So let's look at some actual numbers from that above paper -
These are again averages for the period 1980-82 tropical cyclones that have been classified on the basis of size of the eyewall
Eye-class MSLP(Central pressure in mb) MSLP Change(in mb/12 hrs) No eye 985 +/-5 Large eye ( > 28 kms) 987 +/-8 Medium eye (15-28 kms) 956 +/-9 Small eye ( < 15 kms) 954 +/-12
Note from this table how much the MSLP fluctuates over a period of 12 hours for the smaller eye sizes. Also remember that a vast number of cyclones do not form eyes.
Word of caution if you are trying to view this table from the operational forecaster's point of view it is very likely an individual cyclone may depart from these average values. And while these data sets have been obtained for the Northwest Pacific Ocean it is likely they hold up for other basins as well.