Sunday, July 28, 2013

my favorite lessons 6: bishop takes knight?

Kids love knights. I find myself reminding students 10,000 times a year that bishops are usually better than knights, and they need to have a good reason to trade a bishop for a knight. To overcome their natural tendency to favor the knight, I like to give a lesson directly on the subject.

So what is a good reason to trade bishop for knight? Most of the time, you should either be
1) winning a pawn,
2) opening up a king (in a situation where it's realistic to attack) or
3) creating a real pawn weakness that you can attack.
It's also ok to trade if your bishop is bad and you have no good retreat square for it.

Of course, there are some situations that don't fall into these categories, but I make it clear that students are responsible to me for having a good reason.

I explain to the class that I'm going to give them 8 positions and they have to tell me if white should trade the bishop for the knight, or not. Notice that in most (5/8) positions, the answer is simply no, you should not take the knight.

Here they are: (or here if you'd prefer a cbv file)

1. No

2. No

3. No

4. yes, since white can follow up by taking on e5

5. yes, because white creates a serious weakness: doubled isolated pawns on a half open file. Notice that White needs to continue correctly: 1. Bxc6 bxc6 2. Na4! (otherwise black will get rid of his weakness by playing 2....c5), followed by Rac1 and either Qc2 or doubling rooks on the c file.

6. no


7. Yes, because it opens up black's king. White can follow up by castling queenside or playing Qd2 to attack the weak h6 pawn. Point out that after 1. Bxf6 gxf6 2. Qd2 Kg7 3. 0-0-0, white doesn't have to be afraid of 3...Bxf3, since opening the g- file only makes black's king less safe.
8. No. What white would actually like to do here is play Bg5-h4-g3 and try to trade off black's excellent dark squared bishop.

Thursday, July 25, 2013

my favorite lessons, part 5: attacking and practice in calculating

This position is stolen from Coakley's green book, the chapter "Castles Made of Sand." I use it to talk about how to attack.
    I explain that you normally need one of two things to have a successful attack: either open lines toward the opponent's king, which sometimes happens because you've moved your pawns, and sometimes happens because they've moved their pawns, or more of your pieces attacking the king than enemy pieces defending it. I explain that some players like to attack the king in any position, but it is only a good idea if you have one of these two advantages. This means that if you really want to attack, you have to start by either bringing pieces over towards the kingside, or pushing your pawns to open lines, or somehow getting your opponent to move the pawns in front of their king.
    I ask the class first to brainstorm possible first moves, and ask them to start with forcing moves: moves that are either
1. checks,
2. captures, or
3. threats of checkmate.
After they list all they can (usually 1... Rxh2, 1....Qh5, 1....Qe5/d6, 1...Nxf2), I ask also for any moves that bring more pieces towards the black king. Usually I get two answers to this: 1... Rdg8 and 1...Nf4. This gives us an opportunity to talk about how the former is more effective, as the knight is already participating in the attack, hitting f2 and f4, and preventing the queen from moving along the third rank to defend.

There are a number of wins in this position that you can explore with your students:
  • 1...Rxh2 2. Kxh2 Qh5+ 3.Kg1 Rh8 4. any Qh1/2
  • 1...Rdg8 2.Bxd3 Rxg2+ 3.Kxg2 Qg5+ 4.Kh1 Qh5;
  • 1...Qh5 2.h3 Rdg8 3.Kh2 (3.Bxf7 Qxh3) 3...Rxg2+ 4.Kxg2 Qxh3+ 5.Kg1 Qh2# (5...Rg8#) ;
  • 1...Qe5 2.h3 Rxh3 3.gxh3 Rg8+ 4.Kh1 Qf5 5.Kh2 Qf4+ 6.Kh1 Qf3+ 7.Kh2 Qg2#
    • 2.g3 Qh5 3.h4 Qxh4 4.gxh4 Rdg8+ 5.Kh2 Rxh4#;
    • 2.f4 Qd4+ 3.Kh1 Rxh2+ 4.Kxh2 Rh8+ 5.Kg3 Qg7#)

