People need to get past the "points", and concentrate on the learning, and pattern recognition. When i quit worrying about how "fast" i solved tactics, my rating started going up.
Accuracy first, speed is a by-product.
Well said, indeed. I agree with both these comments.
Yes, a game of chess is trying to have it both ways. You are right about that. You have to not make a mistake, while at the same time, capitalize on your opponent making one.
You are refusing to acknowledge my point. I don't disagree with any of your eloquence about the uncompromising glories of chess. I only question the assumption that the only way to do a chess puzzle is to do it against the clock.
I think that was already covered by a few different people. The puzzles that you get right are against the clock, meaning if you get it right the game goes on with varying degrees of success. If you get the puzzle wrong, it's not against the clock. If you want to take your time getting it wrong, you may, there is no additional point deduction. The point decuction is the same no matter how long it took. I think it's because if you lose the game because of the mistake it makes no difference how long it took you to make the mistake.
So you think that people going slower should get less points if they fail ?
If they get fewer points when they succeed, then it would seem only fair for them to lose fewer points when they fail.
You want people that take a long time to fail a problem to get a benefit for taking too long? If anything, it would be the other way. If you fail a problem that should have had a quick solve time based on your current rating, you would lose more points, since not only should you have solved it, you took a lot longer than someone at your rating should have taken to solve it.
And you don't have to do the puzzles against the clock. Just ignore the clock if you can. And the ratings work pretty well the way it works now. I get a very good mix of tactics, missing some, getting some right. Get some quickly, fail some quickly, get some after some thought, fail some after some thought.
You can approach the tactics trainer as a speed drill or for calculation. You can do both at different times, if you want. I understand the desire to have a higher rating but looking at your stats, you are getting a very good mix; 60% pass rate and a 40% fail rate. Work on keeping the pass rate as high as you can get it and your speed will improve as you learn the patterns and your rating will follow.
So you think that people going slower should get less points if they fail ?
If they get fewer points when they succeed, then it would seem only fair for them to lose fewer points when they fail.
You want people that take a long time to fail a problem to get a benefit for taking too long? If anything, it would be the other way. If you fail a problem that should have had a quick solve time based on your current rating, you would lose more points, since not only should you have solved it, you took a lot longer than someone at your rating should have taken to solve it.
And you don't have to do the puzzles against the clock. Just ignore the clock if you can. And the ratings work pretty well the way it works now. I get a very good mix of tactics, missing some, getting some right. Get some quickly, fail some quickly, get some after some thought, fail some after some thought.
You can approach the tactics trainer as a speed drill or for calculation. You can do both at different times, if you want. I understand the desire to have a higher rating but looking at your stats, you are getting a very good mix; 60% pass rate and a 40% fail rate. Work on keeping the pass rate as high as you can get it and your speed will improve as you learn the patterns and your rating will follow.
Same thought as martin why should being slower help you it should only disadvantage you.
I understand the desire to have a higher rating but looking at your stats, you are getting a very good mix; 60% pass rate and a 40% fail rate.
This actually illustrates my point. If my rating were accurate, which would mean it is giving me puzzles commensurate with my ability, then my pass/fail rate would be 50%/50%. However, since my fails are being counted several times more than my passes, the ratio is skewed (as you observed). This is freshman statistics, and it's frustrating that nobody will even acknowledge the point.
Overall, I'm inclined to stop here. All three of you are just repeating what you said in the first place, and not acknowledging my point.
The tactic selection algorithm gives you a range of tactics, from approximately 400 points below your current rating to 400 points above. That mix is going to have a pretty big impact on how many you get right and wrong, so any modification of that selection algorithm will move the percentages some.
The thing is, taking your time on tactics, can actually give you a higher pass percentage. If you don't solve the tactics until you have 100% of the solution, disregarding time completely, then you can maintain a very high pass percentage. Yeah, you'll still miss some on patterns you don't know, don't understand, or can't calculate.
There are a lot of variables in play but the simple fact is that you go into the problem knowing in advance that there is something there to find. That is a huge advantage already and giving a smaller point loss on failure because you take more time is not something that is a good idea.
My rate is currently 57% pass, 43% failure. My last session I was closer to 49% pass and 51% fail. I was solving pretty quickly, without trying to solve 100% in my head on quite a few of them (I was testing a beta feature). Some I took my time on as well.
The only reason my pass percent is as high as it is, is that I made a concerted effort to keep my pass rate high after the last time I reset my stats when the algorithms changed back in April of last year. I don't do tactics a ton but usually when I do, I'm passing the time and am not overly concerned with keeping my pass rate high, so it has fallen.
Typically I don't concern myself with how the rating algorithm is constructed. As it seems to get tweaked every year or so anyway.
The tactics trainer is just a tool. When I do use it I'm checking the quality of my visualization and calculation ability. Looking for weak points and blind spots in my own thought processes.
Another thing that I do is play out the "supposedly" winning side trying to convert the advantage into a full point. I mean some are completely winning and it isn't necessary for me to even pursue those types of positions.
