Came here for the poker but that hasn't been going well for me lately, so I'm taking a break from poker and in the meantime I've been playing more chess. I've managed to maintain a certain level of chess mediocrity for a long time. This is me on chess.com: "dirty, infect coward and son of a bitch !"

I could delete that comment on my member page of course, but it amuses me. The guy lost to me after hanging a rook and didn't appreciate my refusal of a rematch, I think. At least that explains the "coward" part.

I've been playing mostly 10 minute blitz, 15 | 10 standard, and I've been using the chess.com tactics trainer. Current ratings:
1319 blitz
1576 standard
1471 tactics

So that's where I'm at, and now that I've shaken off the rust once again I'm probably about as good at chess as I've ever been. But I haven't significantly improved in quite a while. I'm over 40 so I'm not expecting too much, and after all this is just a hobby (at my level it had better be). Nevertheless, if I'm going to keep playing, I might as well try to improve. Something like 1500 blitz, 1750 standard by this time next year would be pretty cool as far as I'm concerned. That's about what I'd expect with a reasonable level of sustained effort but if I can't do it I can't do it, and that's fine as long as I try. Or maybe I'll surprise myself and go higher. Or I might get sick of the game and stop playing. But enough of that, moving on...

I used to occasionally get chess fever for brief periods of time, and I'd buy chess books that mostly went unread. As such I'm making a point to not buy more chess books now but anyway, I dug one up on the Scandinavian Defense that I've been looking over. I've decided to give the Scandinavian with 2 - Nf6 (instead of the somewhat more common recapture with the queen) a try, and since most opponents at my level open with e4, I expect to have plenty of opportunities to play it. After 1 e4 d5 2 exd5 Nf6, an unsophisticated player (like me!) would naturally want to play 3 c4 with the idea of hanging onto the pawn. The main line is 3 d4 though 3 c4 is playable, but after 3 c4 c6 the average fish like me (but who unlike me hasn't read up on this opening) would play 4 dxc6, which from what I've read is a bit of a mistake. If I learn how to exploit that, playing this variation should work out well for me. Here's a 10 minute game I played today in that line:
http://www.chess.com/livechess/game?id=1048000881

Cliffs: I somewhat misplayed the opening as early as move 5, managed to create a clearly winning attack on white's uncastled king anyway, botched it completely, but converted the endgame ftw. And incidentally my douchebag opponent decided to make me wait a few minutes for his clock to run out instead of resigning or making a move when faced with mate in 1. Does this kind of thing get less common at higher ratings? One more reason to get better if so.

This line (Tal's gambit) is very sound even vs those who correctly play 3. d4

Good luck! Also, as you've already played 7 correspondence games and are involved in 7, you might be interested in playing more of them for our team

Thanks for the invite, I put in the request to join the team. I've never heard of that line referred to as Tal's gambit but even though it doesn't seem to be common at the top level, grandmasters have played it before which is good enough for me. I'm certainly not happy to see 3 d4 though, and not sure whether to go with the standard Nxd5 4 c4 (and then deciding whether the knight goes back on f6 or b6), or to try the Portuguese variation with 3 - Bg4 which just looks weird to me.

Oops, I must have confused this line (which is called simply the Modern Variation of the Scandinavian) with the similar powerful Tal Gambit in the Sicilian: 1. e4 c5 2. f4 d5 3. exd5 Nxf6 4. Bb5+ Nbd7 5. c4 a6.

I'm not a specialist in the Scandi, but it seems to me that the purpose of the Portuguese 3.- Bg4 move is to either disrupt White's development (if he plays f3, the knight loses a natural development square) or prepare good retreatment squares for the queen (after Qxd5 Nc3) - f5 or h5 (as opposed to the usual d6).

