Skirting Crafty’s weak spots

I lost this blitz game on FICS, but sense that I missed a better opportunity, so here is my analysis game for getting an advantage.

It takes Crafty over a minute to spot 18.Qd3 for White, and before that it thinks Black is winning. A couple minutes after finding Qd3, it realizes the position is even, but at least one minute after before it thinks Black’s advantage is only tiny.

A human would probably choose White right away.

I had to have the guts to tell my chess engine that it was wrong, that White has nice compensation for the pawn, even though this would have been impracticable for me to calculate over the board in a G/90 setting. I didn’t find the continuation right away, but liked the c4 (mine) and Qe2 idea. For years, I would lose such a position to a computer because I cannot calculate like it, lots of ways for White to go wrong, but this time, with Crafty’s help also, I was able to find it. Can you imagine if one us actually beat Crafty by deeper tactics(?). Believe it or not, this is possible. I think it’s the consistency and aggressive play, particularly when challenging human foibles, rather than other computers, that gives an engine such a high possible rating.

Like I say, this is a quad-core, 2.4 Ghz phantom AMD machine (ASUS MB), Dual RAM 2GB in parallel, but acting as a fast 1GB RAM (633 Mhz FSB, plus 2x throughput).

This is why I always look at computer continuations skeptically at first when I read them on a blog. If anyone puts this position up against Fritz or Rybka, let me know how well they performed on this position. 🙂

I would imagine that there are other programs with a more human-like heuristic in their algorithms that don’t go by calculation alone but more by “the look” on the board, as a human would do. They may get it right, right away.

I noticed when looking at GM games on CA that even that engine would not spot the tactics that the GMs did, for a minute or so, mating attack type of attacks. It’s more surprising though, when you or I find these. Crafty does go in over it’s head frequently, though.

I’ll go so far as to say that someone like Capablanca could probably beat a computer that plays this way easily enough. For Capa, he may have looked 12 moves deep, but may have only looked at “the best one” on each move, as he liked to say, so that is only 12 moves that he has to hold in his head, not a million or more, as with a computer.


Move selection process

I am going to detail mine here, in order to be more efficient at it, and to make corrections to it, which I did while writing this.

Part I – 45 seconds
Does my opponent’s last move threaten anything?
What was the purpose of his/her move?
Where is their attack?

Part II – 30 seconds
Initial scan – are their any soft targets or immediate tactics (such as employing a pin)? For/against both players.

Part III – Bulk of time, depends if you are simply carrying out a previous plan or devising another/additional plan.
Enumerate my positional plans.
How would the first plan generally develop?
How would the second plan generally develop?
Determine the optimal move order to prevent counter-threats, if necessary. Blunder-check happens here.
Choose a plan.

Part IV – 30 seconds
Did I overlook any immediate forcing moves, could they be part of a plan? If yes, go back to part III, abbreviated (last 2 steps).

Part V – 15 seconds
Visualize the move, look for any sacrifices or trades that pull pieces out of place/defense.
Make the move on the board and punch clock.

Part III is the only place that I can save time because, otherwise, 2 minutes surprisingly seems like a hard number except in time-trouble (which is mostly a crap-shoot). In time-trouble most thinking is done on opponents time, and is concrete, mostly aimed at counter-strategy.

The best way to get around this 2 minute minimum is to do a lot of the tactical scans on the opponents time, and to enumerate plans during opponents move already. Anticipate opponents move and work out own plans then; this is really the only way around this 2-minute rule.


I don’t “really” need to be lectured on the clock after this one since it’s the only reason that I lost, but you can if you want.

I got there 13 minutes late, he was 20 minutes late, but we both had 80 minutes to start the game, he a couple minutes less.

I got into bad time-trouble for one reason, I didn’t want to play moves I figured would just about lose, but after spending lots of time I felt forced to play them anyway to avoid losing on time.

He exchanged too quickly as I pointed after to him repeatedly after the game (he kept wanting to look at it), and explained that he should move his king one or two squares first even though we played it out a bunch of times first where I showed him he was losing.

Anyway, in the game it was Q and Pawn, I had an extra pawn but 1:23 left on my clock. I knew I would have to make a move anyway before offering a draw (so it really doesn’t get one out of a pickle). I dropped the pawn back because I was too frazzled still and couldn’t decide on move the king (loses a pawn) or e4 or Q moves, which all looked okay except for king moves.

Then I offered him the draw, he refused, whereupon my clock went blank (battery died) and we got another clock and the TD knew it was his 3:23 left to my 1:08. We unexpectedly had time some time right then, and I showed him it’s an easy hold. He refused, and then I said something like “you are totally going to take a crap on me because of the clock, I can’t believe this”. Sure enough, that is what he did.

