I'll have to look at this more closely when I have more time, but just a couple of notes.
1 ENSO is an important piece of the puzzle and a good foundational concept from which to build an outlook, but nothing should be considered as operating within a vacuum....and as far as sample size goes, that is an issue with respect to everything. I think the structure of ENSO is too important to disregard due to inadequate sample size...now, maybe an argument can be made to simple bin them by east and west based, as opposed to central, as well? But what I will say is that I consider central based events as basin wide bc its pretty difficult to have a central based event not be basin wide...and attempting ro distinguish between central based and basin wide definitely harkens back to your point with respect to finite details and sample sizes.
2) I use some discretion as far as strength designation....for instance if two el nino events both have a peak ONI of +1.5, but one has an MEI of 1.2 and the other 2, I am listing the 1.2 MEI as moderate and 2 as strong. The intensity of the ocean-atmosphere coupling is a very relevent and strong proxy for ENSO peak strength designation.