Lessons from blitz
It has been about a month since I hit chess puberty. This metamorphosis included a lot of changes: more reckless, aggressive, antipositional playing style, a desire to play lots of blitz just to gain more experience on the 64 squares. Some commenters acted the concerned parent, but I just had to fly, had to stretch my newfound wings. And just as I'm glad I moved out of my parent's house (indeed it is really weird to live with your parents past a certain age), I am very glad I took this flight.
This meta-blitz has taught me a lot of lessons about chess in general, and the pros and cons of playing lots of blitz more specifically. Note that these lessons probably won't apply to more advanced players who already have superb tactical acumen, an acumen I clearly don't possess.
1. Error patterns
Ninety percent of the games were decided by tactical blunders. The following plots the proportion of tactical errors, sorted by type, culled from looking over about 60 of the games:
Note that the highest proportion of errors are the simplest possible blunder: leaving pieces en prise, either my own or my opponent's. Such blunders make up nearly a quarter of my errors in these blitz games. Second most frequent are counting errors. The rest of the errors, from neglecting to consider checks, to calculation errors, make up a much smaller percentage of my mistakes.
I think this is largely a result of the fact that it is blitz that I'm playing, but I will next do the same analysis on 100 slow games, to see if the patterns are similar. My guess is that in slow games I will have fewer en prise blunders, more calculation and counting blunders. Not to say I don't make outright blunders in slow games, but the proportion should be less (I hope).
This reinforces the importance of using some kind of thought process that includes a blundercheck mode, a step that is easy to neglect in blitz games when I get too caught up in my own plans and schemes. It is quite amazing how easy it is to miss a piece out there hanging out, ripe for the picking.
2. The predominance of short combinations
Going through the errors also revealed a very interesting property of tactical opportunities. There were hardly any complex combinations available in any of the games. Perhaps in 3% of the games, I missed four-or-more move combinations. Most realistic combinations are two or three move, typically one move. This is an extremely useful fact, and should be impressed into the minds of all beginners. When I first started playing chess, I looked at the board as a structure with infinite tactical possibilities that were well out of my reach, I would sit and search for complicated N-move combinations, wrongly believing that they must be there, but that I was just too stupid to see them. My post-mortem showed me how naive my thinking was, and this is liberating.
The law of short combinations also makes sense from an analytical point of view (and could probably be proven mathematically): the longer the imagined combination, the more likely it is that the opponent will have defensive resources, will have in-between moves that are hard to see, the more likely it is that you are simply missing an obvious weakness in your attack or somehow miscalculating the combination. It reinforces the brilliant insight provided by GM Ziatdinov, which I found out about from DK-Transform:
This all makes me extremely happy that I am doing the circles with Chess Tactics for Beginners, which is training me in combinations that are typically no longer than four moves in length. Until I can spot such tactics without error in all my games, the later levels of CT-Art are not time well spent.
The only exception to this short-combination trend that I found in my games were sequences with forced moves (i.e., mating nets, sequences of checks, threats against the queen that couldn't be met, or counting problems with sequences of captures). Luckily these are also relatively easy to calculate through, as there are typically not tons of side-variations as the moves are usually forced (though not always, as even counting problems can be disrupted by a smart in-between move). This is proably why the GMs say that in quiet positions, you shouldn't waste a lot of time working through variations in your head.
3. Memorizing opening variations is useless
Another positive lesson from blitz is that the opening doesn't matter. Coming out of the opening with a 0.5 pawn deficit doesn't mean shit if I drop a piece, and almost all of the games are decided by blunders in which the evaluation graph drastically fluctuates during the game (just as MDLM observed). While avoiding opening traps is important, that is essentially a tactical matter. To reach the levels I want to reach, detailed opening study is not necessary, especially if I have a good grasp of the opening principles.
4. Strategy ain't that important
This is a corollary of the discussion in 2 above. Having a backward pawn doesn't matter if you are down a piece. Until you have stopped dropping pieces to simple tactical combinations, you just don't need to spend a lot of time thinking about subtleties of pawn structure. The only strategic consideration that really influences things, precisely because it correlates highly with tactical opportunities, is piece activity, and not typically of a subtle variety (more of the 'My bishop cannot move so I need to activate it' variety).