Wednesday, July 24, 2013

my favorite lessons 4: winning up a pawn

        I stole this from an endgame book, I'm sorry I don't remember which one. The idea is to give students a roadmap for how to win a simple minor piece endgame when they're up a pawn. Obviously, this is not so simple, and so your objective is not that they win 100% of the time, or even 75% or any particular % of the time, but more that they have an idea of the method and get some practice at it.
      I start by showing them a position
having a student tell me the material, and asking who thinks they could win this as white. (it doesn't matter what the answer is). I then ask who can explain to me what the plan is.
The plan is this:
1. Centralize the king (I explain that you activate pieces in the endgame in the order of their power, i.e. queen first, then rook, the king is worth 4, so king next, then bishops and knights, and generally only after these pieces are activated do you start pushing the pawns.)
2. Activate the knight
3. Make a passed pawn by pushing the pawns on the side you have a majority.
4. Once you've done that, the side with the extra pawn usually wins by some combination of:
   a) pushing the passed pawn and invading with the king
   b) trading knights
   c) sacrificing the passed pawn to win the kingside pawns.
Make sure this is written on the board so students can refer to it later while they are playing.
I then ask for a volunteer to start white out by doing #1, centralizing the king. I move for black, and we play through the following moves:
1.Kf1 Ke7
2.Ke2 Kd6
3.Kd3 Kc5
I then ask for another volunteer to take over and activate the knight:
4.Nc2 Nd5
I ask what this threatens (Nf4+ winning a pawn) and how white can stop this:
5.g3 a5
Here I ask which pawn to push first, and if they don't know, remind them of the general rule that you push the potential passed pawn first, in this case the b pawn:
6.b3 f5
7.a3 g6

8.b4+ axb4
At this point, you've completed step three, and I explain that you now try to advanced the pawn and be on the lookout for tactics that allow you to sneak in with your pieces, or trade knights. In general, you calculate as much as you can. Depending on the level of the class, I go faster or slower through the rest of the game: the exact moves don't matter as much as the kids grasping the basic plan in the beginning. You won't be able to teach technique and endgame control in one lecture-lesson, so don't try too hard.
9....Kd6 [9...Nxb4+ This is a nice example of how white wins fairly easily if black allows the knights to be traded. 10.Nxb4 Kxb4 11.Kd4 Kb3 12.f4 Kc2 13.Ke5 Kd3 14.Kf6 Ke3 15.Kg7 Kf3 16.Kxh7 Kg2 17.Kxg6 Kxh2 18.Kxf5 Kxg3 19.Kg5]
10.Kd4 Nc7
11.f4 Nb5+
12.Kc4 Nc7
13.Ne3 [also good is 13.b5 Nxb5 14.Kxb5 Kd5 15.Ne1 Ke4]
14.Kd4 Kd6
15.Nc4+ Kc6 [15...Ke6 16.Kc5 (16.Ne5 Kd6 17.Nf7+ Ke7 18.Ng5 h6 19.Nf3 Kf6 20.Kc5) ]
16.Ke5 Kb5
17.Ne3 Na6 [17...Kxb4 18.Nd5+]
18.Nd5 Kc4
19.Nf6 h5
20.Nd5 Nb8
21.Ne7 Kxb4
At this point, I reset the position and ask a student to repeat the general plan. Then students choose a partner, set up the position on their own boards, and practice playing the position as white and as black. Ideally, they should play twice, once with each color, and should have 10-15 minutes per side, although you can do it with 5 minutes each if you are pressed for time. Do remind them that playing an endgame with 5 minutes is not at all the same as playing a whole game with 5 minutes, and they should play slowly and thoughtfully as the position is tricky.  
the next day....
I follow that lesson with its sister position:
which you will notice is exactly the same, but with bishops instead of knights. I ask students again how many think they would win the position, and hopefully a few more students raise their hands than last time.
     I then ask what the basic plan is, and of course its essentially the same:
1. Centralize the king
2. Activate the bishop
3. Make a passed pawn by pushing the pawns on the side you have a majority.
4. Once you've done that, the side with the extra pawn usually wins by some combination of:
   a) pushing the passed pawn and invading with the king
   b) trading bishops
   c) sacrificing the passed pawn to win the kingside pawns.
I again show students a model game; you can also have them play first and show them the game afterwards, but I find with difficult lessons like this, many classes benefit from as much teacher-modeling as possible before they do it themselves. They play much better and are more likely to be successful if they see exactly how you do it first.
I ask for a volunteer to help me do step 1:
1.Kf1 Kf8
2.Ke2 Ke7
3.Kd3 Kd7
4.Kc4 Kc6

then a new volunteer for step 2:
5.Bc3 g6

and again a different student for step 3:
6.b4 Bb6 7.f3 Bc7 8.a4 Bb6
9.Bd4 Bc7
10.b5+ axb5+
11.axb5+ Kb7

and again, don't get too worried about covering every detail of the rest: like every endgame it gets a little messy and there are many possibilities for each side. What's below are just examples!

