Chang claims that slow practice is valuable, just not so slow that you change the "gait" of your motions.

Walking isn't the same motion as running; we wouldn't practice running by walking. But not all running practice is a sprint.

He says that it's a mistake to learn pieces too slowly and try to speed up from there precisely because of this "gait" issue.As you speed up, you have to unlearn the wrong motions that do not work at full speed and replace them with different ones.

I find it helpful to play the big phrase choreography—the shape of a phrase, making it a physical gesture with big muscles, actual exact notes be damned. Then add in notes, making it gradually more precise. But for anyone to say that slow practice is a mistake ever is just nuts! Slow practice is how we find hand and arm shapes and angles, how we listen to the sound- esp. for a non button instrument like cello. A musician who really stretches themselves toward greatness tries all the gaits at all the speeds and all the choice emphasis and every wacky thing they can think of to find something of truth in the music. Sometimes we improvise something harder and easier. A great musician is their practice- they are as fascinating as their practice methods- and the practice should be fascinating. The thing is to find one’s own method, not to do anyone else’s method. If one reaches for artistry in music, their practice must be innovative, stylistic, nimble and responsive to everything: training for flexibility. Training for facility is more like what is mentioned here- not bad, but if you want computing, computers are better suited!

economy of motion works wonders, but even then sometimes there are positions which have the same economy but may be dramatically easier at speed, and you need to play closer to tempo to learn that.

Sometimes it's even better to trade away lots of small adjustments to avoid one big one, because you can't hit the big one at tempo.

a lot of drummers are particularly atrocious at low tempos. me included.