By the way, if anyone doesn't religiously follow XKCD like I do, you might have missed this from the other day:



(Large size)

Awesome graphic BJJ, gracias.

Didn't know Vera Menchik died in a missile attack.

So, my new tactics training regime consists of a starting off solving a custom set on chesstempo.com of "Named Mates" as a warm up (5 or 10) and then work on Standard problems to develop calculation skills. I usually give up on a problem somewhere around 7 - 10 minutes, but can go longer if I think I "see" something.

I think I had to do something to it to remove mate threats and only get mates. I didn't put a restriction on the rating of the problems so most are below 1500, and I solve most of them within 20 seconds, but some are very beautiful and it's a good warm up. I think it is has helped me as well because a lot of problems contain "mate threats" which are these typical patterns.

I'm up to about 1700 standard, which isn't very good, but it is what it is. I hover around 1510 on blitz mode. I feel like tactics is a huge leak of mine as no one who has a rating of 1800+ or 2000+ has a tactics rating less than 2000 so I know this is a limit I have to get through to really improve. I had taken a long break from chesstempo to work on Chess Exercises for Kids, but I think this mixed regime is good and work on Chess Exercises for Kids when I don't have an internet connection. It's an excellent puzzle book though, especially the positional and endgame puzzles.

I really don't like tactics trainer at chess.com.

Winning Chess Exercises for Kids is my warm-up / offline book! I love it.

You might want to warm up with a more diverse set of patterns than just mates. The easy material-gain combinations from WCE are ideal, imo. Like this one:



Can you identify any commonalities across (many of) your failed attempts? For example, "ruling out variations too hastily when they superficially appear bad" or "not noticing aggressive pawn moves" or "ignoring 'obvious' moves at first because the solution to a 'difficult' tactic can't be obvious".

I don't think the combinations in WCE are always easy - some of them stump me... of course some of them are easy... most are quite beautiful like the one you posted.

I plan to expand my warmup. I created the named mates because it sounded fun. I found out that it was really fun, but then I noticed that I was seeing these patterns in my more difficult problems as well and they are quite frequent. Often the threat of an Anastasia causes loss of material for example. Or a Morphy or a Boden. Also, a lot of them are really beautiful. I think you would enjoy that set.

Yes. This is a big leak.

Yes. This is a huge leak. This is a real leak in my tournament games too.

I usually look at the obvious... then assume it can't be that obvious. This is a smaller leak then the other two.

I'm curious, what is your chesstempo tactics rating? I'm guessing it's around 2000.

Without doing the research I'm guessing I have "quiet move" leaks, advanced pawn leaks, zwishenzug leaks, and I've even missed some skewers recently on some 1650 rated problems, which is embarrasing.

My solution to the problem:

Yeah, quite a few are toughies. Those are a lot harder than they would be elsewhere. I get primed to expect easy problems, and for some reason, I often try to finish the nine-problem sets in 5-10 minutes, which obviously won't work when a difficult problem pops up.

I'm not sure if this one has a quick fix, but calculating more slowly helps (and usually ends up saving time), and it's useful to cultivate a skeptical attitude towards the apparent failure of promising ideas. More generally,
it seems like a significant % of "strong moves" are positional moves that would be clearly best if safe, look superficially terrible (so much so that they are never considered), but turn out to work for tactical reasons anyway. This is a very simple example, which I'm recycling from another thread:





The maneuver is just so much more desirable than anything else that it's fair to organize our options in a very binary form: "get a knight to e6 while it's possible (1.-Bf6 is coming, after all), or do something else."

So it could be correct to devote almost disproportionate resources to finding a way to make it work. Well not 1. Nd4, which is super easy to verify (the hardest part is just considering it), but these moves in general. I say "could" because there is always time trouble.

A lot of combinations present similar challenges. I have a policy of never accepting knee-jerk "no this line obviously won't work" responses from myself. Even 10-15 seconds of further examination can reveal a promising idea.

Also, when you calculate variations do you mostly visualize the pieces moving around w/o any sort of inner monologue, or do you try to find the correct "idea" before searching for a move?

A couple of examples of what I mean by "ideas":

http://forumserver.twoplustwo.com/sh...postcount=4244

(Format: cues -- "ideas" -- followed by the moves in one variation.)

"Getting the bishop safely onto the a2-g8 diagonal clearly would be great. We have to block the queen's influence on c4." 1. Rc5.

(After some deliberation w/r/t the captures on c5.) "Black needs to try to stop or neutralize the effect of Bxc4+ somehow, maybe by allowing his bishop to interpose the check." 1.-Qb6 and 1.-Qd6.

The rest was almost purely visual, since Black's responses were so forced.

The problem in the spoiler required a lot more verbal thinking.

It seems like getting better at narrowing down the search space with verbal thinking has helped me a lot -- especially with spotting defensive resources.

Incidentally, I think calculation exercises are a lot more about spotting (and neutralizing) all of the relevant defensive resources than finding the primary solution. If I get a problem correct, I don't consider it a success if I blatantly overlook a key defensive resource that just happens to be insufficient. (In Tactics Land, surprise defensive resources conveniently turn out to be insufficient, or ALT-worthy, a lot more often than in actual games!)

Good guess. I got it up to ~1950 standard in February. Blitz was ~1650. But I stopped doing them rated -- caring about my rating was interfering with good habits and encouraging unhelpful ones. (I can't not care.) I have much better ways of tracking my progress anyway.

I'm significantly stronger at tactics/calculation than I was then, though. >:-D.

I'm so tl;dr prone.

Correct. How long did this take? Well, more relevantly, how long did it take to spot the Qa4+ and e4 ideas?

They should just "jump out" at you (in under five seconds).

