Guilty outcomes are the most common legal outcome
While one would expect famous people to experience a higher frequency of false accusations, some form of guilty verdict (conviction, straight guilty plea, or plea agreement) was the most common legal outcome across these 730 arrests.
I need to point out here that the arrest database is incomplete. Of the 730 arrests, 251 (34.3%) are labeled with Unknown outcomes (some of these are ongoing cases, the majority simply do not have a known outcome available).
However, 263 cases (36.0% of all cases, 54.9% of the cases with Known Outcomes) have some form of guilty outcome: guilty convictions, guilty pleas, and guilty plea agreements of some type (e.g. reduced charges). In fact, guilty outcomes occur 7x (6.8x) more often than acquittals:
Two players died before their cases were resolved. Tony Fein died before trial. Jovan Belcher killed himself at the team facility after killing his girlfriend (he shot her 10 times).
This outcome breakdown varies pretty significantly by team (not surprisingly as teams fall in many different legal jurisdictions).
- Guilty Rate was highest for Miami (64% of cases with known outcomes), followed by Detroit and Washington-DC (55%). Guilt Rate was lowest for Dallas (14%), followed by New Orleans (20%) and San Francisco (21%)
- Dropped Case rate was highest in San Diego (42%) followed by Houston (36%) and New England and Pittsburg (33%). It was lowest in St. Louis (zero), followed by Indianapolis (4%) and San Francisco (5%)
- Acquittal Rate was highest in Washington-DC (15%), followed by Baltimore (13%) and Cleveland (11%). Acquittal rate was zero in Carolina, Chicago, Dallas, Indianapolis, Kansas City, New England, New Orleans, New York (for both the Giants and the Jets), Philadelphia, Pittsburgh, and San Diego.
This distribution of legal outcomes is explained in part by types of crimes charged to players from different teams. (We’ll get to this in a bit.)
However, the most common response by teams is “No Response”
Looking across all 730 arrests, NFL teams did not (yet) take action 84% of the time. However, when teams did take act, the most common thing the did was to release players:
Note: The three players who retired after their arrest were Keith Hamilton, Josh Brent, and Rolando McClain.
Team response was highly varied by team, player, the charge and even the timeline of the arrest. In at least* 45 cases, teams released arrested players immediately (i.e., within hours or days following the arrest)—even before the case was resolved. In other cases, players continued to play even after guilty verdicts in trial (Greg Hardy’s case is the most recent high-profile example).
*I use the term “at least” as the database did not clarify the time it took some teams to release players.
Here are some numbers highlighting variance of team response:
- New England released players at the highest rate (27%) followed by Carolina (26%) and New Orleans (20%). On the opposite end of the spectrum Atlanta, Dallas, Green Bay, New York (Giants), Philadelphia, St. Louis and Washington-DC have never release players (according to the USA Today database)
- Suspension rates are similarly varied. It is most frequent in Arizona and Atlanta (17%). It has never happened (according to the USA Today database) in: Baltimore, Dallas, Detroit, Green Bay, Houston, New England, New Orleans, Philadelphia, Pittsburgh, San Diego, Seattle, San Francisco, Tampa Bay and Washington-DC
- Eleven teams (Dallas, Detroit, Green Bay, Houston, Minnesota, Philadelphia, Pittsburgh, San Diego, San Francisco, St. Louis, and Washington-DC have taken no action more than 90+% of arrests. Arizona, New England, and the New York Jets have taken all action most often: in 33% of arrests.
This is depicted in the graph below:
Two-thirds of arrests occur off-season
A potential cause for the lack of action by NFL teams could be that most of the arrests occur off-season (65.7%). However, in 99.6% of arrests the player was under contract. Most NFL contracts have penalty clauses for bad behavior (I cannot analyze this as this data is private):