The purpose of this study is to identify the determinants of cash holding, its relationship with fund performance and investigate the differential role of raw and residual cash holding on mutual funds’ performances and managers’ skill in Pakistan. The raw cash holding is calculated by dividing cash on total net asset and the residual value is measured by cross-sectional regression over the cash holding determinants, each year. The reason for considering residual cash holding is that it has the outlook to capture the relevant information regarding cash holdings, which could not be captured by the raw cash holdings. For this purpose, this study focuses on following objectives; to identify the determinants of cash holdings in mutual fund sector; to examine whether high raw cash holdings or low raw cash holdings outperform; to examine whether high residual cash holdings or low residual cash holdings outperform; to identify the determinants of mutual fund performance; to measure whether low raw cash holdings has better stock selection and market timing abilities or high raw cash holdings; to measure whether low residual cash holdings has better stock selection and market timing abilities or high residual cash holdings. The sample size consists of 190 open-ended mutual funds for the period of 2006 to 2016. Five portfolios are constructed each year on the bases of raw and residual cash holding to avoid the survivor-bias issue. These portfolios are used to measure the risk-adjusted fund performance by using Capital Asset Pricing Model (1964), Fama and French (1993) three factors model and Carhart (1997) four factors model. Further these portfolios are used to measure the macro (market timing) and micro (stock selection) fund manager’s abilities by using Treynor and Mazuy (1966) and Henriksson and Merton (1981) model. This study investigates the fifteen cash holdings determinants in mutual funds’ sector. Panel regression analysis shows that six cash holding determinants are statistically significant namely, size of funds, front-end load, turnover ratio, 12-month fund returns run up, dividend payment and redemption. The result shows that front-end load, turnover ratio, 12-month fund returns run up, dividend payment and redemption have positive and size of fund has negative relationship with cash holding. It indicates that small size funds; high charging front-end load fund; high turnover ratio fund; high 12-month fund returns run up; high dividend paying fund; and high redemption level fund should hold more cash for precautionary purpose. Monthly average raw returns base on lowest raw and residual cash holdings funds outperform highest cash holding funds by 0.8% and 0.1%, respectively. Similarly, the risk-adjusted alphas (fund performance) values for lowest cash holding funds are higher than highest cash holding funds. The fund performance determinants show that fund performance has negative and statistically significant relation with cash, size of fund, dividend payment, fund flow volatility, market risk and redemption, while it has positive relation with 12-month fund returns run up and fund flow prior 12 month. Both Treynor and Mazuy (1966) and Henriksson and Merton (1981) model suggest that the fund managers’ with lowest cash holding (raw and residual) funds, better time the market than highest cash holding fund, whereas lack the stock selection ability This study suggests fund managers to hold least cash, to meet the precautionary motive of cash holding. Substituting redemption with cash reserve, result in increase in the level of fund outflow and it destabilizes the mutual fund performance. High cash holding result in high opportunity cost and it cost investors in term of low returns.