Steps become part of the game
Pocket Pikachu and the Pokéwalker preceded mobile apps. Their principle connects real movement with a companion or in-game progress. The connection is easy to understand: a walk can produce something beyond steps on a counter.
The Pokéwalker also helps distinguish two questions often brought together. Does a device count steps accurately? Does using it lead people to walk more? In 2011, an Iowa State University team reported an accuracy comparison involving 22 children and eight adults at different treadmill speeds. This work concerns instrumental measurement. The announcement said that behavioural results were still being compiled. Calling it an object that had “never been evaluated” would erase the first question; attributing a lasting benefit to it would go beyond the findings presented. Pokéwalker accuracy comparison in 30 participants, 2011.
With Pokémon GO, walking becomes part of a map and organised meet-ups. Players move to encounter creatures, reach a location or take part in an activity with others. The game thus provides several reasons to go outside, whose importance varies between players.
How many steps, and for how long?
Howe and colleagues followed 1,182 young adults, including 560 players. They reported approximately 955 additional steps per day in the first week after installation, with a confidence interval of 697 to 1,213. In this cohort, the difference returned to baseline around the sixth week. The result describes a launch effect in specific participants, using phone data reported in a survey. Step follow-up in 1,182 young adults, Howe, 2016.
Althoff and colleagues studied data from approximately 32,000 Microsoft Band users. The often-cited increase of 1,473 steps per day applies to a subgroup of highly engaged players over thirty days. It is not an average applicable to all Pokémon GO users. Players were identified partly through their online searches, adding uncertainty. Analysis of wearable data and GO participation, Althoff, 2016.
In Yokohama, Hino and colleagues compared 46 players with 184 matched non-players, all aged at least 40. Results notably suggest a relative maintenance of walking during winter. The overall interaction was significant during three of the eight months after launch. The signal lasting up to seven months concerns certain subgroups, including people aged 55–64; it does not demonstrate a uniform effect in all older or sedentary people. Comparison of 46 players and 184 non-players in Yokohama, Hino, 2019.
These studies support discussing an increase in walking in several settings. They invite separate examination of starting the game, regular participation and what remains after the initial enthusiasm. An average return to baseline can coexist with some people continuing for a long time and others stopping quickly.
What happens when the game is offered as an intervention?
A Drexel research thesis defended by Sun in 2024 provides a useful comparison. Thirty-eight people were recruited and randomised, and 32 completed the four weeks. Both groups received a Fitbit, step goals and support; one also received Pokémon GO. Superiority of the GO group in steps at week four was not statistically demonstrated.
The small sample, below the planned target, and missing data limit the result’s precision. A non-significant difference does not prove that two interventions are equivalent. The protocol measures the value of adding GO to an already active programme, rather than the difference between playing and making no changes to daily life. Sun, university thesis, 2024.
Several studies in adolescents have used apps in schools. A 2025 analysis covers 240 students allocated across four apps, including 59 to GO. Of the 29 initially inactive students assigned to GO, 13 dropped out. The programme included distance goals and encouragement through physical education grades. This context matters when interpreting engagement. The 2024 and 2025 publications associated with the same trial registration report identical allocations. They are provisionally grouped in this investigation as analyses of a probably shared cohort, to avoid counting them as independent trials. App analysis, 2025.
What walking allows us to conclude
Here, the Pokémon-specific question is what its characters add to walking and meetings between players. Comparing GO with the absence of a game is not enough to establish this.
Step counts are an interesting outcome in their own right. Discussing physical fitness, disease prevention or a benefit in patients then requires appropriate measures. Game-related walking may vary in intensity and may replace or complement other activities.
The practical question is therefore to understand for whom the game makes movement easier to start and maintain. The roles of social encounters, weather, the map, events and goals deserve to be separated. Koh and colleagues’ review maps a substantial body of work on GO; its richness also shows how much protocols and populations differ. Scoping review of Pokémon GO and health, Koh, 2026.