This also suggests that studying games between GMs is not all that important for beginners. In such annotated collections, typically you have two people playing extremely well tactically, and the entire game hinges on extremely subtle positional factors. In real games between novices at my level, this is not how it works. You are not learning the kinds of things that will make a big difference in your games. Once you have stopped dropping pieces, though, studying GM vs GM games will probably be more helpful. (This is another implicit advertisement for the Euwe's Master vs Amateur book, which shows master versus amateur games).
5. Endgame expertise is not important
I am not losing games because I don't know how to mate with a Q against a R. In most of the games, if they even reach an endgame, the material disparity is so large that subtle endgame knowledge is simply not needed. I know how to mate with the two bishops, B and N, and Q versus N. None of these have ever been useful in a real game. While they say you should 'start' with the endgame, they should also say 'Don't learn too much endgame if you still drop pieces.' Basic K/P versus K is extremely useful at my level, as are the basic mates. Beyond that, not much is needed. Silman realized this and brilliantly incorporated it into his recent book (which I initially thought had ridiculously low expectations of people rated U1400: these low expectations are based in practical considerations about what you will actually need at those levels--more Kudos to Silman!).
6. Experience, for beginners, is key
My coach (an IM), Chandler (in How to beat your dad at chess), and GM Ziatdinov (above) all recommend that beginners simply play. And play some more. Build up skills, practical experience, intuition. Tactical exercises are great, but playing thousands of games, and losing lots of them, will build up a bedrock of experience that is indispensible. Especially when starting out in chess, playing mindful blitz (thanks, Wahrenheit), going over the games quickly afterwards to get a sense for the patterns of error and success, is extremely helpful. Yes, there is the danger of becoming one of those people at ICC with 10,000 games and a rating of 900, but my hunch is that if you have a half decent head on your shoulders, and use it to analyze your blitz losses, that won't happen. There will always be time for slower games once you feel you have hit a blitz plateau.
So, beginners, play. Lose. A lot. Get nailed with the back rank mate 10 times. If you analyze afterwards you will slowly build up a feeling for safe positions, dangerous positions, positions in which crazy fun speculative sacrifices are likely to work, and those in which they are not. Most importantly, you will not fall into the trap of becoming a chess scholar (on the same level as philosophers in academia in the heirarchy of tournament chess), someone who has tons of book knowledge, subtle understanding of the intricacies of move 19 options in some strange variation of the Philidor Defense, but gets his ass kicked in 20 moves by the kid who just started playing 6 months ago.
7. The downsides
There are some clear downsides to blitz overdoses. For one, in slow games, it is hard to get back into the careful mindset required in such controls. I have found that I have started to get bored in slow games: it just isn't as fun as the crazy pie-throwing contest that is blitz.
Second, blitz is largely a pie-throwing contest. A blitz game itself largely tests the knowledge you already have, how quickly you recognize basic tactical motifes, rather than the depth of your understanding of the beautiful subtle aspects of the game. You are also likely to not remember the mistakes you made, and definitely not likely to learn anything new about the game during the game; the longer the game, the less likely you will have these problems. You need to be careful of playing blitz mindlessly if your goal is to improve: if you just play and never think about your losses, you could end up that 900 rated person with 10 zillion games. It's important to play blitz mindfully.
Third, your games will simply be uglier, unimpressive, and messy. That is, blitz games don't have the aesthetic of a well-thought-through slow game.
OK, there's my mind dump. So far, it has been very fun, but I will now play about 100 slow games to get a sense for my error patterns. It will be interesting to see how they compare to the errors in blitz games.
This meta-blitz has taught me a lot of lessons about chess in general, and the pros and cons of playing lots of blitz more specifically. Note that these lessons probably won't apply to more advanced players who already have superb tactical acumen, an acumen I clearly don't possess.
1. Error patterns
Ninety percent of the games were decided by tactical blunders. The following plots the proportion of tactical errors, sorted by type, culled from looking over about 60 of the games:
Note that the highest proportion of errors are the simplest possible blunder: leaving pieces en prise, either my own or my opponent's. Such blunders make up nearly a quarter of my errors in these blitz games. Second most frequent are counting errors. The rest of the errors, from neglecting to consider checks, to calculation errors, make up a much smaller percentage of my mistakes.