But often or not there are some that are unclear to me and to maintain the advantage you have to walk an incredibly thin line where there is only one variation that works.
Having the advantage is one thing, possessing the technique to see it through is another.
Practicing with the TT this way has really helped me learn some new plans that I would have not encountered had I simply moved onto the next problem.
io non ho capito come funziona: devo cercare il matto o devo prendere la regina? 
I did not not know until reading this tip from Martin that you can wriggle out of the huge penalty for getting it wrong by just clicking the '>' tactics arrow
That has kind of spoiled it a bit for me!
I used to be in a real sweat after a minute had ticked by and I knew I was not gonna score any points but I was still gonna get clobbered if I put a foot wrong.
I vote for 'mandatory penalties' once you start the clock on a tactic even if you decline to make a move!
Don't let anybody wriggle out of it with the 'escape' button!
So you think that people going slower should get less points if they fail ?
If they get fewer points when they succeed, then it would seem only fair for them to lose fewer points when they fail.
Does not make any sense.
If anything you should get docked less points for making a wrong move having taken only 10 second to think vs making a wrong move having taken 2 minutes to think.
Either way there should be no wriggling out of it with the 'escape' button or your Tactics score is a sham!
I respect the viewpoint of wanting to be measured against a clock as you do tactics. I really do. I just don't share that sentiment.
After this conversation, my suggestion would be to set up two Tactics ratings, one timed and one untimed. This would be analogous to the three ratings we currently have for game play: Blitz, Rapid, and Daily.
I do not plan to post further on this thread, unless someone says something substantially new.
in chess we never know what will happen next move. Imagine if the guy who played the games from which they took the trouble, told you that some solutions are wrong?
I respect the viewpoint of wanting to be measured against a clock as you do tactics. I really do. I just don't share that sentiment.
After this conversation, my suggestion would be to set up two Tactics ratings, one timed and one untimed. This would be analogous to the three ratings we currently have for game play: Blitz, Rapid, and Daily.
I do not plan to post further on this thread, unless someone says something substantially new.
That has been suggested a number of different times in the past. I don't recall seeing any staff input on the suggestion, in the topics I've read, so don't know if it is something they are thinking about or not.
I did not not know until reading this tip from Martin that you can wriggle out of the huge penalty for getting it wrong by just clicking the '>' tactics arrow
That has kind of spoiled it a bit for me!
I used to be in a real sweat after a minute had ticked by and I knew I was not gonna score any points but I was still gonna get clobbered if I put a foot wrong.
I vote for 'mandatory penalties' once you start the clock on a tactic even if you decline to make a move!
Don't let anybody wriggle out of it with the 'escape' button!
Then what? Wouldn't the result be about the same? Lets assume you ignore every, or most, problems that you cant figure out and move on to the next one. Wouldn't that mean that your rating would rise to the point where you cannot figure out any of the problems? Wouldn't that mean that your rating is artificially high and all or most of the problems you get are far too difficult to solve? It seems to me that would mean a string of failures which would bring your rating back down to where it should be as if you hadn't ignored the difficult problems in the first place.
Wouldn't that mean that your rating is artificially high
......................
It seems to me that would mean a string of failures which would bring your rating back down to where it should be as if you hadn't ignored the difficult problems in the first place.
Thanks for your thoughts IPG.
Yes I think your rating will be artificially high if you skip the problems you cannot do and avoid the -14 monster penalties.
If a player continues with the evasive strategy then how will they ever encounter this 'string of failures'?
I think the enforced and unavoidable penalty would add a bit of spice to the daily engagement!
After this conversation, my suggestion would be to set up two Tactics ratings, one timed and one untimed. This would be analogous to the three ratings we currently have for game play: Blitz, Rapid, and Daily.
I hear ya Koufax.
I am a daily player so I am not used to sweating under the clock either!
Untimed tactics would be good as a separate activity.
But if you want to dance with devil in the timed tactics then you wait for the music to stop, you don't wander out half way through!
Wouldn't that mean that your rating is artificially high
......................
It seems to me that would mean a string of failures which would bring your rating back down to where it should be as if you hadn't ignored the difficult problems in the first place.
Thanks for your thoughts IPG.
Yes I think your rating will be artificially high if you skip the problems you cannot do and avoid the -14 monster penalties.
If a player continues with the evasive strategy then how will they ever encounter this 'string of failures'?
I think the enforced and unavoidable penalty would add a bit of spice to the daily engagement!
I agree and dont see any downside to it other than when someone ignores a problem for reasons not associated with the difficulty, such as distractions or other things that need to be done besides playing chess. In the end though I dont think it would make much difference in a persons rating because under that rule people would probably adjust to the new rule. Such as taking as much time as necessary (15 minutes if need be), rarely getting any wrong, and lowering the problem rating because of the time it takes.
Yes, a game of chess is trying to have it both ways. You are right about that. You have to not make a mistake, while at the same time, capitalize on your opponent making one.
You are refusing to acknowledge my point. I don't disagree with any of your eloquence about the uncompromising glories of chess. I only question the assumption that the only way to do a chess puzzle is to do it against the clock.