The Schematic of Threatening Opening Preparation

Step #1: Exit Your Opponent's Knowledge Base

The logic is simple, if your opponent remembers their current position they will (probably) not blunder. In practice, until the intermediate level almost no try is off limits, at the master level finding a good try requires a strong understanding of what is and is not common knowledge, and at the 2600 level I would expect that finding anything becomes wildly difficult.

Against an intermediate opponent, there are certain well documented traps and gambits that are very likely to work

For example, 17% of Caro Kann players between 1000-1600 on lichess fall into this trap in the Tal Variation:

Additionally, in this rating range the statistically most likely variation in the Tal involves black arriving at a strategically lost position after just 8 moves

Note that at the 2200-2500 level the incidence of black blundering with e6 drops below 2%

(Luckily) Everyone knows less than you think

In general, even for strong players, in depth opening knowledge is very rare. If an opening does not fall into the category of 1. an absolute mainline or 2. a standard concrete opening trap (eg. Scotch Gambit, Tal Variation, Fried Liver etc.) most players will have a fragile understanding of the theory at best. The stronger your opponents are, the more traps they will know, but on the other hand their theoretical knowledge will lead to more predictable opening moves and therefore an easier time crafting surprises.

If you can prepare a chain of two surprises, or even lay a trap relatively deep into theory the likelihood of your opponent having concrete knowledge drops precipitously.

Step #2: Break your Opponent's Understanding

Most players will rely on three resources to find the best moves after confronted by an opening surprise

1. Contrasting the position with the mainline and deciding whether they can keep their plan intact
2. Their general chess understanding and calculation ability
3. Their built up intuition for the thematic ideas in the position

These resources form quite a solid base for making good moves in most positions, and as such most "new moves" in the opening are actually not that threatening, the value of an opening surprise is when it robs your opponent of this security and leaves them to fend for themselves. Here is an example of an opening surprise that does this very effectively.

By the end of black's equalizing line only two games remain in the lichess database filtered to 2200+. An opening surprise is an attempt to give your opponent a test that they didn't study for, by changing the nature of the position we are effectively making a mathematician write us a book report. Our central goal is to create environments where the "path of least resistance"/the human/principled/intuitive moves are eventually not the correct course of action. It is not surprising that almost all strong players react badly to white's e4 break in the example above, because what could be more natural than dxe4?

Step #3: Break Your Opponent's Psychology

What follows is an OTB blitz game I played against GM Olexandr Bortnyk earlier this year

The characteristics of psychologically threatening moves are as follows:

1. They must be unexpected, your opponent not only prepares for the moves that they expect chesswise but also psychologically. An unexpected move is a blow to their confidence
2. They must be combative, if the move does not derail your opponent's plan they have no reason to be unsettled
3. Ideally, they should be tactical, make your opponent feel like they missed something even if their position is sound
4. They give your opponent a few options (not so many that you are likely to be surprised back), there is some security in being surprised but having only one reasonable response, at least they will know that their chosen path is the right one. Not only do they have to burn time when they have options, but they lose the security that their position will be objectively sound after their choice.

This style of opening preparation is inherently time intensive and you need to have a decent level of understanding to pull it off, but it is in my opinion the absolute pinnacle of what opening preparation can achieve.

How To Come Up With Threatening Ideas Yourself

Step 1: Choose a sideline

Step 2: Find a moment where in the lichess database or according to the engine you can play an unexpected, aggressive or unusual move with good results

Step 3: Play the best moves for your side following the statistical main line until your opponent's most popular move in the database is a bad one (if the move isn't showing up in the database, see if you can figure out how to deal with the idea from your opponent's perspective, after all, if you can't they probably won't be able to do so either)