Thursday, January 01, 2009

Safety first!

Fellow patzers: when it is your turn to move, do not waste your time thinking about pawn structure, whether to exchange your Bishop for his Knight, or similar strategic jambalaya before doing a basic safety check.

Safety must be the first and last thing I think about when it is my turn to move. Which pieces are unsafe and are potentially free for the taking? It is all too easy to forget the importance of safety when I get caught up in the beauty of strategy or the drudgery of endgame technique.

I have known this since I was a total beginner in the game, but when I study other topics (e.g., knight outposts) I often forget that safety is the top dog in the priority hierarchy. My description of the situation a year ago is still quite apt:
Obsess about the safety of your material. Don't worry about subtle strategy before looking for obvious code-red tactical opportunities and pitfalls. It is stupid to think about a mosquito on the horizon when there is a gun to your head.
Safety is the wheel of my chess unicycle, and I'm sick of rediscovering it.

The quick safety inspection is also the bare minimum skill required when visualizing the consequences candidate moves. If I make the move, will I be leaving material out there for free?


To maintain my safety vision, I've been getting back to brass tacks, working to maintain and develop my intuitive sense of which pieces are safe on the board. It's a three-pronged training regimen:
1) Safety Blitz
Play blitz games where my only goal is to perform a blundercheck on every move. If I succeed at that, I count it as a win. This gets me into the habit of thinking first and foremost about safety, no questions asked. In slower games I can think about girly things like piece activity.
2) Fritz attack training
Fritz has a training mode where positions are put up and you simply have to click on all the pieces that can be taken. When it takes me too long to click a piece, I stop and examine why that capture was invisible to me. This is very helpful. Some say that missing simple captures is a sign of sloppy thinking. This is often not true. There are patterns to the blunders that reveal deficits in your chess vision.
3) Forks and Skewers
A great free downloadable game that implements de la Maza's forks and skewers drill. As fast as you can, click on the squares that let you fork or skewer the King and another piece.

The above three exercises, done daily, are helping me keep my elementary safety muscles toned. I'd be curious to hear about others' experience with Fritz's attack training mode. I average a score of 13 using a three minute window.


I have written about the topic of safety quite a bit. Making it the highest-priority focus in my games was the key to my initial chess improvement. To increase the odds that I'll remember this lesson, at the end of this post I've appended an annotated set of links to my most memorable posts on the topic. Also, I recommend Heisman's articles on safety, counting, tactics, and all that--he has done more than anyone to emphasize this dimension of patzer chess.

Of course I know that other factors such as combinations, attacks, pawn structure, etc, are important in chess. It is embarrassing to admit it, but I am still crappy enough at chess that I need to actively work to maintain basic safety skills.

The moral of the story is, 'Don't give your shit away for nothing.'

Related posts with excerpts
  • Threats rule in these parts: Don't worry about the mosquito on the horizon when there is a gun pointed at your head!
  • Lessons from blitz: Most tactical possibilities that arise in games between patzers are only one to three moves long. This fact should be impressed into the minds of all beginners.
  • Why start by looking at threats?: Imagine analyzing pawn structure for ten minutes before looking for threats. If it turns out you are about to lose a piece, then you've wasted ten minutes.
  • Develop your blunderstanding: blunders are not random, but follow certain patterns. I need to analyze why I blundered, so in the future I can avoid it. Turn that self-hatred into blunderstanding.
  • My thought process (PDF): Because of the relative importance of mating attacks and material advantages, your primary goal on every move is to keep your own material safe while seizing opportunities to attack the enemy King or kill members of his entourage.

    Blogger likesforests said...

    "Some say that missing simple captures is a sign of sloppy thinking. This is often not true."

    Hmmm... if you miss a hung piece when you're not in time trouble, how could you possibly have looked at all CHECKS and CAPTURES?

    "There are patterns to the blunders that reveal deficits in your chess vision."

    True, but if you make a deliberate choice not to wear a seatbelt and then are killed in a crash...

    ...of course, we could also take this as a lesson to be vigilant when driving on icy roads.

    1/02/2009 01:33:00 PM  
    Blogger tanch said...

    Hello BDK,

    A great post about the numero uno rule in chess: Safety first! :)

    My thoughts process OTB goes like this:

    1. Any forcing moves?
    2. Quick safety scan (hanging pieces etc. )
    3. What are my weak squares, pieces that can be immediately (within the next 2-3 moves) applied pressure on?
    4. Any potential sacs by my opponent?

    After that then do I start thinking about attacks.

    Funnily enough, I'd realised that when I think on my opponent's time, I start to calculate main trunk variations in depth. I also formulate general strategies/ideas in response to the main trunk (using Kotov's method).

    I find that this helps to drastically reduce time wastage and makes more efficient use of my time as a result.


    1/02/2009 02:56:00 PM  
    Blogger Blue Devil Knight said...

    "if you miss a hung piece when you're not in time trouble, how could you possibly have looked at all CHECKS and CAPTURES?"

