Tree pits are attractive stormwater control measures (SCMs) for implementation in dense urban areas because of their small footprint, their potentially low cost and the co-benefits they may bring through improved street tree growth. While they provide street trees with passive irrigation, it remains to be determined if tree pits may achieve meaningful reductions in stormwater runoff. We undertook a streetscape experiment to quantify runoff retention of tree pits in a heavy clay soil with low-rates of exfiltration. We calibrated and validated a water balance model using the field experiment data to identify tree pit characteristics driving runoff retention performance. We then applied the model to different implementation scenarios to ascertain how useful these tree pits may be at reducing runoff to return a more natural flow regime in dense urban areas. The main drivers of runoff retention were identified as exfiltration rates from the tree pits and the connected impervious catchment size. Our results show that it is possible, even in dense urban streetscapes with low conductivity (heavy-clay) soils, to achieve a 90% reduction in annual runoff and to reduce days of runoff to just 15 days per year. However, to achieve this, tree pits need to be sized between 2.5% and 8% of the impervious catchment area, depending on pit exfiltration rates. In practice, achieving these tree pit to catchment area ratios for a dense urban streetscape will require consideration in the planning stages of development works and is expected to be most feasible through implementation of tree pits alongside a suite of other complementary SCMs.