12.Kd5 Bb8 [12...Bf4 13.Be5 Be3 14.Kd6 Kb6 15.Ke7]
13.Bf2 [also good are 13.b6 Bg3; and 13.Be5 Ba7 14.Kd6 Bb8+ 15.Kd5 Ba7, which at first looks like repetition, but white invades after 16. Bg7 h5 17. Ke5]
14.g3 h5
15.h4 Bb8
16.b6 Kc8
17.Kc6 Be5
18.b7+ Kb8
19.f4 Bf6
20.Ba7+ Kxa7
21.Kc7 Bd8+

Now again, return to the original position, have a student repeat the basic plan, and send the class off to practice. Circulate and watch: the most important thing is to catch players who aren't following the basic plan. Don't worry too much about showing kids every forced win that they miss: it's a difficult position and you don't want to undermine their confidence. Keep in mind that your goal here is to give students a basic plan to follow, not to police their endgame technique.

Tuesday, July 2, 2013

favorite lessons part 3: threats

       The idea and first position for this lesson is taken directly from Jeff Coakley's excellent book "Winning Chess Strategy for Kids." Most of my favorite lessons are stolen from Coakley's books, and let me say now that if you're a chess teacher and you don't have all his books, you should stop reading this right now and order them. They're all you'll ever need, I promise.
    That said, I start the lesson (as he suggests) with a general discussion of what a threat is: how in real life, a threat is bad ("I'm going to beat you up after school," "I'm going to tell," etc.) but in chess, a threat is great ("I'm going to take your piece") because it gives you a chance to be winning next move. The more threats you make, the more chances you give your opponent to make a mistake, and the more chances you will get an advantage.
      In chess, a threat has to be specific, so when I ask "what's the threat?" I am really asking "Where are going going to move next turn?" and you should give me a specific answer, like Qxg7, rather than a vague answer, like "checkmate." For a threat to work in either real life or chess, it has to be something that the other guy is actually scared of. So if I say "I'm going to give you a piece of cake," that isn't a threat, and neither is threatening to play QxP if they can just recapture your queen.
    Here's Coakley's position:

    He talks about the following threats:
  • 1. Be3, threatening to take the black queen.
  • 1. Bd6, threatening to win the exchange
  • 1. Qg2, threatening Qxg7#
  • 1. Qd2, threatening the sacrifice 2. Bxh6
and then more complex threats like
  • 1. Be5, threatening to double black's pawns
  • 1. Qe3, threatening to trade queens, since white is up material 
Obviously, you shouldn't just tell the kids this, you should ask them to find the threats. I like to show this position as an example, and then do a couple more, usually one opening position and one endgame. For example:

Threats include:
  • 1. Bg5, threatening the queen;
  • 1. Qa4, threatening the knight for a second time (a good opportunity to review counting attackers and defenders);
  • 1. d4, threatening both to win the e5 pawn, and to play 2. d5, threatening (also winning) the knight (a good opportunity to review pins)
  • 1. Ng5, threatening to take on f7 with the queen or knight (ask which threat is more dangerous). This can lead to an interesting discussion about how to follow up after 1...Nh6 or 1...Qd7. (2. f4 is a logical idea, as are 2. a5 and 2. Bc4)
  • 1. a5 threatening both 2. a6, winning the knight by attacking the bishop, and to a lesser extent 2. axb6, threatening to make black's queenside pawns into targets.
It's also good to talk here about how it's trickier to make less obvious threats, i.e. everyone will see that 1. Bg5 threatens the queen, but 1. d4 and 1. a5 are harder.

Threats include
  • 1. Rhd1, threatening 2. Rd8 with backrank mate
  • 1. h4, threatening to trap the bishop with 2. h5
  • 1. Nd5, threatening a fork with 2. Ne7+
  • 1. Rd7, threatening to take on b7.
Tell students that in their games, they need to try to make as many threats as possible, and to show you when they make a good one. Write down the position in the best student example and use it as a review at the very end of class or the beginning of the following one.