    My point was that even when you look for all checks and captures, you won't necessarily see them. People are subject to what some call visual illusions or strange blocks to seeing captures that are there even when looking.

    Try the attack training mode in Fritz (mentioned in my post). You will see what I mean, or you are past this stage. If you are past it, you should be able to do the attack training super fast. I average a score of 13 using a 3 minute time.

    1/02/2009 03:26:00 PM  
    Anonymous Anonymous said...

    A safety check is indeed needed for us patzers. Well atleast this patzer does so. I am not aware that i do so, i do it unconsious, it has become a habit (atleast in long games).

    However, sometimes leaving the threat excist because your attack is quicker or as deflextion of a piece that is needed in the defense of your opponent leaving a piece hang can be very usefull for oneself. But offcourse, one has to calculate deeply enough to be absolutely certain that it's indeed helping you and not just throwing a piece.

    That makes it so hard and difficult so that in most casses just defending or playing the attacked piece out of the attack is the most smart thing to do.

    1/02/2009 03:34:00 PM  
    Blogger Blue Devil Knight said...

    Chesstiger: yes piece safety isn't the only thing to consider, which is why I mentioned attacks and such near the end. King safety, if poor enough, trumps everything as the nadir of King safety is mate.

    1/02/2009 03:35:00 PM  
    Blogger likesforests said...

    "My point was that even when you look for all checks and captures, you won't necessarily see them."

    I guess I've never had trouble with this. In a random position, if you ask me to enumerate all checks and captures, and I'm not in time trouble, I have no problem. My only problem has been not looking, or rarely I miss O-O+ or O-O-O+ in the context of a chess puzzle, or I miss a move 2-ply into a combination.

    "Try the attack training mode in Fritz (mentioned in my post). You will see what I mean, or you are past this stage. If you are past it, you should be able to do the attack training super fast. I average a score of 13 using a 3 minute time."

    Sounds like a neat drill, but alas I don't own Fritz. I did a 'find the hung piece' exercise in CM11 and scored 5.5, 5.6, and 6.0 sec time averages with 0 errors. A 5-sec time average would be better meaning I could reliably do so in time trouble, but I think is good enough to stay afloat when there is still time on the clock.

    1/02/2009 04:05:00 PM  
    Blogger Blue Devil Knight said...

    LF: Steve Eddins wrote about it some here. He is rated in the 1600s USCF. It would be interesting to compare score in this as a function of USCF rating.

    1/02/2009 04:58:00 PM  
    Blogger Unknown said...

    Here's a data point for you: I average 46 for 3 minutes and I'm Elo 2100.

    1/02/2009 11:47:00 PM  
    Blogger Blue Devil Knight said...

    Aziridine: no way!!! That's crazy high. That's the highest I've heard after talking to a half dozen people. My highest is 25.

    1/03/2009 01:20:00 AM  
    Blogger wang said...

    Funny you seem to be headed in the opposite direction I am.

    I am not playing it as safe lately. I am tired of playing scared. I have not been taking chances as of late and am getting lifeless positions because of it. I wonder how much is related to your repertoire? Are you still gamitteering with white.

    I only mention it, becuae in every opening I play I'm castled by move 8, so maybe that's why I'm becoming less concerned about safety.

    I understand you didn't mean just king safety but when my king is castled the edge comes off for me ALOT. I just get a bit more relaxed and I actually start playing chess.

    The first time I worked through Chess Tactics for the Tournament player I became a much more careful player. It opened my eyes to what I should be doing in order not to keep dropping pieces to combinations.

    Now, several years after my first go around with this book, I am seeing how to make my play more forcing.

    Perhaps all of this is just the ebb and flow of our own chess progression.

    1/03/2009 03:32:00 AM  
    Blogger Blue Devil Knight said...

    Wang: I just mean don't blunder. I don't mean play passively or defensively. In a gambit or sacrifice you give away material because you mean to, not because you screwed up. That's the key difference.

    1/03/2009 10:51:00 AM  
    Blogger Unknown said...

    I don't know anyone else who uses attack training so I don't know if my score means anything. In fact, I'm not totally sure how the scoring works but I figure much higher scores should be possible. My guess is that your score equals how many squares you click divided by the number of minutes you spend. In that case a pace of one click per second doesn't seem unreasonable, and that would be a score of 60. Of course I average nowhere close to that (although my high score is 78) but I wouldn't be surprised if grandmasters could.

    1/04/2009 12:15:00 AM  
    Blogger Blue Devil Knight said...

    Aziridine: yes, I think it divides the total by the number of minutes. I have asked around Chessbase, and even the guy that wrote the article about it (Steve Young or something), and no reply from CB, and he didn't know.

    1/04/2009 02:09:00 AM  
    Blogger Blue Devil Knight said...

