Out of the 14,124 analyzed transactions, 856 (6.06%) would not have met the new criteria, including limitations on parent-child relationships and a maximum child size of 1000 virtual bytes. Specifically, template matching showed that 508 of these failing transactions were identified as LN anchor spend scripts, with an additional 175 having a direct parent fitting the same criteria.

Upon delving into why transactions would be rejected under the v3 rules, it was discovered that most rejections occurred because of exceeding ancestor or descendant count limits, which are often indicative of batch Child Pays for Parent (CPFP) strategies or having more than one descendant in a chain. In particular, 595 transactions hit the ancestor count limit, with 302 seeming to be batch CPFP operations involving multiple anchor spends within a single transaction. Notably, only 10 transactions failed due to surpassing the size limit.

Furthermore, the analysis revealed that 99 transactions that failed due to ancestor count limits were never mined, and 175 were blocked because they could lead to a two-length descendant chain. There's also mention of a specific behavior observed in 19 transactions where an anchor output was spent alongside another unconfirmed transaction output; however, exhaustive analysis on this has not been conducted yet.

Additionally, the examination included a histogram analysis showing the size distribution of child transactions, revealing that nearly all transaction sizes fell within 25 virtual bytes on a logarithmic scale. This data underscores the necessity for careful evaluation of the potential impact of implementing the v3 policy on LN commitment transactions, to avoid unintended disruptions to current transaction patterns.