I think this is largely a result of the fact that it is blitz that I'm playing, but I will next do the same analysis on 100 slow games, to see if the patterns are similar. My guess is that in slow games I will have fewer en prise blunders, more calculation and counting blunders. Not to say I don't make outright blunders in slow games, but the proportion should be less (I hope).
This reinforces the importance of using some kind of thought process that includes a blundercheck mode, a step that is easy to neglect in blitz games when I get too caught up in my own plans and schemes. It is quite amazing how easy it is to miss a piece out there hanging out, ripe for the picking.
2. The predominance of short combinations
Going through the errors also revealed a very interesting property of tactical opportunities. There were hardly any complex combinations available in any of the games. Perhaps in 3% of the games, I missed four-or-more move combinations. Most realistic combinations are two or three move, typically one move. This is an extremely useful fact, and should be impressed into the minds of all beginners. When I first started playing chess, I looked at the board as a structure with infinite tactical possibilities that were well out of my reach, I would sit and search for complicated N-move combinations, wrongly believing that they must be there, but that I was just too stupid to see them. My post-mortem showed me how naive my thinking was, and this is liberating.
The law of short combinations also makes sense from an analytical point of view (and could probably be proven mathematically): the longer the imagined combination, the more likely it is that the opponent will have defensive resources, will have in-between moves that are hard to see, the more likely it is that you are simply missing an obvious weakness in your attack or somehow miscalculating the combination. It reinforces the brilliant insight provided by GM Ziatdinov, which I found out about from DK-Transform:
I teach practical tournament chess. When it comes to tactics, I believe only in clear 1-2 move combinations. These combinations occur in every game, even between strong players, but most people cannot wait for simple combinations.Perhaps for players rated around 1600 this wisdom does not apply: I don't know, and don't particularly care. My goal is to reach 1500 and then maintain that level of performance so that I can enjoy a good game of chess, but get back to a balanced life.
You can win without strategy. If you do not apply effective tactics on every move, you will not survive long. No amount of planning for the next few moves does any good if your forces are destroyed in the current position. This does not mean that long-term strategy is not important, particularly as a context for tactics, but the outcome of most games boils down to which person sees better tactically in the present situation.
In chess, there is strategy and tactics. Strategy involves long-term concepts, while tactics are immediate. Strategy is academic and theoretical; tactics are practical and concrete.
[An] effective method for improving your chess tactics is blitz chess...Blitz is about developing tactical bravery and intuition. Practice this form of tactical training and then analyze your chess "instincts" after the game with a computer.
This all makes me extremely happy that I am doing the circles with Chess Tactics for Beginners, which is training me in combinations that are typically no longer than four moves in length. Until I can spot such tactics without error in all my games, the later levels of CT-Art are not time well spent.
The only exception to this short-combination trend that I found in my games were sequences with forced moves (i.e., mating nets, sequences of checks, threats against the queen that couldn't be met, or counting problems with sequences of captures). Luckily these are also relatively easy to calculate through, as there are typically not tons of side-variations as the moves are usually forced (though not always, as even counting problems can be disrupted by a smart in-between move). This is proably why the GMs say that in quiet positions, you shouldn't waste a lot of time working through variations in your head.
3. Memorizing opening variations is useless
Another positive lesson from blitz is that the opening doesn't matter. Coming out of the opening with a 0.5 pawn deficit doesn't mean shit if I drop a piece, and almost all of the games are decided by blunders in which the evaluation graph drastically fluctuates during the game (just as MDLM observed). While avoiding opening traps is important, that is essentially a tactical matter. To reach the levels I want to reach, detailed opening study is not necessary, especially if I have a good grasp of the opening principles.
4. Strategy ain't that important
This is a corollary of the discussion in 2 above. Having a backward pawn doesn't matter if you are down a piece. Until you have stopped dropping pieces to simple tactical combinations, you just don't need to spend a lot of time thinking about subtleties of pawn structure. The only strategic consideration that really influences things, precisely because it correlates highly with tactical opportunities, is piece activity, and not typically of a subtle variety (more of the 'My bishop cannot move so I need to activate it' variety).
This also suggests that studying games between GMs is not all that important for beginners. In such annotated collections, typically you have two people playing extremely well tactically, and the entire game hinges on extremely subtle positional factors. In real games between novices at my level, this is not how it works. You are not learning the kinds of things that will make a big difference in your games. Once you have stopped dropping pieces, though, studying GM vs GM games will probably be more helpful. (This is another implicit advertisement for the Euwe's Master vs Amateur book, which shows master versus amateur games).
