I wonder if anyone else has had this feeling in Tactics Trainer...

Sort:
DelCheMethod

Oh, I plan to turn that little bastard on again one day! But I want to make sure I'm solid and quick on my thought process before I do. I'm afraid if I bounce back and forth, I'll never establish a solid method as I'm the type that easily goes back to the easy way.

With all that said, I have to really ask myself why would I want to go back? Solving now is fun and educational. Kind oF like doing a Suduko puzzle, relaxing to study. Before it was all "the clocks almost out, OMG - sac the queen" and the follow up "that's just stupid. Who would ever think of those moves".

I really enjoy the new way. Sure I can't brag about how my kick ass 1400 rating is on its way to 2000. But then again, I i like getting the problems right.

Scottrf
DelCheMethod wrote:

I really enjoy the new way. Sure I can't brag about how my kick ass 1400 rating is on its way to 2000. But then again, I i like getting the problems right.

Yeah, but of course you will be. If you're constantly timing out your rating never improves so you're solving easy puzzles. I'd rather see new patterns than stay at 1500 and check that I can still do the smothered mate I've seen 1000 times before.

chrka

I agree with Ozzie. Forget about the rating. As you'll get better it will rise anyway. The past year I have been doing TT for about 15 minutes a day, and I have gone from about 1400 to about 1800-1900 during that time.

However, as soon as I start to think about my rating I start playing too fast and will make stupid mistakes and the rating will drop. So, set a timer for whatever time you find appropriate and try to not look at the rating. Remember, the rating is just there so that TT can give you problems of the right difficulty. You could just look at your percentage or give yourself a cookie each time you get a problem right instead.

DelCheMethod
Ah, that's the rub. Each week, I've been bump up the level of problems. I think I have a 300 point range,. For example, this week I have 1200-1500 problems. Next week, barring significant trouble, I'll bump up to 1300-1600. This will give me review problems with the lower levels and new problems at the higher level. The method I mentioned works fine for the current levels. We'll see how it goes when I hit 1700+ ratings and what tweaks ill have to make on harder problems.
Dutchday

I think the tactics trainer is hateful, since you're practically forced to pick an obvious line before you looked at everything. I'll take a book any day.

In a real game you already know there is ''some threat'' in the position, though it was ''not really working'' last move. It's not really surprising you come up with a nice tactic if you've been looking at the game for +2 hours and 10 minutes for the critical move. 10 minutes may be a bit too much to use on a move, but in the end it works better and a lot of people do it in difficult middle games.

From a psychological point of view, I think the tactics trainer builds up a mind set that is prone to haste and tunnel vision. In fact you should be calm and keep an open mind, while you can still focus. 

Noreaster

Someone said it best earlier you should get a hold of a basic tactics book and study these basic motifs. Set up a board and really study them. Basic puzzles are easy to solve but this is not the real purpose of this excersise. The point is to notice the patterns underlying these tactics. The tactics trainer is good for what it is which is honing your ability to scan for tactics. It is not for exposing the how's and why's of tactics.

Repens
CharlesConrad wrote:

The feeling of spinning your wheels.

This is my standing on the matter as of me writing this:

Your Rating: 1189

Today's Attempts: 59 (Average Score: 45%)      

At some point today I was within a good score of getting back over 1300 (had made it to 1290 range).        

I've never been over 1450 and I've been a premium member for nearly a year. Tactics Trainer is the tool I use most and I have to be doing it wrong because it's claim of sharpening my tactical vision isn't helping me.

Man, some of these exercises leave me cursing under my breath as I miss something plainly easy and my score drops yet again.                    

 

 

1. as said by many posts do not worry about the rating.

2. before you start another tactics session go to the performance button and review the areas you are having trouble with.

3. Re-play through the failed attempts, do so until you understand the theme (do not just click through). 

4. Have the Tactics Trainer pause after each problem so that you can review the successful attempts as well.  

5. I know time is a factor but, try to calculate thoroughly and not click and hope. The moves will become easier to spot.

Have fun Smile

DelCheMethod

After a couple of hours sleep, I have one last thing in my thought process that I forgot to post last night...

6) after I solve the problem, I look at the entire board and try to figure out act a macro level how the players got there. For example, if the problem appears to be a kings pawn opening with a kingside attack, a rook pawn endgame, etc.

6a) if I miss the problem, why? Board vision ( not seeing that Bishop in the corner), missed a pin (skipped my null move step), just being lazy with calculation, etc., Im hoping the after I really identify the cause, the habit I'm hoping to develop will account for these misses.

