The strength of the extreme El Niño episode during 2015/2016 is comparable to that of the 1997/1998 super El Niño. However, the 2016 event was followed by a weak La Niña, while the 1998 event was followed by a strong one. This paper tries to reveal the causes for La Niña events that have different strengths by comparing the two extreme El Niño cases. The mixed layer heat budget analysis shows that during April-November of the El Niño decaying year, the zonal advection feedback and thermocline feedback in 2016 were relatively weak, which weakened the mixed layer temperature decaying rate. The weaker than normal easterly wind anomalies over the western-central equatorial Pacific are the main reasons for the above phenomena. Further analysis reveals that the easterly anomalies over the western-central equatorial Pacific are related to sea surface temperature anomalies in the subtropical Northeast Pacific (STNEP). While the SSTA in STNEP was negative in 1998, it was positive in 2016. Negative SSTA in the STNEP is favorable for the strengthening of trade winds, and thus the strengthening of easterlies over the western-central equatorial Pacific. In contrast, positive SSTA in the STNEP induces southwesterly wind anomalies through the WES (Wind-Evaporation-SST) feedback and thus suppresses the easterly over the western-central equatorial Pacific. In addition, composite analysis also confirms that the SSTA in the subtropical Northeast Pacific is related to the phase transition of ENSO, that is, the case of El Niño-strong La Niña is accompanied by negative SSTA in the STNEP during its phase transition period, while no obvious negative SSTA in STNEP is observed during the case of El Niño-weak La Niña.