Current chess.com ratings:
1287 blitz
1575 standard
1742 correspondence
1408 tactics

Those numbers aren't moving in the right direction so far. I think I've only played one "standard" game since last posting, otherwise I played some 10 minute and 5|5 games, some of which were played in a binge while getting increasingly frustrated with my poor play, then I stopped playing blitz for a while. I'd like to settle on a comfortable time control for the blitz games though. 5 minutes is too fast for me, and 10 minutes is usually ok but if the game goes too many moves I hate the time scrambles at the end. So I'd like to play with some kind of an increment. When I put out seeks at 5|5 it usually takes a while to get any takers. I think 5|2 is a more popular setting so I'm going to start making that my standard game when I want to play blitz. I usually play on my iPad (hate the chess.com browser interface) and to avoid mistakes entering moves I currently have it set to require confirmation after tapping in each move, but for 5|2 I think I'll have to turn that off.

So far I've finished 12 correspondence games: 11 wins, 1 loss. Of course some of those wins were against much weaker opponents while I got my rating established. I'll post three of the more interesting games now.

Game 1:
http://www.chess.com/echess/game?id=104204890

This was a tough win against an opponent near my rating level. At least, his correspondence rating is close to mine. On the other hand he's 1779 blitz after over 1000 games so I don't understand how I could be anywhere near his level. Must not take the correspondence games all that seriously. Anyway in the opening (Caro Kann advance variation) I might be overly eager to exchange pieces - 4 Bd3 and especially 7 Qb5+. It's a habit I've developed from playing against computer programs, where even at the lower settings I play against, the computers still tend to be tactical monsters. So I'm happy to get the queens off the board in general, but this is something to unlearn when playing against humans.

I had an awkwardly placed knight on the queenside throughout the middle game but managed to trade it off for a bishop on move 31. After that play focused on black's passed pawn, which had advanced too far to survive. With 32 Re2 I wasted a move before realizing the rook had to go around (e3 then c3 then c2 though it never got to that point) so the king could occupy e2 and the 3 attackers could take down the pawn. I played Re2 with the idea of putting the king on e3 but of course it gets chased away with Rd3+ once black's rooks are doubled on the d-file. Not sure why black was so eager to exchange off the rooks once the passed pawn was eliminated to reach the lost endgame though.

Game 2:
http://www.chess.com/echess/game?id=104366352

I had no business winning this one. It started as a bishop's opening which I've been trying to implement against e5, and playing f4 when possible as I did in this game. I'm playing according to the recommendations of "Attacking with 1 e4" by John Emms (both with the bishop's opening and also the closed sicilian with 2 Nc3, which I used in game 3). I overlooked 7 - Qh4+ where black wins a pawn, though according to the HIARCS analysis I ran after the game, after 9 Be3 white actually has a slight edge (+0.64 at depth 22). The way I saw it during the game, I had simply blundered away a pawn. By move 25 I saw no way to improve my position and was just hoping for a draw, but black misplayed the ending so badly that I ended up winning it. 32 - b5 in particular made it very clear to me that black had no idea what to do in the endgame.

Game 3:
http://www.chess.com/echess/game?id=104908622

The aforementioned closed Sicilian, which I ended up losing. My 13 Bf3 Nd4 14 Bg2 was a definite "oops, didn't mean to do that". I couldn't come up with a decent plan there. I think it's obvious I should be using my kingside pawns to build up an attack but that doesn't translate directly into moves. As played I wasted too much time before advancing the pawns and gave black time to proceed with his attack first. Luckily for me, his setup after move 25 with the pawns on a4 and b3 and the knight on c2, while claiming territory near my king, just isn't all that threatening compared to what he could have done. So I got a chance to build my own attack, finally. After offering up the queen sac on move 32 I thought I was headed for a win. Black defended about as well as he could, and apparently (per computer analysis) I should have played 33 Rxh7+, which I considered, but wasn't sure if it worked tactically. The line is complex enough that I can't really fault myself for missing it. In any event the computer says the move I played 33 Qh4 should still be winning, albeit less decisively. But only if followed up with the sac 34 Nf5 which I definitely wouldn't have found. 34 - Nc4 was a very nice find by my opponent (if I take the bishop black plays a3 and if I then take the queen I'm getting mated quickly - which I saw during the game). After that I was lost.

