To get better at mahjong, try thinking about new theories.

If you play mahjong in real life for a long time, you'll hear all sorts of theories.
I often play with people who are quite a bit older than me, and the things I hear them say are mostly unfounded, what you might call "occult" or "Showa-era mahjong."

Those of you reading this page probably gather information online and play online mahjong.
I'd like to summarize some of the common sense that has drastically changed since I was a beginner, based on reading books written by top players of the online mahjong game "Tenhou" and actually playing in the top-tier tables of Tenhou.

In most cases, folding is the best option against a reach.

This is something I became quite conscious of after reading "The Scientific Approach to Mahjong" a long time ago.
This used to be the basis in my mind, but recently I've gotten the impression that there are a few cases where you can push even from just one step away from a winning hand.
Of course, the extent to which you're willing to push a tile depends greatly on the situation.
People who always fold when they're close to winning aren't really scary, are they?

I don't mind if the honor tiles are added later.

It's the kind of cry that makes it seem like they're about to give away what they're waiting for, so it's a sound that people tend to avoid.
However, I feel that nowadays, speed in reaching a ready hand is the priority.
Your opponent will discard when they need to, and even if you discard a tile while in a furiten state, you can simply call pon and defend with a single tile.

I don't mind if I only win on one side.

This is the same as the previous point.
Not being able to go for a quick win until you can call 78 with a 6 significantly slows down the game considerably.
You might be able to make pairs of 7s or 8s,Even if you still have 23 and 78 left, keep calling tiles aggressively to get closer to a ready hand.

(Added June 2019: However, this depends on the circumstances.)

Don't avoid Penchan and Kanchan too much.

A common tendency among intermediate players is to avoid unfavorable waiting patterns and instead discard penchan (a type of tile that creates a double-sided wait).
However, while this might be acceptable in the early stages when you're far from a ready hand, it actually slows down your progress when you're only two steps away from a ready hand.

Penchan is a pair, albeit imperfectly, so if you draw one more card, it will become a complete set.
However, even if an isolated tile is in the middle, it doesn't necessarily form a two-sided sequence, and you need to draw two effective tiles to make a set.
If a middle-side kanchan remains, it becomes a more difficult wait to complete than a penchan.
The quickest way to win is to simply prioritize getting a ready hand as soon as possible.

(Added June 2019: It's true that it would be best to sort out the penchan early on if possible.)

Tile efficiency is not just a matter of the number of tiles.

The tile efficiency method, which involves counting the number of tiles that can be accepted, is easy to study, and because the answer is clear, it's an area that's easy to jump into.
However, in reality, you need to consider not only the number of tiles, but also the points you can make, the shape of your ready hand, and whether it will be easy to call tiles.
While it's important to understand the standard number of items that can be accepted, remember that this isn't the only factor.

Even skilled players can't predict the opponent's moves.

When playing against older players, they sometimes say things like, "I knew this was the move to wait for," as if they knew all along, but no one can predict the winning move with a single point, and there's no need to.
It's natural that the winning tile would be among the tiles that aren't part of a sequence and haven't been discarded, and even if you know the winning tile, it's useless if you can't win yourself.

In Mahjong, since there are "entering tiles" (tiles that are drawn into the hand), you cannot predict the winning hand with a single point just by looking at the discarded tile used in a Riichi declaration.
Certainly, when playing with someone who's truly skilled, you can sometimes get the illusion that they can see right through your waiting tiles, but that just goes to show how accurate their assessment of dangerous tiles is.

What I'm trying to say is that since no one can read your hand, if you declare "reach," everyone will be genuinely scared.
Aim for a decisive early win with a ready hand or at least two pre-emptive calls.

Even if the player to the left is a dyer, they don't wring the paper.

There's a common misconception that calling a tile of a certain color when the player to your left has a flush is a bad move, but that's not true at all.

Basically, you discard tiles according to your own convenience.
When you're not in any position to win and the player to your left is struggling, there's a technique you can use: deliberately let the player to your left call a lot of tiles to keep the other players in a stagnant position.
If you can't stop other players yourself, using other people to your advantage can be considered a valid strategy.