Pacing and Your Practice Loop

The cheapest section on the test

Ninety items in sixty minutes works out to forty seconds each — Foundations does the whole-test arithmetic; here it is just the number to beat. Angle ranking is fifteen of those items, and once the method is automatic it is the cheapest fifteen you will meet.

The reason is structural: there is nothing to build here. You are not assembling a solid from flat views, not folding a pattern in your head, not counting hidden cubes. Every item is one repeated comparison — which of these two openings is wider — run against four choices that have already narrowed the question. When that comparison is automatic, an item takes far less than forty seconds.

The surplus is the point. Time you do not spend here is available to the sections that genuinely need it, and the way to bank it is to be quick on the items that are already easy for you — not hasty on the hard ones.

Where the seconds should go

A working target is twenty to thirty seconds an item, with the shape of the spend fixed:

• A few seconds reading the four orderings, to see which comparisons are contested.
• The bulk of the budget on the extremes, because that is the comparison with the most margin and the most eliminating power.
• Whatever is left on the middle pair, and only if two orderings are still standing.

If your time is going the other way round — a glance at the extremes and then a long stare at the middle two — the budget is inverted, and it will show up as misses on items you should have banked.

Watch for the specific way this section wastes time: re-checking. Almost nobody loses a minute deciding; they lose it re-deciding — same two panels, doubt, look again. A second look at a comparison you already made adds nothing. The openings have not moved.

Triage is one rule

An item that resists the extremes pass gets an answer, a flag, and no more time.

That is the entire policy. If you cannot separate the smallest from the largest — not the middle two, the extremes — the item will not yield to another twenty seconds, because you have already spent your best comparison on it and come away empty. Pick the most plausible ordering still standing, flag it, and move.

Never leave an ordering blank. By the time you are ready to give up you are almost never guessing blind: kill two orderings on the extremes and a guess between the survivors is a coin flip, kill even one and you are at one in three. Both beat the one in four you started with, and all of them beat the zero you get for leaving it empty.

The drill loop

Sets of ten, at a difficulty you choose, from the angle ranking practice generator. Ten is the right size: long enough for a pattern in your misses, short enough that you still remember the third item when you review it.

The step people skip is the second one. After you answer, the explanation prints the true size of all four angles in ascending order — read it every time, including on the items you got right. It is the only way to find out whether the pair you fumbled was three degrees apart or thirteen, and those two misses mean opposite things.

Log the cause, not the item

Be clear about what a wrong answer is in this section, because it is not what it is elsewhere on the test. The four choices here are orderings — sequences of the same four numbers — not drawn shapes. Nothing was constructed to trap you; there is no family of tempting wrong figures to learn to recognize, because there are no wrong figures at all. Every wrong choice is simply a sequence that disagrees with the truth about which of two angles is bigger.

So a miss here is always a comparison you got wrong, and the tally has to name the comparison. Three causes cover nearly all of them:

• Length illusion. You ranked by how far apart the ray tips ended up rather than by the opening at the vertex. The signature is a miss on an item with wide gaps and lopsided ray lengths — the differences were well within your reach and you were reading the wrong quantity.
• Orientation. The two angles were turned to different headings and you misjudged which was wider once you could not line them up. The signature is a miss where the two angles you swapped were drawn pointing very different ways.
• Middle pair. The extremes were right, two orderings survived, and you lost the last binary call. The signature is a swap of two adjacent angles a few degrees apart.

They point at three different fixes, which is the whole reason to separate them. Mostly length illusions means you are judging at the tips and should go back to comparing near the vertices. Mostly orientation means drilling at a level where you still have margin while the habit sets. Mostly middle-pair losses is not a method problem at all — that is your acuity threshold showing, and the fix is separation practice, not more care.

Keep one more entry even though it is not a perception fault: selected the wrong ordering. You knew the answer and clicked a sequence that read almost the same. Two or three of those in a tally is a real finding, and no amount of angle practice will fix it.

The takeaway

Forty seconds an item is the test-wide average; angle ranking should cost you well under it, so the surplus can go where it is needed. Spend the budget in fixed proportions — a glance at the orderings, most of it on the extremes, the remainder on the middle pair if two choices survive — and stop re-checking calls you already made. An item that resists the extremes gets an answer, a flag, and no more time; nothing is ever left blank, because elimination has already beaten a blind guess. Drill in sets of ten, read the printed degrees after every item, and log misses by the comparison that failed. There are no distractor shapes here to recognize — only comparisons you made correctly or did not.