This has been touched on a lot, but I haven't seen it a coined term. The topic "I'm stuck" is one example.

Paradigm shifts: Periodic scientific revolutions.
Since I'm lazy: http://en.wikipedia.org/wiki/Paradigm_shift This explains paradigm shift.

After reading Fwiffo's latest blog entry, I've decided to abuse this term by applying it to the periodic stages of improvement in go. So basically, if you're in a bad mood, and you're still playing go, this is usually a downwards slope in your period. You can review it at a better time laughing at your stupid mistakes, this is usually an increasing slope in your period.

My math skills are dreadful, but bear with me. The equation I would use to show the line of the improvement graph would be something like y=(#x+#)*sin(#x+#). Where #s are whatever constants that fits. (one # might not be the same as another #)
This is of course, too basic an equation, because there are definitely stronger periods of improvements than others. so you might want to multiply that by sin^6(5x-2)*cos(3x-5) or something.

so in the case of kghin on KGS, it looks like a wavy line slanting upwards.

but what I'm trying to get at is that the graph could be increasing, or decreasing at the moment, but as long as you stay human, your time of increase will come. That is, unless if you are unable to play, due to death and all his friends.

Hopefully I didn't lose too many people <.<

_________________
LoL forum:
"Complexity is Magic: The Gathering, lots of tiny rules and special cases, and basically every single card is a unique case. Depth is Go: the game has 4 rules, and it takes literally decades to learn how to play that game well."

well, there's an easy way to test,

just do fourier transforms on all the KGS rank graphs, and see if you can see a pattern

http://en.wikipedia.org/wiki/Fourier_transform

_________________
Tactics yes, Tact no...

Hmm... not what I had in mind, but it's much more intriguing and correct for the term: paradigm shifts. I think...

Since I haven't had any prior knowledge of Fourier transforms, I don't know much about this and I'll probably say something totally incorrect, but from what I've just read, it seems like an awesome idea.

However, I was thinking in more individual terms, but then I'd be digging myself a big grave of incorrectness.
If we could also know a certain strength as what kyu, at all times. Then it could be fathomable. We could also pick out the outliers (sandbaggers and other), but I'm going in depth into uncharted territory.

_________________
LoL forum:
"Complexity is Magic: The Gathering, lots of tiny rules and special cases, and basically every single card is a unique case. Depth is Go: the game has 4 rules, and it takes literally decades to learn how to play that game well."

Maybe if you squint a bit and your eyes are watering...

Is it possible to get raw data for rank graphs? If it is, I'd be interested in doing some FFT and looking for results. Of course, you'd have to remove the non-periodic increasing part, maybe through linear regression.

Yeah, I'm not sure actually, It obviously exists somewhere, but I don't think it's ever been made public.

As far as the linear portion, there's no need to extract that, that's just going to be the first mode no?

_________________
Tactics yes, Tact no...

Fourier transform on the derivative of the graphs probably makes more sense than on the graphs themselves.

wow.. we have many math wizes in this forum...
i majored in math but only heard of that term (fourier transformation).
maybe i did learn and forgot. its been too long...

_________________
"The more we think we know about
The greater the unknown"
Words by neil peart, music by geddy lee and alex lifeson

fourier transformation? Isn't that very common in physics?

_________________
长考出臭棋.

Anything sound-related has Fourier transforms out the wazoo.

Yup. NMR imaging wouldn't work without it.

_________________
Stay out of my territory! (W. White, aka Heisenberg)