I believe there was 3-fold repetition 2 or 3 times but I did not make a claim. He did play f3 at one point, so I honorably thought he may have been trying to improve his position and did not make a claim. He never went ahead with any attack although it looked like he was trying to at first.

I looked over to my clock and it said 2 seconds, then I reached to press it and it said zero seconds, I never saw it say 1 second, must have blinked, even though I was looking right at it (it honestly felt like a time-warp had just occurred, and this must be destiny for him). Anyhow, he had 45 seconds at this point (may as well have been 5 hours, relatively speaking). Anyhow, we play on for another 10 moves, him not noticing my zero seconds. I only spent about 2 seconds on each of those moves because he was simply checking me back and forth, even did 3-fold, and then moved his queen one square over, and then repeated this 3-fold checking scheme. I could not believe it. Then he noticed at 23 seconds that I had none and immediately called time.

So I walked away feeling like I had lost because of this game of “chicken” on the clock, and then I end up giving him a post-mortem endgame lesson making him a better adversary for next time. I lose for non-chess reason and then probably he is stronger now that I told him how to win/handle that endgame. Bizarre.

First thought on my mind after the game was try never again to finish a game with <5 minutes on the clock or it will simply turn into a game of "who has more time left." Seriously, why take the draw, even if losing, compared to winning on time. It felt like a FICS game, but at least on FICS I expected that behavior. I knew the writing was on the wall with about him having 9 1/2 minutes left he just started moving and not writing the moves down til later, but I think that that may have been the last move he wrote down anyway, so he was blitzing on my time-trouble and was obviously studying the clock.

The game went on maybe 25 moves longer than I am showing, and he did not even push f3 for a while. He pushed h5, traded pawns on g6, tried to get his king involved but I checked him from h1 with my queen three times. And it was only after all this that I lost on time and he started simply checking the crap out of my king rather mindlessly while never pausing to look at the clocks. I mean, he knew that I was quite interested in a draw, so it wasn’t even about his wanting to simply survive on time that he failed to notice the clocks.

We both missed that he had 31. a5! winning, but I had thought I had been forced to play into his win for quite some anyhow. Funny that he wasn’t able to find the win even though spending quite a few minutes on it around that point. After the game he said “I should have played BxN (before I played Be7)”. I said that “Yes, it would have been stronger so why didn’t you?” and then I continued “I’ve only played it about 25 times in the past anyway.” It made me realized how just about everybody is screwing their game up, in some way, in order to play to the clock. It’s not a totally natural time-control for OTB play.

Here is the game.

Next game I am just going to play for a time control. I’ll play 20 moves to have 30 left, then to move 30 to have 20 minutes left (massive overkill for me), then play it out.

Playing quickly as Black is important in G/90. White can push Black around because White has the initiative, but Black cannot push White around, except on the clock. It’s also important to look for tactical opportunities quickly as Black because White can go for a slow, boring positional win and it’s up to Black to confuse the issue. It’s also a double-whammy playing Black because the higher-rated player usually gets White, so it’s more important to not bother with a correct game too much, other than having it work somehow in your own mind, because you have to make it snappy on the clock. As White, you are usually getting the lower-rated player anyhow so can usually get away with screwing off on the clock more.

90-move draw

This isn’t one of my better games, and I noticed that I had blundered the exchange right away, but this shows why technique is still so important. All of our previous discussions on knight against rook is a draw stuff. 😉

In the game, I was setting up for a cheapo, which I knew he could thwart. If he hadn’t played Qg4 when he did, I was looking at Nd5 sham-sac for the possible win.

BTW, the time control was luckily 10/15 or 10min+15sec increment, which luckily allows for this sort of position to be played out – which is rather unusual for my average game on FICS. 🙂 Incidentally, we both had 3 1/2 minutes left at the end of the game, and on a couple moves I did let the clock go to under 10 seconds. I guess what I am saying is that it doesn’t take an eternity to play this type of position out, it is doable, but it does take some time. So 60 moves of technique at 15 seconds per move is actually 15 minutes of time!

In contrast, before this game I played a guy around 1897 at 15 0. At 3:35 to his 12:00 minutes, I realized I was going to lose on time, cut my analysis short, and immediately blundered a piece. Had I examined the most forcing moves instead of a fantasy attack which wouldn’t work (but I didn’t play anyways, simply made “a move”), I was actually up 1/2 a point by Crafty’s estimate by simply playing NxN followed by Be4, trading his fianchettoed King’s bishop (and you know how much they hate to see that happen). so I had “solved” his English opening, which I and probably most players as Black rarely face on FICS, but lost the battle on the clock. This is the “power” of these awkward openings is that it strains you on the clock and is unfamiliar, it’s not that they are super-hot chessplayers always, IMHO.

Incidentally, I watched two Experts at blitz play a 3/0 game before this and it was atrocious, IMHO, on many separate moves (English opening, again). I’m not as impressed with blitz ratings as I was before watching that.

