The goal of the work was to develop grape quality standards based on physical and chemical attributes that may allow to predict wine quality. A treatments Network was installed in Mendoza (North, East and UCO Valley), San Juan (Valle de Zonda), La Rioja (Chilecito), Catamarca and Salta (Calchaqui Valleys), (Argentina). Different crop loads were tested (30 and 50% shoot thinning, 30 and 50% cluster thinning and control) on Malbec and Syrah. During harvest, grapes were analyzed (berry size, sugar concentration, pH, anthocyanins, catechins, tannins, and total phenols) and wine was made with them. The corresponding wines were also analyzed (alcohol, dry matter, color intensity, shade, anthocyanins, catechins, tannins, total phenols, and polymeric color) and evaluated by a panel of wine tasters. By PCA (Principal Component Analysis) two indexes were generated. These indexes summarized the attributes that better explained the observed variability (80%). The indexes were denominated Phenolic richness (RF, associated to anthocyanins, tannins, catechins, total phenols and concentration) and Oxidative threat (PO, associated to pH and hue). There were not differences in RF between varieties or crop levels. Wines with high RF and low PO were considered the most valuables. Cold climate zones had higher RF than hot ones. In Malbec, cold climate zones and low crop level promoted low PO. Predictors of RF and PO in wine were developed for each cultivar. In order to select the most predictive variables, step wise Multiple Linear Regression was used. The adjustment functions RFpred (Malbec R-2 = 80%; Syrah R-2 = 62%) and POpred (Malbec R-2 = 80%; Syrah R-2 = 62%) were defined. The indexes, translated to quality standards, showed agreement between grapes and wines. The methodology may be valid for other red cultivars, but needs to be adjusted for each case. The standards will allow associations between prices and qualities and also permit achieving a more transparent market.