You maybe right...I haven't done the work with the el nino dataset yet.
There is some subjectivity involved...what I did with la nina is I took a binned data set from a peer reviewed paper and then made some adjustments to the data myself. I consider modoki events ones in which the anomalies are centered well to the west, near the dateline, with either neutral or very weak anomalies to the east of 150 degrees longitude, over the eastern regions of 3 and 1.2. I then have a basin-wide/central pacific group, which I refer to in my work as "Hybrid" events, which are centered around approximately 150 longitude, but usually have el nino (la nina) anomalies throughout the basin. Finally, the east based events are entered to the east of 150* longitude, over regions 3 and 1.2, and have either marginal or no el nino (la nina) anomalies west of 150W.
Its tedious, but if you look at the DM SSTs from every year, you can get a good sense of where to group each season. Its helpful if you scale the SST anomalies down to 0.5 degree increments to really get a better sense of where the event was focused, as the coarser increments can be more difficult to discern, especially with respect to the weaker events.