Traffic monitoring efforts to date have focussed primarily on motorized travel. As more jurisdictions prioritize active transportation, addressing the need for network-level pedestrian data through system-wide traffic monitoring programs is essential to optimize engineering decisions. The purpose of this research was to evaluate methods to better leverage pedestrian count data for cities to estimate annual pedestrian traffic statistics. Using data from eight automated pedestrian counters, this research validated previously established traffic pattern groups in downtown Winnipeg while analyzing and evaluating adjustment methods to estimate annual statistics from short-duration counts. Pedestrian traffic data was collected throughout 2016. Using this continuous dataset, two traffic pattern groups initially developed using short-duration count data were updated. Results indicate that in Winnipeg’s downtown area, the day-of-year factor method for adjusting short-duration counts produces the most precise estimates of annual average daily pedestrian traffic (AADPT), with mean absolute percent error ranging from 19% to 16% for single day and two week counts respectively. Also that adjustment methods using partial day counts produce approximately twice as much error on average than multi-day counts. Additionally, counts beyond three days in duration do not significantly improve the precision of any multiday adjustment method’s estimate of AADPT. Finally, this research highlighted the presence of two additional pedestrian traffic peak periods often ignored by traditional pedestrian traffic data collection methods, these being the noon peak (11:00-13:00) and the evening event peak (21:00-23:00).