Timed Drills: When the Clock Helps and When It Hurts

My son cleared twenty addition facts at the kitchen table without a wobble, until I said the sentence every parent eventually says: I bet you can beat two minutes. His pencil sped up, his shoulders rose, and he missed three facts he had known for a year. Nothing about the mathematics had changed. Only the clock.

A timer is the most powerful and least understood tool in elementary math. Aimed at facts a child already owns, a short sprint builds the automatic recall that frees up working memory for harder problems. Aimed at facts a child is still learning, the same sprint teaches a child to hurry a thought they have not finished having. The difference is not how strict you are. It is whether you checked first.

The gate: two numbers, measured with no clock running

Before a timer appears, run the set untimed and count two things. The first is accuracy. Twenty facts, pencil only. If fewer than nineteen come back right, the set is not ready. Below roughly 95 percent, a timed drill stops measuring fluency and starts measuring the gap, in the most stressful way available.

The second number is retrieval time, and it is the one most families skip. Watch how long a correct answer takes. A fact that arrives in a second or two is stored and ready. A fact that arrives in five or six seconds, with a finger count or a whispered rhythm underneath, is being rebuilt from scratch every time. Timing a rebuilt fact sounds reasonable and behaves badly, because guessing is faster than rebuilding. You do not get a faster fact. You get a coin flip with the answer already selected.

So the gate is boring on purpose: about 95 percent accurate, mostly under three seconds, measured without a clock. Only then is a timer measuring something real.

Sixty seconds, twenty facts, one attempt

Once the gate is passed, keep the drill too small to be dramatic. This is the version we settled on after the two-minute disaster, and it has held for two and a half years:

• Same twenty facts every day for a week. A new set and a new week start together, because changing the questions changes the score and hides the progress.
• Sixty seconds, not five minutes. Fluency work is a sprint. Past about ninety seconds, the last stretch measures endurance and frustration rather than memory.
• One attempt per day. Running the sprint three times in a row turns a personal record into a penalty, and the third run is always worse.
• The score goes on a sticky note on the fridge, compared with the child's own last score and nothing else.

The sticky note is doing more work than it looks like. Seventeen on Tuesday and nineteen on Friday is evidence, and evidence is motivating without being personal. “You need to be faster” is only personal. Two or three weeks of notes is also the honest answer to whether the drill is helping: if the number has moved four or five times and then stopped, the set is done and the timer comes off. Spacing the same set across days rather than piling attempts into one evening is what makes that curve rise, which is the whole idea behind spaced practice.

What a timer cannot do

A timer cannot create a fact that has never been learned. If the seven-times table is still a mystery, a stopwatch makes the mystery urgent. That stage needs counters, a number line, a game, and someone talking through the structure of the fact, which is exactly what a fluency built on meaning looks like. The clock belongs to the next stage, after the fact exists and needs to become cheap to retrieve.

A timer also cannot tell you whether a child understands anything. It records decisions per second, and understanding is not a decision. When the goal is diagnosis, an untimed page with the work shown is worth ten timed sprints, which is the argument behind making practice calmer by default and adding speed only as a separate, labelled event.

The strongest evidence against careless timed testing comes from the children who were never slow in the first place. In Fluency Without Fear, Jo Boaler and colleagues at Stanford's youcubed project report that for about one third of students the onset of timed testing is the beginning of math anxiety, and that at least a third report extreme stress around timed tests (Boaler, 2015). Debating the exact fraction misses the point. The clock lands hardest on the children who cared most about getting it right, and that is a strange thing to build into a daily routine.

Why strong students are not safe

The intuitive guess is that pressure only damages the children who were already behind. The experimental finding runs the other way. Sian Beilock and Thomas Carr had students solve a set of math problems, then added the ingredients of a timed test: a reason to do well and the feeling of being watched. Accuracy fell, and it fell furthest among the students with the largest working memory (Beilock & Carr, 2005). The children who can hold an entire procedure in their head have the most to lose when pressure eats that head-space, because the procedure lives there and nowhere else.

That is why “it is only a game” does not undo a timer. Games still have winners, and a child who is being watched while counting is not playing. It is also why a single classroom time limit can misfire for a bilingual learner converting the question while solving it, or for a child who needs a few extra seconds because of attention rather than knowledge, which is a large part of what math anxiety actually feels like from the inside.

How I run it now

The timer sits face-down on the table and my son turns it over himself. He chooses the day off; Sundays are his, and I do not renegotiate. Two sloppy runs in a row means the clock goes in a drawer for a week, with no lecture attached, and we go back to the untimed set. Twice, that has turned out to be the right call: the facts were not as solid as the score suggested, and one week of quiet work moved the number more than three rushed attempts would have.

Four days a week is plenty. A drill that runs every single day for a month stops being a game, and the number on the sticky note stops moving at the same time.

I also stopped shrinking the clock and growing the set in the same week. Pick one change. A timer that gets tighter while the questions get harder is not a drill, it is a moving target, and a moving target is what makes children stop trusting their own recall.

The last rule is the one I would keep if I could only keep one: the child can always ask for the timer to be turned off, and the answer is always yes. It has been used exactly once. Knowing the exit exists is what makes the sixty seconds survivable.

If you only change one thing

Move the clock to the end of the process, not the beginning. Teach the fact, practise it untimed until it is accurate and quick, and only then, for two or three weeks and never as a grade, let it be timed. A drill that follows mastery builds fluent recall. A drill that replaces mastery builds a child who answers fast and is not sure why.