5. Endgame expertise is not important
I am not losing games because I don't know how to mate with a Q against a R. In most of the games, if they even reach an endgame, the material disparity is so large that subtle endgame knowledge is simply not needed. I know how to mate with the two bishops, B and N, and Q versus N. None of these have ever been useful in a real game. While they say you should 'start' with the endgame, they should also say 'Don't learn too much endgame if you still drop pieces.' Basic K/P versus K is extremely useful at my level, as are the basic mates. Beyond that, not much is needed. Silman realized this and brilliantly incorporated it into his recent book (which I initially thought had ridiculously low expectations of people rated U1400: these low expectations are based in practical considerations about what you will actually need at those levels--more Kudos to Silman!).
6. Experience, for beginners, is key
My coach (an IM), Chandler (in How to beat your dad at chess), and GM Ziatdinov (above) all recommend that beginners simply play. And play some more. Build up skills, practical experience, intuition. Tactical exercises are great, but playing thousands of games, and losing lots of them, will build up a bedrock of experience that is indispensible. Especially when starting out in chess, playing mindful blitz (thanks, Wahrenheit), going over the games quickly afterwards to get a sense for the patterns of error and success, is extremely helpful. Yes, there is the danger of becoming one of those people at ICC with 10,000 games and a rating of 900, but my hunch is that if you have a half decent head on your shoulders, and use it to analyze your blitz losses, that won't happen. There will always be time for slower games once you feel you have hit a blitz plateau.
So, beginners, play. Lose. A lot. Get nailed with the back rank mate 10 times. If you analyze afterwards you will slowly build up a feeling for safe positions, dangerous positions, positions in which crazy fun speculative sacrifices are likely to work, and those in which they are not. Most importantly, you will not fall into the trap of becoming a chess scholar (on the same level as philosophers in academia in the heirarchy of tournament chess), someone who has tons of book knowledge, subtle understanding of the intricacies of move 19 options in some strange variation of the Philidor Defense, but gets his ass kicked in 20 moves by the kid who just started playing 6 months ago.
7. The downsides
There are some clear downsides to blitz overdoses. For one, in slow games, it is hard to get back into the careful mindset required in such controls. I have found that I have started to get bored in slow games: it just isn't as fun as the crazy pie-throwing contest that is blitz.
Second, blitz is largely a pie-throwing contest. A blitz game itself largely tests the knowledge you already have, how quickly you recognize basic tactical motifes, rather than the depth of your understanding of the beautiful subtle aspects of the game. You are also likely to not remember the mistakes you made, and definitely not likely to learn anything new about the game during the game; the longer the game, the less likely you will have these problems. You need to be careful of playing blitz mindlessly if your goal is to improve: if you just play and never think about your losses, you could end up that 900 rated person with 10 zillion games. It's important to play blitz mindfully.
Third, your games will simply be uglier, unimpressive, and messy. That is, blitz games don't have the aesthetic of a well-thought-through slow game.
OK, there's my mind dump. So far, it has been very fun, but I will now play about 100 slow games to get a sense for my error patterns. It will be interesting to see how they compare to the errors in blitz games.
posted by Blue Devil Knight at 7/26/2007 11:15:00 PM
30 Comments:
first, thank you for the acknowlegement.
it is true, while i dont play chess year round, instead studying in massive if not heady doses (in bed, at the beach, at the toilet, during sleep or naps i often see things the best),
lately i have been hitting it super hard as i do several times a year, as friends here know.
and of course, i have a daily excel spreadsheet on effort and hours spent (a secret, i dont want mental health folks knocking at my door), but played 748 games in 31 of the last 34 days. that translates to 24.13 games per day.
i am tired, but twice came within TWO, then ten elo of my goal. i cannot stop till done.
but, then again, as Charlie Munger said when Warren Buffet asked him if it was ok if he buy TEN PERCENT of the General Dynamics Corporation (the largest specialty defence contractor on the planet, tanks and submarines: chump change for the likes of these two dynimos):
Buffet: "He said it was ok if it kept me out of bars..."
