Snake looks like the simplest game ever made. It is not. Here is how the scoring works, why most runs end the same way, and the three habits that separate a 20-point run from a 200-point one.
Almost everyone has played some version of Snake. It arrived pre-installed on Nokia phones in 1998 and became, for millions of people, the first video game they ever played on a device they owned. The rules fit in one sentence: eat the food, grow longer, don't hit anything.
What makes it worth writing a guide about is that the difficulty is entirely self-inflicted. The board never gets smaller and the snake never gets faster in most versions — you are the obstacle that grows. Every point you score makes the next point harder to reach. That's an unusual design, and it means good play is less about reflexes than about the shape you leave behind you.
Every Snake death is one of these, and knowing which one keeps killing you tells you what to fix:
The instinct when the snake is short is to cut straight across the middle of the board toward each piece of food. It's the fastest route, and for the first ten points it works fine.
The problem is what it does to your shape. A snake that has been criss-crossing the centre ends up as a tangle of loops through the most valuable real estate on the board. When you're 60 segments long and food appears in the middle, you'll have nowhere to put yourself.
Instead, spend the early game travelling along the perimeter. It costs you a few seconds per point, but it keeps the centre of the board completely clear — and the centre is the only place with enough room to manoeuvre when you're long.
Strong Snake players don't improvise a fresh route for every piece of food. They settle into a repeating circuit — usually a boustrophedon pattern, which is the technical name for the way an ox ploughs a field: down one column, back up the next, all the way across, then return along the edge.
The advantage is that a repeating path produces a predictable body shape. Your tail always vacates the space you're about to need, because you laid it down in a known order. It's slower than opportunistic play, but it's the only approach that scales to very high scores, and it's what most Snake-solving algorithms converge on too.
You don't need to plough the entire board every lap. A simplified version that works well in practice:
The single most useful habit in Snake is asking one question before every turn into a confined area: if I go in there, how do I get out?
Self-trapping happens because the snake enters a pocket that was big enough on the way in, then discovers the exit closed behind it as the body kept feeding through. Since your tail only moves forward when the head does, entering a dead end with a long body is genuinely unrecoverable — no amount of reflex saves it.
If the answer to "how do I get out" isn't obvious in under a second, don't go in. Circle and take that food on the next pass.
It depends on the version. Some implementations increase speed at score thresholds; others hold a constant tick rate and let length alone provide the difficulty. If you find your inputs are being dropped as the game speeds up, it usually means you're queueing turns faster than the game processes them — press once per intended turn rather than mashing.
Games that survive for decades usually share one trait: the rules are learnable in seconds but the skill ceiling is nowhere in sight. Snake qualifies. There's no random element to blame, no hidden information, no opponent — every death is traceable to a decision you made several seconds earlier. That's uncomfortable, and it's exactly why people keep starting one more run.