Don't play handicapped engines unless you can commit to passing up on easy exploits. It can hurt you. As you pointed out, quite a few of your exchanges are really suspect.

Yeah, White has a lot of development and various other positional trumps. One of the things 6.-Nfd7 does is open the diagonal for the queen, and Qh4+ becomes harder to miss after you explicitly notice that.

Bf3 moves a piece to an unprotected square. Whenever you want to do that -- in fact, whenever you want to move a piece to a square with an equal number of attackers and defenders -- you should glance at all the ways it can be attacked. It's hard to miss moves like Nd4 after doing that. Do the same for your opponent's moves.

This is a general principle, but I'm not a hypocrite. Heuristics for spotting concrete lines are more valuable than positional principles.

Current chess.com ratings:
1330 blitz
1594 standard

Not much progress but I've started playing more chess again. I did manage to beat Computer3-HARD in 3 | 2 blitz for the first time. Warning: It's ugly.
http://www.chess.com/livechess/game?id=1266142172

If I play the HARD computer more, and the MEDIUM computer less, the games should be a little less silly on average. Although the game I won is a bad example. In order to increase my chances, I'm taking notes on the computer's opening repertoire. And also starting off with slower games at 10 | 10 but the eventual goal is to win frequently at 3 | 2. Totally treating this like trying to beat a level in a video game, by whatever means necessary.

Here's a win at 10 | 10 in the open Sicilian, which I've just decided to start playing after watching John Bartholomew's excellent video series on climbing the rating ladder (my comfort zone is closed Sicilian with 2 Nc3).
http://www.chess.com/livechess/game?id=1268212616

Looking through the database afterwards, I found that the computer only plays 4... Qb6 occasionally in that line, and 6... d6 a little less than half the time, but can you believe that it traps its queen with 7... Qb4 every time it reaches that position? Useless information as far as general chess development is concerned, but quite relevant from a beating the video game perspective (though it doesn't do Qb6 that often so maybe not). Did they just sprinkle in some randomly ****ty moves into its opening book? I don't see how else it would come up with that, the search can't be that shallow.

I've also been playing people from time to time.

Those comps simulate weakness by randomly playing bad moves.

It's better to play against comps with short depth searches, but not 1 ply or something silly. They are still really good at basic tactics, so it's good practice for picking up on threats and not blundering.

Yeah I don't know why they don't have computers like that, instead of the random terrible move style of handicapping. Limiting the number of nodes searched per move would do it. In the case of chess.com, their computers have specific blunders built into the opening books, and then they also will make random sacrifices later in the game. EASY does this constantly, MEDIUM a little less, and with HARD you can't rely on it doing the nonsense sac in any given game but it still happens. I don't know how many games I've won against EASY and MEDIUM in the Italian game where it ignores the pin on its f6 knight and I win the queen. It's a specific position and I believe it captures the pawn on e4 every single time.

http://www.chess.com/explorer/?white...=13&id=1623141

That position and make sure the computer is set to play black. 541 times out of 551 in the database it played 7... Nxe4 (that must not include blitz games maybe its only turn based? even though I selected online & live). Easy rating boost there for lower rated players who have no shame.

And here's something else weird about the chess.com computers, or perhaps the rating system. Last I checked HARD (for example) had a higher standard than blitz rating. But it's way easier to beat at 10 | 10 than at 3 | 2 for obvious reasons. I don't think it plays any differently between the time controls, the only difference is the opponent gets more time to think at the slower time control.

Played a bunch of blitz games badly for the most part, and got the rating under 1300 again. Maybe that's where I should be right now when I'm not just playing against the computer though. I'm getting over my fear of the open Sicilian as white, even though I don't know it very well, and I'm finding my opponents don't really know it either. My two most recent games were both Sicilians and I should have won both, but in time trouble my endgame technique was spectacularly poor.

http://www.chess.com/livechess/game?id=1270248634

17 Nf7 doesn't really accomplish anything, should have played c4 to win black's pinned knight (after Qa5+ 18 Kf1). Missed it again on the next move.

