How to test your pool skill level

Why “how good am I?” is so hard to answer

Match results measure your opponents as much as they measure you. A good night against a weak field and a bad night against a strong one can produce identical scorelines. Practice feels productive whether or not it is working, because you naturally spend it on shots you already like. And the drills most players run are the ones they have run before — which is exactly why they get easier.

A useful skill test has to do three things that casual practice does not.

1. It has to measure a whole skill, not a single shot

Potting a long straight blue tells you about one stroke. Clearing a table tells you about pattern reading, position play, speed control, safety judgement and nerve, in the order the table demands them. That is why the run-out — the whole table, start to finish — is the right unit of measurement for a skill test, and why Runout Rank scores tables rather than shots.

2. It has to be unpredictable

Any fixed set of layouts decays into a memory test. The tenth time you set up the same drill you are not measuring run-out ability, you are measuring how well you remember the answer to that particular table. A test worth repeating has to generate its layouts, so the pattern in front of you is genuinely new each time.

3. It has to be repeatable in the same form for everyone

Here is the tension: randomness makes a test honest, and it also makes two scores incomparable. If your ten tables were harder than mine, our scores mean different things.

The fix is seeding. The layouts are generated by an algorithm from a fixed starting number, so they are unpredictable to the player but perfectly reproducible for the program. In Runout Rank, the seed is the level plus the test number: Level 3 Test #12 is the same ten tables for every player, on Android and iOS alike. You cannot practise the answers, and you can still compare your score to anyone else's — without an account, a server, or a leaderboard.

The rules that make a score honest

• One attempt per table. Best-of-three measures your best day, not your standard.
• No skipping. The layouts you would rather avoid are precisely the ones carrying the information.
• A fixed number of tables. Ten is enough to average out one unlucky roll, short enough to finish in a single sitting at a real table.
• A stated pass mark. Seven out of ten clears a level in Runout Rank. Knowing the bar before you start is part of the test.
• Record it immediately. A result you write down an hour later is a result you have already flattered.

What to do with the number

A rating on its own is trivia. The number is only useful if it points somewhere, which is why the important output of a test is not the score but the edge level — the rung you cannot yet clear. That is where practice pays, because it is the only level where the layouts are still asking you a question you cannot answer.

The practical loop looks like this:

1. Test at a level you think you can clear.
2. If you clear it, test the level above until one beats you.
3. Practise at that edge level, logging attempts so the run-out rate is real.
4. Retest the same level when the rate has moved. Compare the ratings, not the feelings.

How often to retest

Often enough that the number tracks reality, rarely enough that each retest reflects actual work. For most players putting in a couple of table sessions a week, every two to four weeks is about right. Retesting after every session mostly measures noise; retesting twice a year tells you nothing you can act on.