Alan Tongue
Pretty much a regular
Business of Betting with Jason Trost
We look how the betting industry really runs by interviewing its insiders
businessofbetting.com
by Business of Betting Podcast | Sep 2, 2019 | Betting | 0 comments
Summary
- Betting content is represented as profitable suggestions or predictions
- It is incredibly difficult to provide positive expectation value (+EV) predictions over the long term
- Predictive information is hard
- Entertainment focused betting content is easier
- +EV is hard to explain
- Predictive and +EV is boring
- Does the inherent conflict matter?
The Inherent Conflict with Betting Content
In April 2017, the Business of Betting podcast was launched, partly, with the objective to try something new in a space that is not dripping with innovation. That being the betting, gambling and wagering content space.
At that time and arguably still very much the same today, a typical betting content show (radio, tv, podcast etc.) is around 30min and consists of picks, tips, brief handicapping analysis, best bets, morals, locks and guarantees. This is then cut up into 30sec clips and reduced to 140 characters for broad distribution. By its very nature and putting aside the actual intent of the producer, betting content in 2019 is catered to the recreational bettor as an entertainment and fan/bettor engagement product.
Don’t get me wrong, I am not critical of this approach and completely understand the business objectives producing this type of content for the proposed audience. In saying that, it is not necessarily sustainable in a world where there are no bandwidth barriers (unlimited YouTube videos and low-cost podcast creation to a worldwide audience) and no distribution barriers (social media and blog/website creation is inexpensive or free). Someone will always be able to do it better, cheaper, faster and more intelligently.
It is very difficult to truly know the answer but something like 1-3% of bettors (let’s go with 2%) are able to win long term. If you estimate that most of the 2% have no interest in sharing their secrets or models, you are left with a very select few that may share some of their information publicly and for the rest, they are in the 98%.
So, what is the conflict?
Generally, betting content is represented as profitable suggestions or predictions that will empower the bettor to make winning selections (and money) in the short term.
Generally, it is incredibly difficult to provide positive expectation value (+EV) predictions over the long term, therefore, the betting content is very hard to categorize as anything other than entertainment or negative expectation value.
There are some people who are able to take certain betting content for what it is, that being entertainment. For the vast majority, it is viewed as actionable and treated as such, leading to bets being placed or financial decisions being made based on the betting content.
The stark reality is that the 98% are usually representing their information as +EV predictions.
Predictive information is hard
The 2% are not sharing what they do, how they do it and the proprietary information that comes with that. The problem for the media ‘gambling expert’ is that they probably don’t bet their own selections or have skin in the game to fully appreciate how hard it really is to climb into the 2% category. It is probably not simple to convince certain so-called gambling experts that they are harming the bottom line of their audience rather than augmenting it.
The issue with the general gambler is that they believe that their favorite gambling expert is really able to give them an advantage or ‘the right side’. Further, it is not hard to see or find those who boast small sample size positive results as evidence of legitimate betting expertise (rather than what is probably is….luck). The ability to market this luck is prevalent across the internet and it is not a new phenomenon.
Short term variance and luck can be as long as an entire NFL season where anyone can go on a winning streak (see Personal Gourmet’s 70% ATS record in the Westgate Super Contest – this is not a betting market but used for illustration purposes of how widely distributed short-term results can be). Even if you can put together some sterling results, you still need to battle the typical barriers like getting outs to bet and disciplined bankroll management to ensure you don’t over bet your edge and deplete your bank, even with +EV selections.
Predictive and +EV is really hard for the 98%.
Entertainment betting content is easier
“The wrong team is favored and I love this team as a home underdog. They are coming off extra rest and the defense has been impenetrable so far this season. I will be taking them on the money line and expect them to win the game.”
This analysis is easy to find and will probably resonate with inexperienced gamblers. The problem remains that the information mentioned is clearly known and therefore included in the price or line available, usually at a rate consistent with its impact on the result. It is also impossible to understand the size of the edge based on the above analysis or what number this person has for this game (if they have done that work).
The vast majority, some would say all, doing their analysis or public explanations in this simplified manner will find it nigh on impossible to win long term against the house edge. Major betting markets, such as the NFL or NBA, are very efficient and without sophisticated and modern technology, mathematical, analytical and process driven betting systems, the odds may not be in your favor.
It may be pertinent to note that this does not mean ‘gut feeling and only watch games’ guy is always inferior to ‘math model nerd’ guy (sorry for the stereotypes), it just means that over a large enough sample size, the reality is that generally you are far more likely to be successful with a more modern approach. This is due to a number of factors and biases that exist, including ‘our eyes lie’, recency bias, loss aversion, confirmation bias and so on.
+EV is hard to explain
Betting shows that have 30 minutes to get through the high-profile NFL games, a few tantalizing college football fixtures, Warriors at Rockets, some college basketball futures and a recap of Heisman trophy line moves, certainly do not have time to extrapolate on why their models like a certain game or why and how a team of researchers, data scientists and analytical minded traders are about to extract value from certain games (shout out to the more qualitative professionals who can make that method work).
In addition to that, models typically contain vast amounts of information, variables and levers, which will generate a price or number on a certain team/game. Even if you wanted to, it is not always possible to identify the exact levers that have resulted in a game or team representing value. A single WR/CB matchup is not the sole or major reason for a betting opportunity in most cases.
