Summing Up Hours of Any Type of Practice Versus Identifying Optimal Practice Activities: Commentary on Macnamara, Moreau & Hambrick — and Macnamara et al.'s Reply

Methodological exchange over the 63 sports effect sizes (N=2,765)

Exchange in Perspectives on Psychological Science 11(3). Ericsson's commentary (pp. 351-354) presents NO new data: it argues the sports meta summed virtually any sport-specific activity (group training, watching games on TV, play, competition) rather than narrow teacher-designed practice with immediate feedback. The reply (pp. 355-358) re-runs the meta on the subsets Ericsson demanded.

The definitional defence does not survive its own test: every restriction Ericsson demanded leaves the estimate between 17% and 22% — still a small minority of variance. Macnamara et al. also document that Ericsson repeatedly cited the very studies (Helsen 1998; Hodges & Starkes 1996) he now says should be excluded, and called amateur bowlers 'experts' while rejecting amateur-containing samples elsewhere. Their charge is that the DP construct is applied inconsistently and is therefore unfalsifiable. Ericsson & Charness (1994) had set the bar for 'expert' at merely 2 SD above the population mean.

Ericsson dismissed the Mosing (2014) and Hambrick & Tucker-Drob (2014) twin evidence as not generalizable to 'expert musicians'; the reply shows this is inconsistent with his own usage of 'expert'. Rejected on definitional, not empirical, grounds.

Escalated: Ericsson & Harwell (2019, Front Psychol 10:2396), Macnamara & Hambrick (2020, Psychological Research), Hambrick et al. (2020, Front Psychol 11:1134), Ericsson's posthumous 2020 reply. No decisive test exists.