    I bet Anand could hit 100. :)

    1/04/2009 02:10:00 AM  
    Blogger katar said...

    this safety issue is handled if you spend a brief moment after each move by your opponent to identify his/her biggest threats if s/he were to move AGAIN. safety and threat detection are synonymous. so i dont think you need a separate "safety radar" if you are visualizing threats first and foremost, as you describe in your "Start by Looking At Threats" post. somehow chess seems more complicated to me when i visit this blog, perhaps because you seem to take a verbal/scientific rather than visual/intuitive approach. just compare the number of diagrams with colored squares and arrows here compared to say, Temposchlucker or whatever. i long ago gave up on reading Tempo's blog b/c he is either too hard to follow or off the deep end sometimes. i think it's your great skill at writing and your verbal/"scientific" approach to chess that make this blog one of the most popular of all chess blogs on the web. but i wonder if it hinders your board vision and corresponding visual pattern recognition. like learning to drive, the translation of visual to verbal to visual is clunky... the goal while driving is to react intuitively to visual stimuli without even engaging the logical faculties at all. just food for thought, and only my opinion which of course could be wrong, especially as i am just an average player.....

    1/05/2009 01:43:00 PM  
    Blogger likesforests said...

    You have me interested now. I ordered a copy of Fritz 10 for $18. Will let you know how it goes.

    1/05/2009 01:54:00 PM  
    Blogger Blue Devil Knight said...

    Katar: My PBS style serves me well in my job but not as much in chess. My next Rowson summary (finished a draft, will polish it up the next couple of days before posting) discusses exactly the contrast you mention between word-based explanations (what he would call pseudo-explanations), which is not the medium of actual chess thinking (during a game) versus imagery and visualization, the medium of productive chess thought during games.

    OTOH, this is a blog and how else am I to communicate but with words? During a game, I spend probably 2/3 of my thinking time in image-based thought, and about 1/3 telling stories about what I want to do and planning (e.g., my Knight is awful, I really need to activate him).

    But definitely some proportion of my blunders comes from getting caught up in my own stories about the game and forgetting to really look at and think about the position in front of me.

    You are right that safety is just one subset of threats.

    PS You're a damned hippy.

    1/05/2009 03:55:00 PM  
    Blogger Blue Devil Knight said...

    Likeforests: awesome! I predict somewhere between average score of 15 and 50. :)

    1/05/2009 03:56:00 PM  
    Blogger Diamondback said...

    The question is if one hangs a piece towards the endgame is that sloppy chess or is one just burnt out by playing a six hour game ?

    I believe the majority of Class C and below player who hang a piece may be focused on the overall basic strategy of the game to win a won game and not on the safety of piece placement.

    1/06/2009 04:32:00 PM  
    Blogger likesforests said...

    My Fritz X arrived! For Attack training I scored Avg:23 High:32.

    Since Check & Attack Training help you spot checks & captures faster I could see working at this improving the SPEED at which one does their per-move blunder-check, freeing up time for other ideas.

    I like that they use a database of real games. Maybe I will use a DB of my own games or master games in my openings--I bet that would make training more effective.

    But I still say anyone who looks for them consistently--with as low a speed as 6--should rarely hang a piece or miss a hung piece. ;)

    1/07/2009 09:11:00 AM  
    Blogger Blue Devil Knight said...

    Diamondback: that is an excellent point. When I am tired I overlook things.

    LF: that is really good. I sometimes sit there a minute trying to find what I've missed. It is very helpful as it shows major oversights in my elementary board vision. In such cases, in a real game, I would likely overlook the capture. Indeed, yesterday I missed a queen moving sideways that could capture a Knight while playing attack training. This is the exact blunder I made in a tournament last month.

    I'm trying to get where I consistently score over 20. When I'm really tired I score much worse, which is good to know.

    Of course the attack training game is different from a real game of chess, where you typically keep track of these safety issues as the develop during the game. Hence, in real games it tends to be much easier.

    1/07/2009 10:33:00 AM  
    Blogger katar said...

    brilliant writing and content as usual.. your chess blogging rating hovers around 2600. the only one i can think of that's higher is Kenilworthian at 2750, and maybe Drunknite at 2650.

    1/08/2009 05:46:00 PM  
    Blogger Blue Devil Knight said...

    Thanks Katar but my hunch is you meant that for the next post on Rowson?

    1/09/2009 09:48:00 AM  
    Blogger katar said...


    well, your writing is consistently brilliant but that comment was specifically intended for the subsequent Rowson post. ahh, this bloggery is so confusing.

    1/09/2009 06:39:00 PM  
    Blogger yi Zhang said...

    likesforest: I find your comment to
    "Some say that missing simple captures is a sign of sloppy thinking. This is often not true." to be somewhat ignorant because learning to not hang a piece is not just a matter of concentration, its a matter of practice and repetition.

    your comment to
    "There are patterns to the blunders that reveal deficits in your chess vision." also is nonsensical. After all the mistakes you make define your weaknesses. we don't make deliberate choices to blunder, it happens because a) we need to work on it more or b) lets say your an A player and you hang a piece, then fine your interpretation would be correct.

    1/11/2009 12:05:00 AM  

    Post a Comment

    << Home