How much tactic training should you do per day?

That's what I like about Chesstempo tactics wise, not a social site so your ratings are pretty much private.

The more the better, but also keep in mind that very often, tactics are NOT the main limiter of your rating.

I suspect it's actually more common at the class level to make so many positional errors and positional blunders that losing tactics are just super easy to fall into. But looking at the tactic and thinking you could have studied more tactics to avoid it isn't the right solution - the solution is to understand the positional elements so you won't even get into that sort of risky situation in the first place.

Strong players play games where it seems like their opponents have all the blundering/losing chances, and they almost have to intentionally drop pieces to lose. Often times, they can even blunder pieces, but their position is so strong that they win anyway.  

If you're doing tons of tactics but your rating isn't going anywhere, this is likely a big part of why.

Saw an interview with Magnus recently where he stated that all sub 1800's blunder somewhere in their game.  He suggested tactics as a solution and volunteered that he still did 3 to 4 hours a day.  Susan Polgar does an hour.  Obviously if folks like this still find it invaluable, we should make time for it also.  Except for beginners who need tactics training just to stop hanging pieces, we of course need a balanced approach, but it is interesting to see that even elite titled players use tactics to stay sharp.

Depends on how much time and effort you want to invest.  

What do i do?

I will do a couple hundred a day reviewing known patterns.  Kinda like a refresher, and a good warmup.

When it comes to learning new tactics, im a firm believer in David Preuess's asessment that once you miss 3 tactics you stop.  I then thouroughly learn, and understand the 3 i missed.

Just my opinion.  Good Luck.

Yes, I agree with that. Preuss's advice to stop when you *miss* three puzzles and reflect on those.  This is not my area of medical expertise, but from reading on the subject of learning physiology, I would add that this does not need to be a 24hr day, but per session with a reasonable break to rest between to allow transference to long term memory. 

I don't agree with that Pruess advice at all.  

Invariably, if I do tactics that are 'too easy' for me, my play suffers. My thinking process gets lazy, and I start not looking for deeper complex tactics or even tactical elements. 

The best puzzles for me are ones that are slightly too hard so I can't really get them right without the solution. If I spend even 5+ minutes trying to solve one of them in my head doing all the calculation, that's def higher yield for me than banging out hundreds of 'too-easy' problems.  

I don't think you're improving your tactics at all if you quit after missing three puzzles.  That's like quitting free throw practice in basketball after missing three shots in a row. The solution isn't to stop - it's to analyze what's going wrong, and try and fix it by more practice.

The advice is not to 'quit', but to stop and reflect.  I guess 'pause' is a better word.  Otherwise, as you say, the person would just be making the same mistakes over and over which is inefficient.

The research is based on how many new patterns you can learn in one session.  During the session the new patterns are stored in short term memory.  They get transferred to long term memory during 'down time' (usually a good sleep, but any prolonged period of not taxing your brain I think will work).

If you overtax short term memory, a lot of it gets dumped due to lack of working memory to hold it, before it gets to long term memory.  Again, this is innefficient, as you will forget a lot of what you learned and need to relearn it.  

That's the theory as I understand it - but perhaps I read too much about it at one time and didn't retain enough in long term memory LOL.

Its like reading.  You can read fast and not comprehend much, or you can read slow and really absorb what youre reading.  

There is something about tactics, and openings where some feel quantity over quality is supposed to be impressive.  

But whatever works for you.  I just know what works for me.

there is more to to playing lots of tactics IMHO, than merely "learning" new patterns.

you could be

• building your ability to focus for an extended time
• disrupting your normal tendency to disregard moves that hang peices or look counter-intuitive
• practicing calculation
• trying to understand what Kinds of tactic pattern you struggle with (espacially if you write this down)
• building an appreciation for and enjoying finding a tricky win in a mind-numbing puzzle

myself, I find the developing long-term focus to be an important and little recognized benefit of extensive tactics training. 

whens the last time you lost a peice because you were getting tired of playing chess and wasn't putting the best into your game?  

you can work on that with the puzzles without the sting of defeat...

