Is the Tactical Trainer seriously flawed?

I have gone through about 300 of the Tactics Trainers problems, and so far, I have noticed some really weird things:

1.  A lot of the solutions given aren't the best lines in the position.

2.  Many of the solutions are unforced moves, where many other things could have happened.

3. Some of the "problems" are so obvious that they aren't even problems, like simply moving out of check, or an obvious re-capture.

I feel like some of these problems were just submitted by random and weren't computer checked or quality checked.

Am I the only one who is concerned about this?  Do the problems get any better later on?

Nobody agrees with me?  Maybe I should post some examples...

Yes please do!  Examples are awesome.

Yeah, examples would be good, particularly for (1). For (2), usually if "many other things could have happened" then those other things are usually worse for the other guy, so they're as good as forced. For (3), check out the pass rates, you might be surprised at how many people get these "obvious" ones wrong.

I could be wrong, but as I understand it, the vast majority of the problems are computer generated and automatically trawled from real Chess.com games, and are then subject to modification by the moderators. I don't think you'll find many, if any, problems in there that someone has ineptly constructed by hand and just submitted without them being checked. It's possible that, after getting the first move right, there are alternative subsequent moves which are equally good and for some reason the computer has evaluated one as slightly better and decides the other one is "wrong", and the moderators try to weed these ones out when people report them.

Well I did a Tactics Trainer problem today which was ridiculously innaccurate.

Here it is:  http://www.chess.com/tactics/server?id=32059

A man claiming inaccuracy really ought to furnish proof. Where's the analysis?

Yeah, it would be really helpful if you pointed out what you thought the inaccuracy was.

Ok, let me rephrase that.

Although there are many tactical positions in the trainer which have incorrect solutions, this was an example of something else...

The one I posted is more along the lines of my #3 complaint, which is that many of the "problems" are beyond obvious.

In this case, white is already up a huge amount of material, the bishop is being threatened by a pawn, taking the pawn is simple and obvious, there is NOTHING to analyze, its not a chess problem, it doesn't belong in any kind of tactics database.

The issue I have is there are TONS of these types of pointless positions in the tactics trainer.  That is why I asked if there was any quality control at all on this thing?  After all, they are charging people money for this service.

Why do 36% of people get that one wrong, then? You might find it "beyond obvious", but the 1,800 or so folks who've got it wrong so far clearly didn't. The evidence doesn't appear to be on your side, with this one.

My two cents:

cent 1: OTB, a clear win in three is not worse than a clear win in two. Once you find one way to win, why look for another?

cent 2: Not nearly enough (in fact almost none) defensive combinations. To be fair, this is a common problem in most tactics trainers, and even books, not limited to chess.com

"There are no Tactics involved, Your just looking for the first move in a fixed progressive line leading to a conclusion.

a: the science and art of disposing and maneuvering forces in combat     b: the art or skill of employing available means to accomplish an end"

No tactics involved? They fit precisely with your second definition there.

How is mate in X not employing available means to accomplish an end?

"The Tactic therfores is finding the right move at the start so that whatever the opponant does - when he has a choice - you have the correct response ready."

The tactics trainer is made so that after your first move, you get an advantage no matter what the response of the opponent. There is analysis and source for alternate lines, the puzzle normally progresses down the most challenging responses.

If you don't understand the puzzle, ask in the comments section. For most people there is no need for a why, and it would take a lot of work.

I think the issue with tactics trainer is that they can not accept more than one "good" response.  So, many tactics that should be continued are stopped to soon. 

Definitely agree with number one.

As for the thread starter.

For 1) you're unlikely to be correct, they are created by a computer. I see a lot in the comments at the bottom that people think they have a better move, they are almost always wrong.

2) Most of the 'unforcing' moves are mating nets which the opponent wont be able to escape from. Not every move has to be a capture or check, I think these are some of the best problems.

3) As people say they still have a low success rate. And it's useful when you have an obvious recapture to be trained to look for something better first, and it helps board vision.

I disagree with all three points.

I used to read the comments, but they're full of people whining that because they don't understand the solution then the solution must be wrong.

So now if I don't understand then I look at them in Fritz, I've yet to find an error or better alternative solution and I've yet to  find an arbitrary or unforced response.

As for the ratings, surely it works the same way as the human-to-human one does, with everyone's response thrown in the pot to create a relative grading system?

Karl, from my experience, I would say that it is VERY rare that in TT, there is a stronger move for the attacker (= side of the user solving the problem) than given in the main line solution.

If you think you have spotted such a rare exception, I would suggest you thoroughly check with an engine first to make sure you haven't overlooked anything. If you then still are sure, tell me about it, and I will fix or delete that problem in TT.

I agree that there are some problems in TT which fit your description of "beyond obvious and thus not really a tactics", to such a degree that users from all rating ranges will agree on feeling that they are not interesting.

We are weeding them out as we come across them, in order to continuously improve tactics trainer.

If you come across them, feel free to simply to pm me a list with their links so I can delete them from TT.

I've checked out and reported a lot of TT problems to moderators, and those problems that are defective get removed or edited fairly quickly.

Right now, I work problems in the 1700-2000 range, and I'd say that only about 2-3% are defective, and that number is shrinking.

You definitely should check out a problem first before you report it, however.  Besides saving the moderators from unnecessary work, you gain a better understanding of the position in question.

Is the Tactics Trainer seriously flawed?

No

Tactics Trainer is a valuable resource.

Most of the time it is the main reason why I sign in to chess.com.

Are there occasionally problems that bug me?  Yes.

Is it seriously flawed?  Not even close.

Amy

Yeah, definitely need to check with a computer.  I often think "doesn't this move win too?!" when I fail to solve a puzzle... but >99% of the time that means I missed something and it's learning time.

Thank you for looking into the problems. I normally post comment on Tactic Problem page maybe its better to pm you on this one.