How To Study Miniature Games

In Part I, I listed all the reasons you should study miniature games.  Today, for Part II, I explain how to do it.  Specifically, I describe the process I've developed. I started doing this as a kid and then tweaked it as I got older.  No one taught me this; I just figured it made a lot of sense, and the results were excellent.

A fair warning: this process takes work.  It's not hard to get started, but the process has no hand-holding: it's you, a chess board and a lot of thought.  If you do it, though, I guarantee you will get results.

And if it seems intimidating, then check out my new free course on miniature games.  I walk through 12 of my games and describe the thought process, the common mistakes, the thematic maneuvers ... basically, everything worth extracting.  I tried to make it as fun as possible, and with any luck, you will finish it and be itching to go analyze games yourself.

It's also free.  Free is good.  Check it out, but in the meantime, let's dive in.

How To Train Miniature Games By Yourself

First, Get Games

Here's how I did it when I was younger.  I had access to the famous "5334 Chess" book.  It's a staple in bookstores and most libraries that I've seen, so it should be readily available.

It's a brick of a book.

This included 600 miniature games organized by theme.  So there are roughly 75 games that involve a sacrifice on f6, another 75 with a sacrifice on f7, etc.  This is enough for a lifetime; I never got through all of them, or likely even half of them.

They have minimal annotations, which is fine for our purposes.  Basically, you get the base game, then a diagram right before the main combination.  As such, you have a clue that a combination is coming, and you know that the mistake must have occurred sometime before that diagram.  Otherwise, it is all active learning.

Nowadays, I use Chessbase, so I can search its vast database and filter out games by move number.  Specifically, I search for games under 25 moves with a decisive result.  I sometimes narrow it by opening if I want to focus on just one.  I typically want to see "good" players, so I include a restriction on ratings as well.  Here's what that search field looks like:

If you don't have Chessbase, you can search ChessTempo's database for miniature games on a free account.  This seems to be the easiest free alternative: Go to that link, click on "Advanced Search" on the bottom and use these criteria at a minimum:

Edit: I have since found Caissabase, which offers a free database to download and includes instructions on how to use it.  I haven't used it, but it will definitely do the job.

Second, Study Games

Once you have some games, review them.  First, I skip to the end position.  Is it a mating attack?  Was it a combination to win material?  If so, that's worth studying.  If it's just a straight blunder, someone dropping a piece from pure blindness, I generally skip those games.  There's not much to learn from unforced errors.

Once I know how the game ends (and I decide it is worth it), I will start from the beginning and flow through the whole game, generally 1-2s each move.  I now know how the game went.  I then start at the end and go backwards.  So if it ended with a mating attack, I look at the attack; if it's a combination to win material, I start at the first part.  I want to fully understand this: I spend time checking side-variations, figuring out why it works, why other moves don't, etc.  

Don't just assume the game is sound!  Lots of attacks work in practice, under time pressure, but fail to adequate defence.  Make sure it is winning.  Check other moves and other replies.  In particular, check your opponent's checks and counters.  It's very easy for even good players to miss how an intermediate check completely refutes an otherwise sound combination.

Of course, if you are an amateur reviewing master games, the masters are stronger than you and probably right ... but probably doesn't mean always!  Satisfy yourself.  I try to reach a point where I feel I would have played the same move if I were playing.  If I'm still unsure, then I haven't analyzed enough.  If it is truly a winning combination, then I should not move on until I know it is winning.

NOW THE IMPORTANT PART.  So important I made it a new paragraph so you don't skim over it.

Once I see how the game ends and I'm satisfied I understand it, I look for the losing mistake.  At what point did the game go from saveable to lost?  I try to identify the exact moment this occurs, and then I try to find a better move.  For example, maybe Black castles straight into a mating attack.  Okay, the losing move is pretty obvious.  Don't castle into a mating attack.  But what could Black have done instead?  And how might White have responded?

I can't stress the importance of this step enough.  This is where you get all the benefits.  You start recognizing when moves are errors, and you start recognizing whether it's a small error (letting you get easy play) or a large one (letting you get a decisive attack).  By looking for alternatives, you train your decision-making, and by looking for how the game might continue, you work on calculation and evaluation.

It's crucial you do this at least once, for the losing move.  It's also beneficial to go further and identify all the mistakes, if possible.  What moves made the loser's side worse?  Did the winning side have better moves along the way?  This is excellent stuff, but consider it extra credit.  Bonus points.  You can skip it, but never skip looking for the losing move.

This entire time, you don't touch the engine.  This is all you.  Reach your conclusions, write down your thoughts if necessary, and only now double-check with an engine.  Ideally, the engine agrees with everything you said.  Woohoo.  If not, play with the engine and figure out why.  What did you miss?  And why did you miss it?

Modern engines are strong, so they are rarely wrong ... but sometimes they are too strong.  They sometimes conclude an attack is completely winning long before a human would recognize it.  Other times they dream up impossibly complex saving attempts.  There's not much we can about this.  In such cases, my goal isn't perfection but rather being on the right track.

I strongly recommend that you record these games and your thoughts somewhere.  I studied hundreds of miniature games when I was younger, but I left no notes.  That's a huge loss.  Keep a record and in a year's time you can look back and see not only a large work product, but also a wonderful collection of tactics, attacks and combinations.

How to Train Miniature Games ... with Smithy

If the above sounds like too much work, or if you are still iffy on the whole concept, then I invite you to look at two resources.  First, check out Part Three of my Chessable course, Smithy's Opening Fundamentals.  It contains 25 miniature games, fully annotated.  You will see all the lessons I am trying to draw: recognizing mistakes, taking advantage, how patterns emerge, analyzing final attacks, etc.  It's free.  It has good reviews.  I'm quite proud of it.  Check it out.

Second, if video is more your thing, you can check my new course on miniatures, Smithy's Minis.  It features 12 of my own miniature games, where I applied all the lessons I've learned from doing this type of review.  Games are organized in a logical way, with lessons building off each other, and the focus is always on the thinking, on the root cause, on making sure you understand why the right moves are right moves.  It also features games of an amateur club player against other club players, so the types of mistakes should be relevant and familiar to the target audience ... namely, other club players.

It's also free.  It's brand new, so it has no reviews yet.  Honestly, it started as an idea for a YouTube series, but NM Matt Jensen suggested it would be better as a course.  I think it turned out really well.  Please check it out and let me know your thoughts.

Conclusion

Combined, Part I and II make for a really long article, so I will keep the recap short.  

Miniature games provide many benefits.  Study them.  It's worth it.  Get a book or get a free database.  Look at some games.  Go over them.  Learn the lessons.  Look for the losing mistakes.  Strive to find better moves, and then better moves against those better moves.  Confirm with an engine.  Enjoy, and be ready for quick wins to flow in.  Maybe not right way, but they will be coming.

If you want me to be your guide, check out my Chessable course or my new Chess Goals course.  Both are free.  Both will improve your chess, and I'm confident if you apply what's inside, you will be sending me your own miniature games.

And I can't wait for that to happen.