I take a cue from Botvinnik.  I practice with the TV on as it helps me block out external noise.  I know a guy that practices only when its dead quiet, and yep you guessed it.  At tournaments hes "that guy" that is bothered by every little noise.  

We jokingly tell him that he can probably hear a gnat fart from 500 feet away.  

A lot.

I feel like we're not giving the OP a strait answer- but on the other hand This isn't so easy

Tactics training is just one part of a balanced diet for the improving chess player, and you can and should sometimes make a big deal out of it, and perhaps at other times, just a minimal effort to keep it in your head.

When you ask, however, how many HOURS a day- you I think assume to much. its not a neccesary part of every Day although it absolutely should be regularly done.

its also very neccessary to know how much you train a week - 5, 10, 20,40,perhaps 60 hours?  I don't know.  10-15 is where I stand on most weeks.  and at this point it is less than 5 hours (for sure) per week on tactics.

but when I first got into tactics training- It was nearly the Only thing I did to improve in chess;  play games and do tactics puzzles.   I could see a 10 hour week of that.  but I wouldn't recommend you keep it that intense for week after week.

--

so to be exact, and assuming a 15 hours week (and being honest with the hours); 

• 9hours a week; 2 hours 4days a week and a day of 1 hour.
• (after several weeks) 4.5hours a week; 1 hour 4days a week and a day of :30
• (maintain)  3hours a week; 1 hour 3 days a week

thats my recommendation. its a lot but I think there's a lot of progress to be made (espacially at first) in really Deeply and quickly recognizing tactics.

And that 'limited short term memory' theory is likely bunk. 

If our brains were that puny, we wouldn't be able to accomplish much. College and postgrad school made it pretty clear to me that I could absorb entire textbooks, pretty quickly, with reasonable time and a lot of effort. 

I could do the same with chess if I had similar time, but alas, I don't - chess is like last-priority in my life events, unfortunately. (Family, then job, then fitness/health, then home tasks, and only after all the important stuff is done, then I get to dabble in chess!)

 Anybody who looks at my tactics stats can see I spent an awful lot of time doing tactics.  That's what happens when you're retired.

 I have over 800 hours and counting!

Scientific knowledge is just the 'least wrong idea so far', in the words of Prof Brian Cox, but I doubt the theory is 'bunk' considering the amount of evidence for it.

Working memory has had a lot of research done on it, and it is a key skill in chess (for visualisation of analysis trees), it is absolutely limited and starts showing significant decline after age 26 or 27.  

The amount of short term memory that is transferred to long term memory is pretty well researched as well, and is also limited.  That's not to say we don't have good short term memories: when I was younger I had an almost photographic memory and could cram a textbook in a few hours.  However, I could only remember a fraction of it after a week or two. 

Deep learning, to reach mastery, of a subject requires repeated transfer of information from short term memory to long term memory, I believe over a period of time.  

Sorry, but I think that's inappropriate.

Sure his standard chess average opponent is low, but he has only played 38 games. has beaten higher rated opponents, and has a FIDE rating over over 1800.  Also, his bullet and blitz ratings are very high and his average opponents in those are high too.

Please have another look at his stats and reflect.  I'm sure you don't want to come across like you are bullying the guy.

Note two things:

1) "Absolute best" is not a real person. There are plenty of trolls and PLENTY of them on this site. Some are claiming it's the A.L.I.V.E. infestation back again, who knows anymore and who cares... really. 

2) Unless you're preparing for an event or something, training for tactics is a waste of time. They will go in one ear and out the other. You only need to see tactics as they come up as part of normal study, otherwise I don't see the point.  

alexm2310 don't talk to me. I was giving my advice after being asked for it. Your anecdotal examples aren't worth much, I'm not going to counter your flimsy arguments about what players "should" do. 

The entire "tactics training" fiasco has been a disaster and a failure from the start, which is what I've been saying for years. All of those "knight's errant" guys ended up a complete failure, unable to grasp even the basics of proper chess play. de la Maza guys even worse. Training tactics a lot is not only useless, it's also draining and it's inhuman. It's not fun. I don't want to play against people who "study" and drill like machines. No titled player that I've seen has ever gone on youtube and said "okay let's do tactics puzzles".