A database, the largest one to date, of magnetosheathencounters by geosynchronous satellites during 1986--1992 and 1999--2000and upstream observations by Wind, IMP 8, or Geotail in the solar wind areused to estimate the forecasting capability of the models of Chao et al. <2001>, Shue et al. <1998>, and Petrinec and Russell <1996>. For each of the1-min resolution data points obtained by GOES spacecraft, we check the followingtwo things: if the magnetosheath was observed by the spacecraft, and if eachof the three models predicted a magnetosheath encounter by the spacecraft.Three parameters are defined to quantify the models' forecasting capability:probability of prediction (PoP), probability of detection (PoD), and falsealarm rate (FAR). A higher PoP and PoD with a lower FAR imply a better forecastingmodel. In the 1986--1992 period we found that most of the magnetosheathencounters observed at 6.6 REare detected. In particular, the Chaoet al. <2001> model predicts the lowest FAR compared withthose of the other two models. We have also studied the magnetosheath encountersmade by GOES spacecraft for the period from 1999 to 2000 using Wind and Geotailas the solar wind monitor. This independent database of magnetosheath encountersduring 1999--2000 confirms our previous anticipations. The PoD of Petrinec and Russell <1996>model (94%) is much higher than those of Shueet al. <1998> (74%) and Chaoet al. <2001> (84%) models. The Chao et al. <2001> model has a higher PoPthan the other two models. The values of FAR for Chao et al. <2001>, Shue et al. <1998>, and Petrinec and Russell <1996> models are 25,27, and 40%, respectively.