I really underestimated black's center pawns. Shouldn't be a difficult win but I had to be somewhat careful there. A few times I could have won the e pawn and it's pretty easy after that. Time pressure + panic = loss.

http://www.chess.com/livechess/game?id=1270260865

I present this game as a candidate for worst endgame ever played, by both players.

When I was looking for a response in the bishop's opening for when my opponent mirrors my first two moves (something other than going into the Italian game), Stockfish's analysis suggested something strange but apparently playable.

1 e4 e5
2 Bc4 Bc5
3 Qg4

It's not great, but if black plays anything other than Qf6 white gets a nice advantage. After 3... Qf6 4 Qg3 it's probably about even (ie, Stockfish gives white +.08 at depth 42). Of course, Computer3-HARD doesn't play 3... Qf6 there, but it does play 3... d5, which isn't going to go very well for black.

In the endgame, Kf1-e2 is the "go to sleep" plan, and it's nearly impossible to lose after that. I noticed a general tendency after the queens came off to delay (and not attempt, really) centralizing the king.

Edit: Looking at the next game, trying to do everything with pieces and pawns and leaving the king at home is actually a trend. On move 31, Black is completely tied down, so you can gain time by playing Kc1-d2-e3 before munching the pawns (which are doomed no matter what) and save yourself a lot of hassle in potential time trouble.

In the first game especially Kf1-e2 is really obvious once you mention it, but not something I look for immediately, so in time pressure I never got to it. I know in the abstract to activate my king in the endgame, and of course I do it once it's down to just king and pawns (plus maybe a minor piece), but if there are other pieces it does seem to be a trend that I want to put the pieces on mop up duty first. In blitz games I have lots of trouble in winning positions when I'm up material but need to hold back a wall of opposing pawns, and not activating the king in time is probably why.

http://www.chess.com/livechess/game?id=1270357567

A win against a higher rated opponent. Not a very good game, I was on my heels the whole way after getting outplayed in the opening, but managed to stay in it until he blundered in time pressure (he missed 28 Qxe8+ Qxe8 29 Rxe8#, then hung his knight a few moves later).

3 Bb5+ gave me trouble and I wasn't sure how to respond. I blocked with the bishop on d7, which is fine (or Nbd7 is ok too), but when he didn't immediately exchange bishops like I'm used to and instead played the apparently standard 4 Bc4, I didn't know what to do after that. 4... Bg4 inducing f3 is the main line. Not the first move that comes to mind for me, but since I play the Nf6 Scandi, noted.

Are you doing any studying or training besides looking over your games a bit?

Also, is there a reason you play against computers so much rather than players?

There's been no serious study of any kind, just playing mostly blitz games against the computers while putting off other things and to see how high I could get my rating. I went on a run and managed to get my blitz rating up to 1465 at one point, but I think all I really did was I got very efficient at beating the computer. Maybe I got a little better at the time management aspect of blitz in general. In the few slower games I played my rating actually went down, and if I went back to playing humans I expect the blitz rating would drop. That experiment has probably run its course, at least with the easy and medium computers. It's just habit forming, and especially when I have other things to do, a few quick games with the computer becomes tempting.

Going forward, no more blitz games with the easy and medium computers would be a good policy (playing against the hard computer still has value though). Why do I play against computers so much? Partly I really hate losing against people. It's a psychological thing that I don't think makes much sense, but it bothers me to think there might be somebody on the other side of the games I lose laughing at my mistakes, or something.

Weirdly, it's not that uncommon for the easy computer to have a higher blitz rating than the medium computer, but I'd like to stop concerning myself with such things. As I said, playing the computers gets addicting.

Hmm, just playing blitz to get blitz higher is mainly going to be variance. You won't likely improve that much over time (well, at the very beginning you will get better as you get used to the time control and such).

