You're comparing two different numbers. The parent is referring to net firearm crimes while the data you cited is "mass shooting deaths".
If you check CDC data through 2016 (https://wonder.cdc.gov), the non-suicide firearm-related death rate is effectively flat for the date range they have data for, varying at most by about 0.1%.