A close friend at work told his wife: "David quit drinking--he has chess instead".
yes, i spread the word to all you sharp geniuses and barely sub-geniuses, but please credit the most excellent Underpromoted Knight for reminding me of it's existence.
my next Wild Card post will do one better, but that must now wait till i can write more, or find time to write more.
-------
have you discovered the verity of increment?
not 5/0 or 10/0 or 3/0
but instead:
3/3, 2/12, and my beloved 0/4
(3:04+ for 43+ moves.
i have gotten, i can assure you, to try and at once face EVERY opening variation in my reporatore in the last four weeks, and then of course my 1350 games at icc, nov to April, with months off between?
its all there, for me to data mine after im done--hopefully soon, then move back to 2/12 after slightly more than half a year...
combine the two? 2000 games since Jan or Dec and I have taken months off.
when i go to bed, i cannot sleep, but my head is swimming.
but this is deeply compressed, carbon like chess KNOWLEDGE.
that tests intuition.
i beat a 1900 the other night, so im getting closer.
but far from that...
lastly, i agree, again, dont study openings, as i have said again and again, but this doesnt mean we dont get, like a gigantic fractal, to see how it spins and learn from our learning...
...
i cannot tell you, how many times i think: "oh, im in this mess again..."
all the endings, the R and P.
the B vs. N.
the K vs pawns.
play 2000 games then let me know.
------
it is not in Euwe or Dvoretsky or Soltis, but all inside you.
warmest, dk
There were hardly any complex combinations available in any of the games.
I have to chew on this one. I guess you have let Fritz have a look at this?
Hi, very nice post. I'm curious as to what time controls you have been using in your blitz games?
BDK,
Great post as usual. I mostly agree with your points of view. I only differ slightly at the opening: I think it is a good idea to have some very basic knowledge of the plans of the opening. This helped me a lot.
I also agree with tempo about the long combinations. They do appear in my games frequently, but 95% of the time they are really hard to spot.
As you said, however, the basic, 1-2-3 moves tactics are the bread and butter of chess. You have to master them to improve.
Cogent, concise, compelling.
Excellent post, full of wisdom, a must-read.
"... I looked at the board as a structure with infinite tactical possibilities ... I was just too stupid to see them."
Totally resonates with my experience. Although I haven't had the opportunity to test my new approach in OTB tournaments, one of the benefits of moving away from such a wrong-headed belief is that online blitz play is more enjoyable.
Regards, HDK
Tempo: yes, I fritzed the games--basically all of the big fluctuations in the evaluation function were from low move-number combinations (one move tactics are most common, and then it drops off exponentially with number of moves).
Perhaps this doesn't apply at the more advanced levels, where you no longer make one-move mistakes, so those opportunities don't even come up!
That would make sense: the higher your rating, the more that curve is damped down for small numbers. I would assume the curve is the same for the larger numbers, which means at high levels tactical opportunities simply don't come up that often.
It would be very interesting to get empirical frequencies of one- two- three- up to N move tactical opportunities in games, as a function of rating.
It would be a lot of work, but incredibly helpful.
Oops. Missed the preview button.
Anyhow, I know Ziatdinov is a GM, but he is just wrong- I am going to respectfully dissent here.
Blitz simply does not give enough time to a non-Master to play non-hope chess. The main benefit of blitz is to give a superficial player the opportunity to level the playing field with a deeper player.
When both players are playing superficially, there is a greater random factor- anything can happen, and this benefits the weaker player.
Even at slower clock times, it is a CONSTANT struggle to play real chess. I can see how blitz has a certain appeal, because you are kind of relieved of the horrible burden of being forced to play real chess on every move of every game. The tense, nervous strain can sap the fun out of things, sometimes.
This is where Josh Waitzkin and his Zen-style philosophy might come into the picture.
OTOH, I can look back at some really cool, deep positional and tactical ideas played both by me and against me in slower time games. These games are keepers and I have plenty of them. ( plenty of throw-away slow games, too, to be fair)
I would be surprised if any of the blitz games are keepers.
FF: Uh oh. I could just refer to my comments at this post where you first started warning against the evils of blitz and how it is good for nothing because Heisman says so. Now that I have played some blitz, it is more clear that your worry was misplaced.
The goal in blitz is obviously not to have ten minute deep thinks. Playing real chess is not the only way to improve at chess: sure it's the only way to do well in slow games, but that obviously isn't the goal in blitz. Rather, my goal is to learn tactics, gain experience, and have fun.