If it took significantly longer, drilling simple motifs (one-shot tactics, and two-move combos) might be a pressing need for you. I'd consider those more than just warm-ups. You can address the rest of your leaks this way too, I think*.


*ETA: Actually, missing quiet moves could point towards a need for more verbal thinking, and missing zwishenzugs could be a product of calculating too quickly.

Fantastic indeed!

There is no plateau ~ There is no spoon #TheMatrix

Hmm, I have a question this time. What should Slav players who don't particularly enjoy the Semi-Slav or most other QGDs play against 1. c4?

1. c4 Nf6 2. d4 (or Nf3) c6 transposes to the Slav or ends up in a Reti, which is fine, but 2. Nc3 c6 3. e4 isn't great. I guess there is 2. Nc3 d5, which can either transpose to the Slav or head towards the Reversed Sicilian, which is fine too.

Is there any transpositional weirdness I should be aware of?

The transposition to the Caro/Panov after e4 you mentioned might be a problem if you don't want those lines in your repertoire, and you should expect most players to avoid d4 anyway (this is my experience as an amateur--IM+ have used it as a transpositional tool to get some line of the semi-slav they like, FM- just don't feel like playing d4 that day).

I've mainly been playing c6/d5 with an early dxc4 (mainly to avoid having to transpose to the semi-slav, which was otherwise difficult for me to avoid in practice), but I don't feel like it's changing the world. I also think the most common continuation I've faced has been g3 with White holding off on deciding the right spot for the queenside knight and maybe playing Qc2 or the like first.

Edit: Hopefully someone else answers your question though :-)

Need to read through this in a bit, after I have my coffee, but taking your advice I created a set.

This problem was the first I encountered and it took me 56s.

I invariably answer 1. c4 with 1...c6. If 2. Nc3, I'll play 2...d5.

You need a line after 1.c4 Nf6 2.Nc3 d5 3.exd5 Nxd5 4.d4 (also 4.e4)- 4....g6 is the Gruenfeld and 4....e6 is the Semi-Tarrasch.

IMO it makes more sense to play 1...c6 and learn some lines in the Panov after 2.e4.

Primarily the pieces moving, with a smidge of internal dialogue which is pretty transparent and simple. "The knight is pinned to the queen, if I can remove the knights defence, or get more attacking units on it..." That sort of thing. I probably do a lot less internal dialogue then I should, as the dialogue actually suits me a lot.

I think I will try to work on this a bit as it aligns with my general thinking processes.

A lot of times I will "solve" a problem, but it goes like this: I see a really strong variation, I double check "all" of the possible defences, I go for the variation, and it's right, but chesstempo plays a totally different defensive move then I had considered. I always hate that because it means I missed the best defence. I'm not sure how to fix this.

I'm guessing it was less than 30 seconds for the puzzle, but not sure of the time for the ideas. I solved it when I first woke up, so I was significantly weaker than normal but I saw Qa4+ but didn't see a follow up to it so looked for other candidates and saw the idea of e4, but the bishop can just take it, and then I put them together. So I investigated Qa4 before looking for e4.

I feel like I'm posting a lot here without contributing anything, but here is another "warmup" which took me 3:32 to solve. In reality, I spent 3:27 seconds trying to find a tactic on the f7 square or a mate on the h-file before seeing the real idea and solving it.

How do I fix this? What do I work on so that I can remove this blindness / fixation and see the simplicity?

All-ways look for the most forcing moves in a position, which is usually a check. In that position a check wins the rook right away. There is a book all about this idea called "Forcing Chess Moves" I like it, although it is a bit overkill.

This might sound cliche, but I think the only thing you can really do is to solve more and more of these problems. The more you see this motif, plus many more, the quicker you'll solve them. And as you begin to see the same motifs over and over, you'll get better at spotting them faster.

I don't think there's anything special you can do to see the simplicity of certain problems. In tactically rich positions, there are often times you'll have to spend time on various potential tactics and there's no way around it. Sometimes the "tactics" aren't there after all, like a mate on the h-file in the previous problem. The only way to figure out it doesn't work is by calculating.

And also, you won't know a problem is a simple one or not until you solve it. And OTB, there won't be a hint or anything either. So just drilling and hammering problem after problem and absorbing motif after motif is the way to go

Thanks guys, 1. c4 c6 seems to make the most sense.

Sometimes this is enough.

Cultivate the habit of scanning the board for "tactical seeds" before you pick any candidate moves or calculate anything. This seems very important.

thought process:
can we take on f7
no way it's protected too many times
hm the guy on e7 is loose
wait doesn't that check just win

The stronger you are, the faster you home in on the most important things in the position. Don't ask me how.

goldaxe, thought you might enjoy this game. Smyslov beats Olafsson (position after 17...Bc6) from a position with very similar pawn structures and same minor pieces. Black is probably actually slightly more uncomfortable after 17...Bc6 due to less space. It's always fascinating to me to watch guys beat strong GMs from positions like this.

http://www.chessgames.com/perl/chessgame?gid=1125875

I started thinking in the position after 22...bxc6, trying to play as white, since the position doesn't look too bad. One slightly annoying thing is what to do with our bishop. Obviously its best location is on the diagonal, but we should always be wary of playing f4 in this type of position, because the weaknesses and possible knight jumps to g4 might come back to bite us. So I hit on the move c5, and was quite happy with myself. Played through the next couple moves, and Smyslov's dxc5 is definitely the most impressive move of the game for me. So now I'm not sure what to play

Yeah the more I think about it, I reckon Black is just very slightly better here, he is going to have the safer king after g6-Kg7 later, while white's bishop just doesn't have an effective role, since Black will play c5 later.

And defending these sort of positions is exactly what I find most difficult. I have no idea how I should arrange my kingside pawns, for example