Sunday, November 30, 2008

Cobwebs, go away!

This weekend, I played my first couple slow games in three months, so I would have something for my coach to go over with me.

likesforests vs ICC 1978 - I'm paired up against a stronger player. When he responds to my English with ...c6 I decide to try something new and transpose into a Semi-Slav / Moscow. To my surprise I'm dominating him the whole game--then I blow a piece-up endgame.
lf-1978.pgn


ICC 1711 vs likesforests - A thrill-ride, especially after I forget my opening lines. I take his gambit pawn and try to hold on for dear life, but I'm forced to return it to stave off a crushing attack. A tactic in the late middlegame decides the game rather abruptly.
1711-lf.pgn


Not too bad, except for the embarrassing last move of my first game. :)

12 comments:

From the patzer said...

How much time did you have on the clock when you did lose that knight in the first game? I mean was it a total oversight or was it the nerves under timepresure?

Nice knight winning tactic in the second game! Always nice to see somebody punishing the mistakes of his opponent.

likesforests said...

I had 2 minutes left, but I played too quickly because I was worried about falling beneath 60-90 sec (I need that much time to mate with B+N, which was a possible finale.)

With a few moment of reflection the winning path seems obvious: 1.Ne2+ Ke3 2.Kg5 Kd2 3.Kg4Ke3 4.Nxg3 and now either use the knight as a shield while I march my pawn to promotion, or allow the B for P sac and mate with B+N, both endings I've practiced.

Yup, the second win was fun. :)

likesforests said...

I think my first lesson will be very short. "Stop hanging pieces!"

Kaan Kara said...

Or it might be something like:" use your time economically" and even it might be sth like "dont lose time with creating unefficient ways of winning" :P

Kaan Kara said...

By the way i really wondered what was the tempo of the game ? 60' ?

likesforests said...

Kaan Kara - Hehe. Those would help too, maybe more! Since they were played on ICC, my coach will know exactly how much time I spent on each move and there are no "lost" moves at the end of the game. The time control was G/45.

Blue Devil Knight said...

To risk being the Debbie Downer, it seems a bit strange to play so few slow games when trying to optimize improvement (as I mentioned here in the study you cited, they seem to think study is less important than other factors...perhaps playing?).

To put the shoe on my foot, I am suboptimal all the way. No coach, wasting time studying openings, playing too much blitz, that's not the way to improve.

Blue Devil Knight said...

In swimming, we usually do one length drill followed immediately by one length just swimming.

E.g., one length dragging my fingers across the top of the water to practice keeping my elbows up, then the next length just swim to integrate the drill into my full stroke.

The same applies in chess, I think. When I was in the Circles, at any rate, my play was actually disrupted as I was in this weird tactical drill mindset.

Anyway, I only bring this crap up again as we never really finished the discussion at the last post.

likesforests said...

Dear Debbie Downer from Duke,

"seems a bit strange to play so few slow games when trying to optimize improvement"

Absolutely. Around 3-6 slow games per month is optimal. By my own measure, I need to play more.

In Aug I played 19 25+5s. In Sep I played 8 25+5s. In Oct I played 2 25+5s. Not very consistent?! This was my first "slow" G/45 since... the World Open 4 mos. ago.

"as I mentioned here in the study you cited, they seem to think study is less important than other factors...perhaps playing?)."

Regarding the study,

"The median number of hours of study of chess in an average week was 4.5 hours (6 hours for the masters)"

To me this implies that 6 hours of study/week is healthy.

"but this varied widely; from 0 to 50 hours. There was a correlation between amount of deliberate practice from the age of serious play and peak rating (and with current rating). However, the correlation was quite small, suggesting that other factors are
more important."

His conclusion is not worded as clearly as I would like. You and I can agree studying 0 hours/week is correlated with stagnation. And there must be some # of optimal hours of study, beyond which we get very diminishing returns.

Also, note table 3. Do you feel playing or study is more important? 14% say playing, 14% say playing, 70% say they are both equally important.

So I fall back on the median # of hours studied in his survey, and other studies which survey # hours of deliberate practice obtained in 10 years and correlate that to one's peak rating. That turned up at least 6 hours/week as optimal if one wants to reach master.

...again with the caveat that, there's no guarantee that what it took in the past is what it will take in the future. The world of chess is changing more rapidly than the world of basketball.

"To put the shoe on my foot, I am suboptimal all the way. No coach, wasting time studying openings, playing too much blitz, that's not the way to improve."

Funny enough, your post on the Slav made me review it, and that's why I employed it against the 1974 successfully. I've never replied to 1.c4 c6 with 2.d4 before. I had an edge out of the gate. Thanks!

Chessaholic said...

Nice games! Well, despite the fact that you're not playing as many slow games as you would like to, you're still playing a lot more of them than I do :) But I will finally play in a tournament again this weekend, hope I am not too rusty. We shall see.

As far as 60.Kh5 in the first game - we've all been there :)

Blue Devil Knight said...

LF: that seems reasonable. I was not trying to be a dink shanker, but just finishing up the discussion from last post.

I don't completely trust someone's judgement of what is most important, but think it is more crucial to look at correlations between the variables and rating (e.g., if they all said that they were equally important, but in fact study wasn't correlated at all with rating, but slow game/postmortem was highly correlated, we would learn they are just not very good at evaluating what makes them better). Of course that isn't what they found: study was important too.

My bias is toward playing slow games/postmortems (or studying that closely simulates play conditions) probably because of Rowson's book. But my old coach (an IM) clearly got a lot better by combining study and play.

Also, the truth is probably it depends on the person. As our age increases, the optimal ratio of study to play likely changes. It also depends on the type of person we are, for instance what you enjoy (if the fun is gone then why bother?). So I don't want to give the impression that there is some universal formula.

It is too bad there aren't more well designed longitudinal studies of this kind of thing, starting with a pool of, say, 1000 chess players of different ages and tracking their progress over time. The self-initiated web questionnare has some serious problems methodologically (there can be selection bias, for instance, and now this could be even worse as they've released preliminary results and using them to promote more responses!).

OTOH, it is hard to get a grant for longitudinal studies even in medicine!

It also probably depends on what aspect of the game we're talking about. Perhaps for openings and endgames, study is really important, but for the middlegame (aside from gaining basic tactical competence) Rowson's simulation-style study is better.

If there are any millionares out there that want a good study done, contact me and we'll get it rolling! :)

likesforests said...

"I was not trying to be a dink shanker, but just finishing up the discussion from last post."

I was just kidding around. Your thoughts are always very welcome. :)