.... for dynamical models. In Ensemble Forecast Systems for medium-range forecasting, there is a tool called "relative measure of predictability" (RMOP), which compares the current uncertainty in the system to past uncertainties over the previous standard period (I believe it's 30 days for the RMOP). I don't think that would necessarily help here ... but something similar might allow assessment of the ENSO predictability ... comparing current uncertainty with past uncertainty at this time of year. You may not have enough data for a statistically robust "climatological" estimate of uncertainty in early boreal spring ... but it might be interesting to try and see if anything useful emerges.