What tempo should you start Stick Control at?
Search the book for a metronome number: there isn't one. Stone's only tempo guidance is a range — from extremely slow up to extremely fast. So the honest answer has two parts: a convention, and a method.
The convention: 60 → 150
Generations of teachers converged on roughly the same plan for the Single Beat Combinations: start near 60 BPM (quarter-note click, two 8th notes per click) and consider an exercise solid somewhere around 150 BPM before moving on. For the Triplets section the same stroke rate arrives earlier — a ceiling near 110 BPM is equivalent work. These aren't magic numbers; they're a range within which the real method operates.
The method: raise tempo by ~3%, and only after clean runs
Three findings from motor-learning research shape the progression:
- Steps should be percentages, not fixed BPM. Human tempo discrimination is roughly proportional — around 2–6% of the current tempo (Drake & Botte, 1993). +3% feels like the same-sized step at 60 and at 140; +10 BPM does not.
- Aim to succeed about 85% of the time. Learning is fastest when practice is hard enough to fail sometimes and easy enough to mostly succeed (Wilson et al., 2019). If every run is clean, the tempo is too low to teach; if most runs fall apart, it's too high to learn from.
- Start each day slightly below yesterday's peak. Speed gains consolidate between sessions, substantially overnight (Simmons, 2012). Yesterday's edge is usually free today — but only if you walk up to it instead of starting on it.
What this looks like in practice
Keep a per-exercise "anchor" — the fastest tempo you've been clean at twice. Begin each session one step below it, climb +3% after each clean 20-rep run, drop −6% when you lose the pattern, and stop for the day after two failures: pushing past that point trains tension, not control. StickTempo's tempo ladder automates exactly this — it's an opt-in toggle on every exercise page, and unlike conventional speed trainers it never advances on a timer. More on that difference →