Why I Stopped Using Online Opening Trainers

Chessable.  Chess-Tempo.  Chess Position Trainer.  These are just some of the opening trainers out there.  They all work more or less the same: you train your openings on a real board, and spaced repetition determines what positions you are to train.  They differ in features but essentially serve the same niche.

I spent 3+ years using Chessable.  I spent another year on Chess-Tempo.  And now I've stopped.  Completely.  This post will explain why. The short answer: these don't provide enough value for the time investment.  The longer answer... well, read on.

The Time Cost

Let's start here, as it's the reason I quit these cold turkey.  It ultimately comes down to the nature of spaced repetition systems.  In theory, you are supposed to recall something right before you would have forgotten it; this puts it deeper into your long-term memory, and you can go progressively longer before you need to recall it again... but it always comes back up again.  Things never graduate or retire; they always come back.

Doing it correctly also means doing it every day.  If you miss a day, then you have missed the ideal recall time for a group of moves.  That throws the schedule out of whack and likely defeats the purpose of spaced repetition.  In short, if you want to reap the benefits, you need to commit time daily.

We can see the problem emerging: you train everyday and things never go away.  Now add into the mix learning new material.  The more you learn, the more you have to train everyday, and the more items that will never go away enter your schedule.  By design, this is unsustainable longterm.

I thought it was fine.  At my height, I logged into Chessable 4-6 times a day, generally for 5-15min at set times: when I wake-up, morning break, lunch, afternoon break, before dinner, before bed.  No session felt that restrictive, but it adds up.  How much?  Here's my four-month snapshot.

As a legal professional, I have software that tracks my time on various sites for billing purposes.  In 2023, I used it to track some personal stats, just for fun.  This was my Q1 total.  When I saw this, I stopped the next day.

5min here, 15min there doesn't seem like much, but it adds up.  Seeing this total flabbergasted me, quite honestly.  Now, by itself, this isn't necessarily a bad thing, and it's just one side of the equation.  The question becomes, are the results I am getting worth this time investment?

My Terrible Results

During this time, I primarily played blitz on lichess.  I've spent a huge amount of time learning the Sveshnikov and Classical Sicilians.  Some of these lines approach move 30.  Let's see my games.  Out of 549 games starting with 1.e4 c5:

• Sveshnikov 2...Nc6: 257 total games
• Open Sicilian: 157 games (61%)
• 6. Nb5 (by far the main move in the course): 79 games (30%)
• 7.Bg5 (the start of the mainline): 49 games (19%)
• Classical: 2...d6: Only 42 games
• Open Sicilian: 31 games (74%)
• Rauzer 6.Bg5 (by far the most critical): 9 games (21%)
• Silly 6.Be3 (barely any trainable coverage): 8 games (19%)

Forget move 30, I am not even hitting move 7!  In only 20% of my games, 1 in 5, am I reaching the main tabiya.  This is literally the starting point of these Chessable courses.  The vast majority of the time, my in-depth opening knowledge is useless.

Worse than useless, actually.  Subjectively, I still felt uncertain about what moves to play.  Can you imagine that?  Investing 100+ hours into something, not feeling like I know it just a few moves in.  That is completely unacceptable. When I learned most of IM Dembo's Grunfeld book back in the day, I invested far less time and had much better knowledge, recall and subjective confidence in my moves.  Granted, I was playing correspondence at the time, not blitz, so I didn't feel the same time pressure to make my moves quickly, but my conclusion stands: Chessable is much easier than a physical book ... but not more effective.  Not even close.

Why?

I can think of several reasons. First, studying on these platforms is nothing like playing in a real game.  Chessable has some pretty glaring technical limitations, but this extends to all of the options I've tried. Fundamentally, these simply train you to react to moves.  You don't think about your opponent, you don't look at their plans or ideas; your focus is solely, 100%, on your own rote replies.  You get very good at shooting off several moves in the blink of an eye, often without fully knowing if there are other reasonable replies for either side.  It's not programed in, so it doesn't matter ... until it comes up in a game, and it does matter.

Said another way, because you only make your own moves, you tend to ignore the opposite side.  I've reached the point where I knew how to react to moves XYZ, but show me any given position and I wouldn't know what my opponent's best moves are.  By design, these programs only teach you half the opening; you learn your side, but not your opponent's.

Second, reviewing moves is not the same as reviewing the plans, the ideas and the structures of the opening.  Rote reviewing the moves does not engage these more fundamental concepts, and it often feels like this gets lost in the shuffle.  Knowing that 7.Bxf6 is the best move doesn't do much if you can't explain why it is.

Third, their strength may by a weakness.  As they say, easy come, easy go.  With a click of the mouse you can load a variation and speed through all the moves.  I know I've definitely had times where a position came up and I had no memory of ever seeing it before, let alone learning it!  Things happen so quickly, so effortlessly, that you might not fully turn on your brain.

Compare this with a book and a real board.  I set up the pieces and move them myself.  When I explore a side variation, I have to manually go backwards and re-set the board.  Often this means going all the way back to the beginning and starting over.  If you explore 3-6 side variations like this, that means you have 3-6 more repetitions through the beginning moves to reach that point.  You naturally get more exposure, and it's no surprise I remembered it so much better.  Books are harder, or at least less convenient, but that very fact lets it sink in that much deeper.

Finally, all of these programs confuse "doing something once" with "knowing it."  The entire premise of spaced repetition is to prompt recall of something right before you would normally forget it.  This is entirely useless if you didn't learn something in the first place.  In many cases, I would go through a given variation and not understand all the details.  I would benefit from going through it again, probably several more times ... but the UI of all these sites discourages this.  Very easy to go and "learn" new material; very hard to repeat the line you just did.

This seems completely opposite of how we learn ... well, basically everything else.  Imagine you are learning a new language and there's a whole list of vocabulary to learn.  Are you going to read each item once and then try again 4hrs from now?  That's the default Chessable mode, and it's silly.  No, you would go several times through the list, testing yourself along the way, making note of which ones aren't sticking and trying them again.  Maybe you make some mnemonics to help, or you might handwrite the entire list out just for practice.  After you do all this, after you are fairly comfortable with the material, does it make sense to enter spaced repetition mode.

None of the programs operate like this.  They simply assume you saw something once and so you know it.  That's silly.

Conclusion

I stopped using opening trainers and reclaimed dozens of hours each month.  My opening knowledge has taken only a modest hit, and it would take just a small amount of work to get back to normal.  I don't miss the grind at all, and I will never go back.  Ultimately, I think all of these platforms are fundamentally broken: they give the illusion of progress.

I am not necessarily advocating that other people should stop using these sites.  I am merely documenting my experiences.  If yours match mine, you might consider stepping away.  I certainly have no regrets.