I know Heisman says that the way to get better at blitz is to play slow games. This is nonsense. That is one way to improve at blitz: it is also possible to get better at blitz by playing blitz mindfully. My blitz rating went up 200 points in the past month, for instance, probably because it's the first time I actually analyzed my blitz games rather than just let them sit on my computer.
It is also possible to get better at slow games by playing blitz, especially if your main weaknesses are lack of experience and tactical crappiness (bearing in mind the caveats from my post on the downsides of playing blitz).
This is trivial to prove. Take two kids, both only know the moves of the game. Have one play blitz for a month (10 games a day), the other do nothing. Then have them get together and play a slow game. Who will win? If blitz was the evil skill killer you portray, the kid who does nothing should do better than the kid who plays a bunch of blitz. The key is that the blitz kid will build up experience, and some simple tactical skills (e.g., don't drop pieces) that will still be there in the slow games. His main worry will be that he needs to slow down, be more patient, and not play the slow game like he is playing blitz. There are tradeoffs.
It would be interesting to compare two kids if one had been playing slow games, the other had been playing blitz, so that there was the same total amount of time played. Then have them play blitz and slow games against each other.
At any rate, this debate is fairly silly because it's not like blitz and slow chess are mutually exclusive. I look at them as complementary approaches for an overall balanced improvement plan. One tends to build up thinking skills and appreciation of strategic factors (which even in slow games are not that important at my level, but they can at least be appreciated), the other tends to build up raw experience and tactical skills. Which is more important for the beginner? Clearly, raw experience and tactical skills!
I agree that blitz games tend to not be keepers (indeed I already listed this as the third downside of blitz in my original post). They aren't games you'd want to publish in Chess Life. That doesn't mean they aren't helping you on some aspects of your game.
I have had more than a hundred students and nobody had enough will power to finish this tactical training method. Is it my students or is it me? Well, take only thirty minutes a day and slowly memorize 1000 problems; take a year or two if you have to. It comes down to will power, and that I cannot provide.
The above is from Grandmaster Z link . Congratulations it appears you have successfully done what 100 of his students did not do.
BTW, I normally play 20+6 time controls are your Blitz games quicker than this.
Your findings seem to match MDLM to the letter. Are any of your findings inconsistant with his experience?
Very, very useful post full of impactful information--I'm going to do some commenting and expansion on it over at my blog. It really brings a lot together in a concise package. Thanks!
Tak: I play 2 12 or 3 8 usually. I really hate games with no increment as then it becomes a mouse race (as DK puts it).
Warheit: thanks for the comment: your phrase 'mindful blitz' on the earlier post has been very useful in guiding my approach.
i can assure you, i have dozens of keepers, amazing positions, filled with endgame ideas, cannons of tactics, and worthy of SLOWER study.
in massive rapid or blitz play, one can mine for and find tons of archtypal positions, that illustrate fundamental endgame problems, opening structural problems, and of course, tactical problems.
even if the quality of decisions is reduced, the positions themselves stand on their own, in tabula rasa form,
AND THE FACT THAT THEY ARE YOUR OWN positions MAKES YOU really WANT TO LOOK AT THEM.
my goal is not only to practice play, en masse, but to DISCOVER positions, which will endure in memory, and, believe you me, they do.
i am willing to venture, that i have discovered more in my last four weeks than half of the critical readers contra rapid/blitz here in their last four months.
if the spin is fast, so is the discovery.
the caviat is to ACTAULLY look at the games. that is the validating factor.
I agree with many of your points- but you don't have any support for the denial of heisman's claim that slow chess builds blitz skill. Might be interesting to survey the relative amounts of time that top GMs spent playing blitz versus slow.
BDK- BTW, I researched it, and my hunch that blitz serves as a tool to allow patzers to beat good players, has been borne out by scientific research.
http://users.fmg.uva.nl/hvandermaas/papers/chesspaperfinal3.pdf
So if you become a killer chess player, and you still enjoy getting beaten by patzers, blitz chess is your best shot! (;-)
FF said:
I agree with many of your points- but you don't have any support for the denial of heisman's claim that slow chess builds blitz skill.
