Yield estimation of grapes is crucial for efficient and cost-effective management of vineyards. Presence of spatial variability within and across vineyards poses a great challenge for development of standard models to accurately predict grape yields. Traditional methods of yield prediction are highly costly and time-consuming, requiring a large amount of human labor. Precision Viticulture (PV) involves the use of remote sensing technologies that are non- invasive, time and cost-effective in defining the spatial variability of vineyards and estimating grape yields. Normalized Difference Vegetation Index (NDVI) is the most widely used satellite-based vegetation index whose relationship with crop vigor is well-defined. However, NDVI has inadequacies due to its susceptibility to atmospheric effect, saturation phenomenon, and sensor quality, due to which it gives unreliable prediction in many cases. To address this problem, this study attempted to compare the effectiveness of satellite imagery-derived vegetation indices namely, NDVI and Enhanced Vegetation Index (EVI) in predicting the yield of grapes (Vitis vinifera L.) vineyard in Western New York, USA. Sentinel-2 images were used to calculate NDVI and EVI values, which were correlated with ground-measured yield of grapes. Linear regression analysis showed no relationship of either index with yield of grapes. Pearson's correlation test resulted a correlation coefficient of -0.188 between NDVI and yield and -0.23 between EVI and yield and the correlation coefficients were statistically significant at p<0.05. The results revealed that Sentinel-2 remote sensing based vegetation indices do not effectively predict the yield of grape vineyards. Further study needs to be done to realize the best remote sensing platform and best vegetation index for reliable yield estimation of wine grapes. Identification of such a technology has great potential in revolutionizing the viticulture industry through increase in efficiency of vineyard management and yield estimation.