I hope this step will tie back to gameplay and being able to confidently look for shots when I haven't been told there is a solution.

NimzoRoy

Try doing them unrated, and just keep on going back to each one until you've got it down pat. If they're unrated the time doesn't count so you don't have to worry about being able to solve them at blitz speed which is totally bogus unless you only play blitz chess.

Rated or unrated get every position down pat before moving on to the next one. Once you do a TT position as rated you can go over it as many times as you like, because it's only rated the first time.

As a Diamond member you can also watch videos and do the chess mentor, so start doing them too instead of getting bummed out and/or pissed off at the TT constantly, like I used to. It looks like TT is practically all you do here you've only played a handful of games in each category so start playing more chess, watch videos, try the chess mentor and cut back on the TT for awhile - way back.

rooperi

Lol, just came accross a TT problem: 0432835

Avg time 7070 seconds (almost 2 hours!)

Attempts 111

Success 0.

rating 2000-2100

Doggy_Style
rooperi wrote:

Lol, just came accross a TT problem: 0432835

Avg time 7070 seconds (almost 2 hours!)

Attempts 111

Success 0.

rating 2000-2100

Surely, no one would sit in front of a single problem for two hours. I suspect a bug with the problem or at the very least, incorrect recording.

Scottrf

Will probably be explained by the time being based on correct solves. Without one it might just show the total time spent on the problem or something.

rooperi

Nope, it's real, read the comments

Some issues, though, I get engine evals of -100, but you gotta nail the mate in x.

It's a 7 move problem, but after 3 I think a 1300 can really beat a 2700

PAMetalBoss
ozzie_c_cobblepot wrote:

+1 to the "100% certainty solving" idea.

Instead of focusing on the rating, focus on the percentage correct. In my case, I was very dismayed at some point that I had a minus score - more incorrect than correct. After building my way back up to even, I am now at almost +900.

Correct me if I'm wrong, but doesn't the statistics for tactics trainer tell you what problems you passed and what problems you didn't pass? If you take too much time, even you find the right move it will count as a fail in your stats. In order to pass a problem you need a socre of 60%. It is possible to get a 60% without getting all the moves correct, for example if there are 4 moves and you get 3 of them right in a fast enough time, you have passed the problem. 

Scottrf

That would probably be removed if reported. I know they aren't adding new ones with that level of ambiguity.

Scottrf
PAMetalBoss wrote:
ozzie_c_cobblepot wrote:

+1 to the "100% certainty solving" idea.

Instead of focusing on the rating, focus on the percentage correct. In my case, I was very dismayed at some point that I had a minus score - more incorrect than correct. After building my way back up to even, I am now at almost +900.

Correct me if I'm wrong, but doesn't the statistics for tactics trainer tell you what problems you passed and what problems you didn't pass? If you take too much time, even you find the right move it will count as a fail in your stats. In order to pass a problem you need a socre of 60%. It is possible to get a 60% without getting all the moves correct, for example if there are 4 moves and you get 3 of them right in a fast enough time, you have passed the problem. 

A pass is a pass, if you take 10 seconds or an hour.

However, 3 out of 4 is counted as a fail, not a pass.

Pass with time elapsed (average time x 2) gets you 20% times the ratio of the problem you get correct. e.g. 2 out of four moves with time elapsed gets you 10% and a fail.

You are correct though in pointing out a flaw: you can make 3 natural moves out of four quickly and miss the key move (scoring 75%), which would lose you the game, and score higher than someone who takes an extra 20 seconds and calculates it accurately to the end.

reboc

To improve your chess game, good teachers like Dan Heisman tell you to play slow games (ie games of at least 30 minutes per side) and take time making calculations, etc.

I suspect the same thing applies to studying tactics. The goal is not to hit a 1300 rating on Tactics Trainer, the goal is to consistently improve. It takes time (and a certain type of careful thought) to learn increasingly complex tactical patterns.

I find it hard to discipline myself to do the slow/improvement work when so much of internet chess is geared to blitz (and blitz is so fun!)

ozzie_c_cobblepot

@Scottrf is correct, passing is passing. There is no 60% magic threshold.

SmyslovFan

My problem with the tactics trainer here is that there are some puzzles where you make three perfect moves and reach a winning position, but the puzzle keeps going on. So what if the fourth move leads to mate in 5 instead of mate in 3. They're both winning. 

Stop the puzzle when the position is clearly winning.

PAMetalBoss

Yeah, I don't think it's fair that if you find a mate in 5 instead of a mate in 3 that you should fail the problem. There are tons of problems where taking 1 or 2 extra moves makes you fail. I mean, it's only a matter of seconds.