I don't deny this. I just deny that it is the only way, and I am not convinced it is the best way. Most blitz players probably don't spend a lot of time analyzing their games, and that could explain the lack of improvement in blitz players.
Here is the link to the paper FF mentions. That looks extremely interesting, with important implications for how we think about chess thinking.
Also, I think DK makes a very good point: all sorts of interesting and analysis-worthy positions come up in blitz. Even if you screw them up at the time, that doesn't mean you can't spend a whole lot of time analyzing them afterwards. And since they are from your actual games rather than books, they are more likely to stick in memory and be relevant for future matches.
What stands out, for me, was the
study by
Calderwood, Klein, Randall (1988)
that showed that while all classes of player suffer from faster time controls ( no surprise there!), that the weaker you are, the MORE you suffer from fast time controls.
In short, by the time you have reached Kasparov's level, blitz will bring you down to 2600 from 2700 - but a novice will be brought down to chimpanzee level (;-)
dear funky fathom, i value many of your comments. but your are filled with accustation and presumption.
since i do not depricate you, then why must you throw stones?
:
218W, 420L, 30D = 657 games
can you do that? can you? have you ever? only 33.1% of games wins.
does that sound like being beaten by a bunch of patzers?--patzers, in this instance, folks below your skill class? please.
70% of the folks i play are 150 to 200 elo above me. many of those loses are to folks 300+ above me.
i played a bunch of games with a 1900-1950 bullet player (u realize, these are NOT 1/0 or 2/0 or even 3/0 but increment bullet), and lost
eighteen games in a row. i won the nineteenth.
you really think id play him again and again if i didnt have a slight advantage (at times almost just winning) 25% of the time? i was. but that is why he is 1900, he can convert games, even when i had an edge. by outfoxing me, or seeing IT faster.
so, my point is, stop the presumption. you dont know who i play, or how much or by what degree.
you give credit for nothing, which says much about you. sometimes i beat folks 200+ above me two times in a row, then win a third to somoeone 100+ elo, but then playsomeone rated way above me, and loose, then loose, then loose some more. such an approach cannot be without merit.
you could say, good effort DK, you could say that. but you dont. the focus is on your repudiation and disparagment. i will see if you post, then see what you say! yes, we will.
if the hardest trained runner or cyclist is running up mountains or riding up them, day after day, even if they are 'tired', they are getting a good work out.
that, i think, is what we are discussing.
yes, exactly as i thought:
you dont post, because you are not strong but wish to be able to snipe but not allow anyone to talk to you or put your thoughts in stone, except as a sniper.
if you are so strong, show us. show us how strong you are.
what is your rating?
what is your background?
what are your percentage of wins?
lets see how many patzers youve avoided. damn you.
go to any place such as playchess, icc, fics or even yahoo, and count how many persons have a 33% win ratio?
less than 5% if not more like 3%
you really, really have me pissed off. 750 *&%^% games, hours and hours, many, many loses against ++ elo, way beyong what you face.
tell me otherwise.
damn you for saying what you said, you bastard!!
If I may suggest an advice: don't try to educate your parents. It's a waste of time.
correction, sorry:
229w, 420L, 32D= 8=681 lattest.
Very impressive, well-structured post BDK. I've always felt that blitz has to have merits and this post highlights those beautifully. I assume the pool of blitz games in your study is from 100 games? I think you especially struck a chord for me when you say (regarding blitz play) "If you analyze afterwards you will slowly build up a feeling for safe positions, dangerous positions, positions in which crazy fun speculative sacrifices are likely to work, and those in which they are not." For me, this is what blitz has allowed me to realize which I likely wouldn't have speculated or investigated about in slow play.
cratercat: many thanks for the kind words. I analyzed probably more than 100, but the graph is from about 60 games or so.
DK-- I love to see passion on the internets. Though I think FF was trying to be funny but it may not have translated well.
Transformation- I have never addressed a comment to you ever.
You sound kind of unbalanced.
Who cares about your statistics?
I have no idea what you are talking about.
