We quantitatively investigated the clinical severity and progression of diseases with ataxia, as measured with the Scale for the Assessment and Rating of Ataxia, and examined the potential application of the Scale for the Assessment and Rating of Ataxia for future therapeutic trials. Severity of ataxia was assessed in 238 patients with spinocerebellar ataxia type 2, spinocerebellar ataxia type 3, spinocerebellar ataxia type 6, spinocerebellar ataxia type 17, multiple system atrophy-cerebellar variant, or Gerstman-Sträussler-Scheinker disease. Among them, 119 (50%) were longitudinally examined three to seven times, in a period of 8 to 38 months, resulting in a total set of 535 assessments. The differences between spinocerebellar ataxia and multiple system atrophy-cerebellar variant were ascertained cross-sectionally and longitudinally. Gerstman-Sträussler-Scheinker disease had the fastest progression, followed by multiple system atrophy-cerebellar variant, spinocerebellar ataxia type 17, spinocerebellar ataxia type 3, spinocerebellar ataxia type 2, and spinocerebellar ataxia type 6. Patients with multiple system atrophy-cerebellar variant had a faster progression in gait, sitting, speech, and total score than patients with spinocerebellar ataxias. For a randomized, case-control trial, a sample size of 47 for spinocerebellar ataxia and 85 for multiple system atrophy-cerebellar variant in the treatment or placebo arms would have a sufficient statistical power to demonstrate the efficacy of a new therapy that would retard ataxia progression by 1 point per year as measured by the Scale for the Assessment and Rating of Ataxia. The results will have a significant impact on the planning and implementation of future therapeutic trials of spinocerebellar ataxia and multiple system atrophy-cerebellar variant.