I think playing endless blitz game against computers is a one-way ticket to zero improvement fwiw. Obviously just my opinion but if you actually want to get better I'd highly recommend at least playing longer games against humans (and then going over them afterwards) and studying (although I can appreciate that may just not be fun and so not worth it for you).

Perhaps one other thing is that if you play humans a bunch you will care less and less, eventually not worrying about losing to them vs. a computer at all. I guess I can understand to some extent since for video games I often do not want to play vs. humans just the AI over and over again. Although in that situation I feel it's a pretty different situation vs. chess.

When I was 1400 blitz, chesstempo[dot]com alone get me to 1600. I think tactics are essential and when I was 1400 I played some sessions where I'm in "tactic" mode, means I'm only looking for tactics and neglect things like doubled pawn, weak square, etc. Just focusing to identify tactics because 1400-1600 players hang pieces (not in one move) a lot.

200+ points is a lot to improve in one year. I've had 2 big rating increases within a year, and its certainly always fun watching your rating climb in such a short period of time, but it isn't easy.

For me, my two biggest rating increases didn't happen by accident.

First - I had been playing chess for just over a year, mostly online. I played in 3 tournaments and had a provisional rating of around 1450. I then moved to las vegas and played in one of the top 5 run chess clubs in the country, run by Chris Bird who has sinced moved on to being a big time TD.

At this time i bought The Amateurs Mind by Silman and was going to the club 2x week. Chris and another expert were always helping me go over my games and I was 1650 in about 9 months. Not long after, Chris moved on, and the club became much less organized. I was still steadily improving. After a while I reached a plateau around 1790'ish, but was I probably 100 points stronger. But the club was frustrating, there were a lot of players u1600 and over 2000 but very few A/B players. So I was getting stronger, but my rating wasn't going up because I couldn't beat the experts yet. After a while I got frustrated at the clubs organization, the fact my rating had been stagnant, and also poker was picking up. So I stopped going to the club for 2 years.

The second major jump I had is related to this comment. I not sure it applies as much for people under 2000.


In the 2 years off from live chess, I didn't study but I played a ton of 3m chess online. This really improved my tactical play, but more so my chess intuition. I had to really learn how to instinctively develop plans while being mindful of tactics, very quickly.

When I returned to the club, I think I went a full year with only 2 losses. I went up 200 points. I've moved twice since, back in LV now, but I don't really have time for tournaments or clubs the last couple of years. I've gradually moved up to 2100 rating, probably 2150 strength, while only playing slow chess 2-5x year. I still play tons of 3m chess because it is mostly what I have time for, and when I can I'll enter an occasional tournament.

I've reached another point where my progress is really slowing. A lot of it is lack of playing, but most of it is probably the game is really friggin hard and the higher you get the harder it is to improve considerably - especially just playing blitz. While I would love another 150-200pt boost, I don't see it happening unless I dedicate hours to proper study.

I agree with SicilianTaimanov that chesstempo would be a great place to improve. I really should do it too, but 15-20min a day there working on tactics will make you so much better.

best of luck in your improvement

Honestly I don't think 15-20 mins/day on tactics is enough. During my chess days I used to solve 50-75 tactics per day and one tactic usually take 1-2 minutes (some might take 4-6 mins). After solving about 12k tactics, I went to 1800 blitz online. I was able to beat few national masters at regional tournament after 2 years.
Another book I recommend is Understanding Chess Move by Move by John Nunn. It completely changes my way of thinking and get into GM's mind on understanding chess.

Funny thing is, I prefer studying chess to playing chess, but in poker I can't even allocate 30mins per day to review HH lol

studying chess is nearly always fun. studying poker is nearly always soulless

I posted a youtube video of three 10 | 10 games I played against Computer3-HARD. Nothing special about the games and they certainly weren't cherry picked for being high quality, just the games I happened to play when I started recording. Then a brief computer analysis after the games.