Good question... there are some papers looking at whether El Nino variability increases (vs. Neutral or La Nina winters), but in my view is there are some conflicting results.  Also there is some pretty decent results showing that the variability doesn't change as much as the shift in *mean* anomalies (my colleagues paper here:  https://link.springer.com/article/10.1007%2Fs00382-014-2336-9 ).  While the year-to-year pattern can be noisy, there is information to be gained by the average cahnge in temperature and precipitation.  If you play your odds over the long term, then this skill becomes more evident.  This is why we issue our outlooks as probabilities and not in inches or in degrees... b/c we know the event to event variations can be large.