Why single moves are not enough
Water Sort punishes locally good moves. Any individual pour that merges two same-coloured blocks looks like progress, but if it uses the space that a buried colour needed, the board dies four moves later. The cure is to plan in groups: a short sequence with a purpose, usually two to four pours long.
The five-step routine
Inventory the board
Note the top colour of every bottle, then the fragment count of every colour (see the strategy guide). Ten seconds is enough once you are used to it.
Choose a home for one colour
Pick the colour with the fewest fragments and decide which bottle will hold all four units. Prefer a bottle that already has that colour at the bottom — it needs no extra space at all.
Work backwards from that home
For each fragment of that colour, ask what sits on top of it and where that liquid can go. That question generates your real move list. If one of those answers is "nowhere", the colour is not ready yet — go back to step 2 and pick a different one.
Check the plan against your free space
Add up the units you need to park temporarily and compare with the free units available to those specific colours. If the plan needs five parked units and you have four, it fails — and it is much cheaper to discover that now than halfway through.
Execute in the order that keeps space longest
Among the pours in your plan, do the ones that free a bottle first and the ones that consume universal space last. Same moves, different order, very different safety margin.
A worked plan
1: yellow, cyan, cyan · 2: cyan, yellow, green · 3: green, green, yellow · 4: yellow, green, cyan · 5: empty
Cyan is at 1+1+2 — three fragments. Green is 2+1+1 — three fragments. Yellow is 1+1+1+1 — four. So neither is close, but green has two units already stacked in bottle 3, which makes bottle 3 the natural home for green.
Working backwards from bottle 3: one yellow sits on top of the green. That yellow needs a home. No bottle shows yellow on top, so the only target is the empty bottle 5. That single pour costs universal space, but it directly unlocks a green home, which is exactly the trade-off worth making.
- 3 → 5. Yellow parks in bottle 5. Bottle 3 now shows green on top and can receive green.
- 2 → 3. The green on top of bottle 2 joins the pair. Green is now three units in one bottle, and yellow is exposed in bottle 2.
- 2 → 5. That yellow lands on the yellow you parked. Bottle 5 now holds two yellow — the space you spent is doing double duty. Bottle 2 is a single cyan.
- 2 → 1. Cyan onto the cyan pair in bottle 1, which now holds three cyan, and bottle 2 is completely empty. You have traded a parked yellow for a whole empty bottle.
- 4 → 1. Cyan from bottle 4 completes cyan. Green is exposed in bottle 4.
- 4 → 3. Green completes green. Bottle 4 is a single yellow.
- 4 → 5 then 1 → 5 when the last yellow is exposed — yellow finishes in bottle 5.
The plan was decided at move 1 and every later pour followed from it. That is what planning looks like in practice: not deep calculation, but committing to a destination and then clearing the path to it.
When to abandon a plan
- A pour you counted on turns out to be illegal because a colour changed on top.
- You need to park more units than you have universal space for.
- The plan finishes a colour but leaves every remaining top colour distinct — that is a deadlock dressed up as progress.
Abandoning early is cheap. Use undo to step back to the branch point instead of improvising from a worse position.
Practising with the hint button
On the game page, write down your intended next three pours, then press hint. The hint is the first move of a genuine solution, so if it contradicts your plan, your plan probably needed space it did not have. Reconstructing why is the fastest way to improve.