At my level of chess skill and strategic/positional understanding, I can stare at the board for 10 mins but the reality is that’s all I’m doing. I generally have force myself to take my time but then I tend to mull the same 3 (fairly obvious) moves over and over in my head … I have a really hard time getting very deep into a position – mostly because I am unable to visualize, scrub back and forth between, and ultimately evaluate more than one (fairly short) short line in my head. That is not to say that I don't appreciate the theory of the deep think and have no doubt about its long term benefits - which is why I continue to play slow games - however, it just seems beyond my cerebral short term chessic RAM. I get where Dan’s article is coming from but my question is, if I am unable to really perform a deep think, wouldn't I be better served being exposed to many tactical situations in as real a game setting as possible? O.K… I’m gonna go back to my blog to delete comment spam now…
I've been going through my reckless Blitz phase ( Catch me on ICC as Blunderpone).
I like it as a supplement to practicing tactical calculations... I also like Pie.
When doing CT-ART or CTS... you KNOW there is a tactic. doing a 2/12 or 3/8 pie throwing contest has been beneficial for helping me think on my feet. I've been playing the bots as a training... I start with a lower one... and work my way up... when my rating meets within 50 ELO.
I think its helped my OTB game.. but i have to learn to slow back down... dial it down so to speak.. as I do one real slow game a week at teh club.
So, yeah... I've converted.
FF a character disorder is the opposite of a neurotic: instead of taking too much responsibility, it is always someone else that did it.
how old r u young man?
give us your win/loss stats,
otherwise shut the f_ck up. you must be the one playing patzers. not rating, but rating delta. im done with you now.
does GM Seirawan email you for your opinions and offers his, five and six times a day?
anyone here know anyone who fits that bill?
FF:
who cares about my statisitcs? the 2,000 persons who have been to my blog in only 13 monnths, GM Yasser Seirawan, to name but a few of the people who care about or have some interest in my statistics and opinions:
http://www.coloradomasterchess.com/Informant/Ratings%20and%20Expectations.htm
where do you stand according to this article?
i bet you are the one who plays patzers. give us your great delta of + elo to -elo opposition. or you have zero credit here. DONE.
Well in my opinion, blitz does have some redeeming value. Personally, I agree with the idea that raw tactical/strategic skill is raised by playing standard time limit games and not that much by playing blitz.
Here are the uses of blitz as far as I see it:
1. Improving ability to play in time pressure. I used to constantly lose games because I kept playing too slow and getting into time trouble and then making blunders or timing out. Whenever I played blitz, I would constantly make blunders every few moves. Then I made an effort to improve at blitz and not blunder. I raised my blitz rating up signficantly (up to 1400 on FICS, still about 350 points below my standard time limit rating). Now I don't have major issues with time trouble and when I do get in time trouble I play better than I used to.
2. Practicing a new opening. If you feel like trying out a new opening because it looks interesting, it is a good idea to play a dozen blitz games where your opponent plays against this opening, to see how it works out before trying it in a standard time limit game. People also tend to use tricky variations in blitz, so if you play blitz first you can run into some of the sharp systems against your opening and prepare for them.
3. Improving intuition in terms of endgame strategy. I think that in blitz it is a crucial skill to be able to see things like potential passed pawns, bishops of opposite color, whether or not your bishop covers the promotion square of a passed pawn, perpetual checks, traps, and ways of getting compensation when you lose material.
For example, I had a standard time limit game a while ago where I ended up losing a piece to a tactic, but I found a way of getting two pawns for it. Then my opponent was lookng to set up a way to win the exchange (which would put him up a rook) and I let him take it. However, while he was preoccupied with this tactic, I managed to get 2 more pawns and soon got another one. This made me have 5 pawns for his rook (he had like 1 pawn left and it was a rook pawn) and I actually ended up winning this game. Playing blitz games often requires you to quickly find resources for stalling (to win on time), drawing, and getting compensation. I would say that playing blitz helped me get compensation in the game mentioned above.
4. If you're an attacking player, it probably is good for improving intuition in attacking (I'm not an attacking player in general, so I can't vouch for this).
5. Also, eventually after playing chess for a while you get to a point where nothing really surprises you (that is, there are moves you don't see, but you don't have a reaction of "What is THAT!?" when your opponent plays one). I think playing blitz enough helps to get to such a point where you say "oh, they are putting their bishop on that square and their knight on that square, I've seen this setup a million times". In general such a feeling is good because you won't be surprised as often.
Transformation, chill out, it didn't seem to me that he was aiming personal stuff at you. He was just giving his own opinion in general.
thanks Dean.
nice comments Dinomike.
Post a Comment
<< Home