There are many unknown factors behind the mass shooting incidents (defined as 4 or more victims per incident, dead or wounded). It is not an easy matter to figure them all out. Here we try to exhibit some seasonality of the incidents which is above the specific reasons behind each incident.

Previous blogs on this topic are here
US Mass Shootings 2013-2015: Incidents vs Population
US Mass Shootings 2013-2015: Geo and Temporal Distributions
Spatial Analysis by Quartile Partitioning (Analysis methodology used)

Data source: http://www.gunviolencearchive.org/

Trends
This is charting the incidents as time series.

At nation level, 2014 has got a bit less incidents than 2013 and 2015. It is about once a day!
At state level, we see that some states are getting significantly more: Georgia, Massachusetts, Louisiana, Maryland, Pennsylvania, etc.

And some states are getting significantly less: California, Kansas, Nevada etc.
From the quarterly trends, we see right away the chart exhibits seasonality. It is natural for us to have a closer look at the seasonality and try to find the patterns.
Patterns
In this chart, we try to explore the uneven distribution of the incidents and we do see patterns. So here are some insights:

Quarterly: A third of the incidents took place in Q3 which got the highest number of incidents. It is the most murderous of all seasons.
Monthly: The hotter months got more incidents: May-September, almost twice the amount of colder months.
Weekly: Weekend is significantly more dangerous than workdays. Half of the incidents took place on weekends. Saturday is 2x that of workdays and Sunday is the most dangerous, 3x that of workdays.
What about Sundays in August? They could likely be the worst days with the highest number of mass shooting incidents.

There could be more sophisticate methods to detect seasonality in data, such as Discrete Fourier Transform. However we will not go that far in this article today.

The interactive version can be downloaded here.

Questions
What are behind the patterns? I don't know. Do you know?