The article, which was first published online January 3 in the Journal of Gambling Studies (paywalled), is titled Patterns of Daily Fantasy Sport Play: Tackling the Issues.
The authors, with one exception, are all affiliated with Harvard Medical School and the Division on Addiction of the Cambridge Health Alliance. The article is significant because it finds that “typical” and “heavily involved” daily fantasy sports players on DraftKings “tend to mirror” behaviors of those equivalently involved sports bettors.
This of course is not news to those who have been watching the DFS industry for the last few years. But the big news is that DraftKings was the primary funder of the research study.
So DraftKings paid for a study to show DFS players behave like sports bettors? Not quite.
What did DraftKings pay for, then?
DraftKings provided funding for the study and gave the researchers access to data. The researchers were given access to DraftKings’s play data. From there, they selected a sample of 12,041 DFS players to include in the study.
Access to the kind of data these researchers received is highly valuable from an academic research standpoint. The authors noted that no other study had been conducted relying on this same level of data; instead others had to rely on self-reported data, which is less desirable.
Self-report data is less desirable because people do not always respond truthfully, particularly when being asked questions that may be embarrassing or be attached to a social stigma. Gambling is a subject that might, either consciously or unconsciously, lead an individual to report data less accurately for fear of violating social norms.
Academic research that is funded generally does not entitle the funder to veto or copy-edit an article. The integrity of the academic publishing industry, and academia more generally, relies on the concept of academic freedom.
The idea that researchers report their findings accurately without regard for the interests of research funders. The results here certainly appear to indicate that is what the authors of this study have done.
More about the research process
The Journal of Gambling Studies uses a double-blind peer-review system, the gold-standard in academic publishing for ensuring quality and reliability of articles. Under a double-blind system, neither the author nor the expert reviewers know who the other side is.
The idea is that if the reviewers do not know who the author is, they will not let any bias (positive or negative) creep into their analysis of the article. The reviewers then make a recommendation to the editor on whether to accept the article, request that the author make some revisions to the article, or reject the article.
This article would have received the same treatment whether DraftKings funded it, or no one funded it. Under a properly functioning academic publishing system, there is no preference for author or funder, only for the quality and rigor of the work.
What is in the DraftKings article
The article itself does not provide a ton of new information of interest to the casual follower of legal sports betting, nor for the casual fan of DFS. But there are a few noteworthy points worthy of discussion.
First, it should be noted that a lot of the findings here supplement the tremendous work done by Ed Miller and Daniel Singer in their 2015 Sports Business Journal article.
The first point to keep in mind is that this was designed to understand what the actual play patterns of NFL DFS players look like. The authors note that:
“[s]ome aspects of DFS might make the activity more risky than traditional fantasy sports.”
This study is the first with access to one of the two DFS giants’ customer play records, which makes the data set important. This information is not easily accessible, and the authors were granted a tremendous privilege to be allowed to examine the data.
It should be noted that the data that the researchers received anonymized the players’ information, effectively making it impossible for anyone to know who was in the sample.
The age of players
The researchers found that the average age of a player in the sample was 34, but there was a range of player ages from 15 to 82 years old. In footnote 10 of the article, the authors note that 10 players in the sample were under the age of 18.
This finding, though not especially relevant to the article, is something that DraftKings will likely need to cover in future discussions with regulators: how and why those under 18 were allowed to register with birthdates that would make them under 18 (or 21).
Heavy users
While most users in the study engaged with DFS on a moderate level of activity, paying a median of $87 in entry fees across the 2014 NFL season, there is a group who engaged with DFS at an extremely high level.
But this high level of engagement group is not uniform, as some have a high level of time investment (entering lots of contests) and others have a high level of money invested. The authors note that this suggests that a one-size-fits-all consumer protection model is ill-suited to DFS.
Overall, the authors found that there were similarities between DFS players and other studies of internet gamblers. But as the authors’ acknowledged, this is a very early step in understanding DFS play and its relationship to problem gambling.
What to make of this
There are a few things to consider. First, this research is an important step forward in understanding how DFS relates to other forms of gambling. It should be noted that this sample is now a bit dated, as it examined play during the 2014 NFL season. We are many attorney general opinions, and hundreds of millions of dollars in advertising separated from that season.
The second consideration is questions raised by, but independent of, the study itself. In this category, what was DraftKings thinking?
An early piece of advice law students receive is, do not ask questions that you do not already know the answer to. This is generally pretty sage advice. If DraftKings did not know how this study was going to turn out, why grant access to the data?
It is unclear what the company’s internal analyses showed, but this study is not supportive of the arguments they have continued to advance, nor those advanced by the organization formally known as the Fantasy Sports & Gaming Association. That group continues to promote the idea that “the behavior of fantasy sports players differs greatly from sports bettors.” This study appears to be a strong rebuttal to that proposition.
DraftKings could have hired consultants to run this study and retained the information internally.
Finally, there is likely not much to come from this study in terms of fallout, but there are certainly questions about how a 15-year-old was able to register as a 15-year-old, and for how long this was the case.
