Decarbonizing the transport sector through electric vehicles (EVs) requires understanding how EV charging affects the electricity system and how smart charging can support future energy systems. The aim of this study is to characterize the flexibility of smart charging by analyzing real world private EV usage patterns. The analysis is based on logging data collected from 394 privately owned EVs and survey responses from their owners. The results reveal substantial flexibility potential for smart charging. The share of EVs available for smart charging varies depending on the assumptions on to which extent they are connected for charging, ranging from about 20% to 90% during daytime on an average weekday. The average state of charge (SOC) during the entire logging period is 55% or higher for 95% of EVs. In weeks with more than three logged charging events, approximately 40% of the charging events could be eliminated, according to a mixed integer linear programming model that minimizes the weekly number of charging events while accounting for revealed SOC preferences. Furthermore, the electricity spot price leads EV owners with hourly electricity contracts to charge during hours with spot prices that are, on average 19% lower than those with non-hourly electricity contracts.