Played a provisionally-rated player

After only two previous games his rating is 1480, but he’s played thousands of blitz games online and such. It was fun talking to him after the game.

The crazy thing is that I saw 7 moves deep accurately that I was winning after 12.e5 Ne4, but we both missed that after …Nf8 13.Bc5 Nc6e7 prevents the loss of the exchange. We both played quickly on this move, but me moreso. The irony is that Bc5 was still Crafty’s choice, in order to pull the N at c6 from defending the a7 pawn.

I did get there late (although now I realize that the parking lot attendant will be there even after 6pm), so I lost maybe 12 minutes off my clock, and had 7 minutes left at at the end of the game, so it was an incremental improvement over the last game. By the end of the game, I needed to speed it up, so I was no longer making best moves like finding Nc7, even though I had seen the combination that it is used for.

As bad as it may be that I frittered away time in the wrong places, it could just as well be said that he didn’t spend enough time analyzing, which he acknowledged after the game, for he still had around 65 minutes left the time he played 17…Qxb. So, there’s the opposite effect to mine.

Before this game, I was worried that I was “chessed out” and would blunder, but I actually played very slow in the beginning and then it began to come to me.

We analyzed after the game and told him if he had played …Qb6, that I would have played NxNc6 QxN, then we saw that Bf3 kicking the queen was pleasant for White, and I told him that if fxg, that he should recapture with the h-pawn and then we looked at tactics. Which realistically reminds me that I cannot count on getting a winning advantage out of the opening. Sooner or later it’s going to be almost even leaving the opening, and I will have to stick with a sustained attack, merely trying to outcalculate my opponent on move-by-move basis for as long as possible.

Well, I made it back to 1800 on FICS, which isn’t too difficult for me these days; at least it is a cold comfort to leave it at 1800 there.

Bringing down an open-Sicilian

Is one of the greatest satisfaction for me in chess.

I only wished that my games OTB went like this, but alas people will put more effort into those ones.

This was a 15/0 game on FICS. I had 8:32 left and he/she had 11:19 left.

Since I’ve got here, it seems my only games as White were two Owen’s Def against 1400 level, which I think I can handle now, and two Open Sicilians as White against high 1700 level – one Najdorf and one Dragon; talk about catching no breaks. Then as Black I see a lot of “come and get me openings” from White, those are the hardest to play positionally.

That last game against the Bird, I found out that ..d4 with ..f6 and ..e5 to follow, even if I had played it right away, was too slow to gain more than equality (although equal games can be won/lost). One has to play patiently against those systems. Really, there is more that could be calculated in those “slow” positional games than there are even with open games and quick fireworks.

Pawn structures

Here are a couple of interesting games that I played recently that revolve around the pawn structure. I noticed Tommy was looking at the Alekhine, so here that one first:
In the second game, I have to admit that 14.Na4 was hardly necessary as I had dxe followed by Bxa7 winning a pawn, but it made the game more interesting. 😉

My plan now is actually to stop playing on FICS and study openings or games or endings instead. I have the “Big Book of Busts” and just studied Owen’s Defense, so I’m ready to take that down next time I face it, although I really studied simply out of interest.

I have MCO13 and Chess Assistant, but it doesn’t do me much good. MCO is a little too, “Okay, here’s how crazy Joe handled this opening in a correspondence game, is down material but is unclear.” Well, that is a fun way to learn tactics, but studying that is almost of negative value to my rating. The super GM’s games on CA, the opening book part, remind me of lines where people are simply trying to deviate from the main-line and mix it up, really not helpful on the class level.

At our level, I think a concise explanation of an opening from a Master is best. For example, Soltis book on opening analysis for Advanced player is concise and hits the major themes of what can happen (not just boring stuff of “okay here is the main-line – yawn”). I actually gave away Alburt’s book on the Alekhine many years ago because I found it so boring, the main lines with sideline games in the notes(with not much explanation). Big Book of Busts is rather incisive as well.

But here’s my point. I used to go over this stuff and was not tactically adept enough to figure out why +- was +-, but I can figure that stuff out now. I spent a long time with Crafty going over my games afterwards, even blitz games (yes, TONS of blunders no one notices in blitz games, as you might expect). Anyhow, at G/90, if I really want to have fun winning vs. just playing and losing, I really have to look at some of these openings because there is not time to figure them out OTB, RollingPawns is right about that. At G/2hrs. at my old club, there was time and it was fun, but at G/90 I simply don’t see such a possibility, or I would have to play it quickly in any event.

Incidentally, in the second game Black stopped me from playing 24.Qc1, a useful device which is known as “Alekhine’s Gun”. Sierawan once wrote a book ‘Take my Rooks’ that covered games using this technique, but it also had something to do with a ‘double-rook sacrifice’.