )

I’ve been reading the SmartGo books version, which I find really convenient for testing out things :slight_smile:

2 Likes

Update: After that huge loss to a 14 kyu, a run of six wins and six losses (including three upsets against 17 kyu or weaker opponents) took my rank back down to 16 kyu, but four consecutive wins brought me up to 15 kyu.

Results vs 15 kyus: L+12.5, W+3.5, L+3.5, W+16.5
Results vs 16 kyus: W+16.5, L+6.5, W+R, L+R, W+R, W+10.5
Results vs 17 kyus: L+0.5, W+R, W+36.5, W+R, W+49.5, L+29.5, W+R

2 Likes

Nah what’s the point to focuse like that on your results?
Just let the system qualify you at this or that level, and enjoy your games and experimenting new ideas.

Or how you will handle the heat when entering later the SDK group?

2 Likes

I think it’s interesting and I’d say its relevant to thread :slight_smile:

Is there something in particular about the bold games?

Winrate is looking decent vs 17kyus, and 50:50 ish against 15-16kyu :slight_smile: Keep working and keep improving! :slight_smile:

1 Like

The bold games are the more recent ones (after the huge loss to the 14 kyu) while the earlier ones are from the winning streak period.

Per the thread title, I wanted to know whether I had improved, because that would impact my selection of study resources.

O fine, I didn’t get that you rely on the effectiveness of your study with winning strike to plan further topics.

I think if enough people tell you, “don’t study Y until you’re X kyu” maybe you start to follow the advice. It is a typical format of advice I’ve heard.

On the one hand I’d say study whatever you enjoy :slight_smile: (But I probably amn’t improving that fast so it could be bad advice, but it’s fun :slight_smile:)

At the moment what I try to do is play a couple of games and if I think there’s something in particular after reviewing that I need to work on, I try to study that. It doesn’t always happen, for instance I still need to sit down and study moyos properly. Sometimes there’s better ways and ideas to make or reduce moyos than I would do in a game.

Another thing I’d like to improve for myself is my endgame. I think I’m getting a bit better with the first step in just trying to prioritise moves I think are bigger point wise, even if I don’t have time to count the game or calculate the size or a move (I’m probably in byoyomi at endgame).

So that might be something I know I probably need to study but sometimes I just read or do tesuji or Life and death problems for fun. Or maybe I was just really interested in learning about the monkey jump so I look at that :slight_smile:

One thing I’ve found is that any one thing from tesuji or life and death books won’t really help in every game. Like if you wanted to practice the tombstone tesuji/two stone Edge squeeze or killing the L+1 group it probably won’t come up often. The good thing is that when it does maybe you can respond to it :slight_smile: eg above tesuji at move 38 Game without stones and L+1 group move 20 Tournament Game: Blitz 13x13 Swiss Tournament 2020-09-15 19:30 (65498) R:3 (GeorgA vs shinuito). Now whether it was the right move in those is another story :slight_smile:

In terms of study, what are the biggest mistakes you’re making? What could you do to improve them? Identifying your current weaknesses will tell you more about what you should be studying than looking at your current rank.

1 Like

Losing is sometimes the proof that you are renewing your ideas.

3 Likes

I can hardly believe this. I just beat a 14 kyu, albeit by 0.5 points!

Seven consecutive wins took me to 13 kyu, but I will not get too carried away because I am behind in a few games that are close to completion.

In my case, I am not sure how well I am applying what I have studied. For example, in past reviews, I was advised to try harder to play big and sente moves, but my attempted sente moves often end up being endgame moves too early or worse, thank-you moves. Likewise, I know common shapes and extensions, but think I am using them wrongly.

One definite weakness is counting (and ruleset differences). Relying on the score estimator is understandable for a TPK, but less excusable for a 15-16 kyu. In one virtually-completed game, the score estimator said I was ahead by 0.5 points, so I passed, then my opponent passed, but in scoring phase, after the dead stones were correctly marked, I lost by 0.5 points.

3 Likes

A lot of what you’re describing might come down to reading. For sente moves, are you considering both what your opponent’s response is going to be, as well as how much it’s worth? Endgame moves often are sente, just only in the endgame. If your opponent loses 8 points for ignoring your move, that’s a great endgame threat, but if you play it while there’s a 20 point move on the board you’re playing in the wrong spot. Thank-you moves, similarly, often arise because people don’t really read out the local situation to figure out what the strategic picture is.

Looking at a few of your recent games, you tend to play very passive moves and let your opponent get away with whatever they want. In the game you linked, consider move 71. Here, your opponent played a hane, and you chose to back of. That cost you about 10 points, and you even came back later and fixed a defect that resulted from that in Gote, for a few more points. There was no reason to back off, though: the atari you defended against by backing off instead of cutting doesn’t go anywhere. You descend once and then your opponent doesn’t have any real threats. Move 109 is a similar example, but smaller: you back off solidly instead of pressing at L3. Your shape after L3 is way worse, but it doesn’t matter: the ataris that your opponent have don’t go anywhere and you’re so strong locally that you never have to worry about an invasion. L3 also has a follow up at M2 for about 6 points in gote, where your opponent’s defense is also gote. Either you get the 6 point gote move in the endgame or you get a move somewhere else. The pull back is way too passive, and those kinds of moves seem like a common trend in your games.

This game is another good example of the same kind of behavior:

Your opponent surrounds you, and you resign in what should be a winning position. Your opponent’s corner isn’t alive yet, and they’re going to lose stones on the bottom defending it if you press the attack. Even if you don’t see that it’s attackable, the stones surrounding your group are riddled with weaknesses. You’ve got an easy push and cut at L9/N9, for instance, and your group is no longer surrounded, even if you can’t find another way to get an eye locally. You’re only considering defending your stones, here, and not looking for ways to conterattack your opponent or press their weaknesses. Move 35 in that game is another example of the sort of passive defense that shows up a lot in your play. You connect under in Gote when there’s no reason not to just play O7 and capture the cutting stone.

Either reading more accurately or more consistently making yourself read out all of the relevant sequences will address a lot of those issues. If there’s a weakness in your own shape: does it matter? What actually happens if your opponent tries to cut you? If there’s a weakness in your opponent’s shape, ask the same: does it matter? What happens if you cut it? Can you kill something or reduce something by attacking?

For study, you should be reviewing your own games and trying to identify the moves where you think you went wrong and figuring out what the best alternatives are. In particular, look for the kinds of moves you’re bad at making in games, namely moves that attack your opponent. Then ask other players to help review: the OGS community is great at responding to review requests! (Until you get to around 6k-5k. Grumble grumble.) External reading about topics like shape is useful, but it’s no substitute for opening the hood on your games and seeing what’s wrong with your moves.

Also: counting is moderately pointless at kyu levels. A good move is a good move, regardless of whether you’re ahead. In the game where you lost by 0.5 points, was there a one point move on the board somewhere that you missed? If so, you should’ve played it, regardless of the score. If there wasn’t, counting wouldn’t have made one appear. Were there any big endgame moves you should’ve played that you missed? Figure out what they were and play them in the future, regardless of what the score is. Knowing the score more accurately won’t make them appear, and you should be cognizant of the endgame regardless of how close the game is.

4 Likes