The Truth Behind Sports Handicapping Algorithms For Our Betting Picks

Ross Benjamin

Saturday, January 23, 2016 6:29 PM GMT

Saturday, Jan. 23, 2016 6:29 PM GMT

Ross Benjamin is one of our writers at SBR and he’s been a professional sports handicapper for over two decades. Read his insightful article that states the facts and discloses the fallacies.

“Numbers Don’t Lie and Liars Don’t Figure”
There are very few if any successful sports bettors or professional handicappers that don’t rely on statistical data in some way, shape, or form. I scoff when I hear individuals claim that numbers mean nothing, and their gut instincts is what they solely rely on. My old saying is, “betting with your gut is the shortest route to a stomach ulcer”. The majority who make these types of claims are phone touts or bettors with a small sample size. In a majority of cases, these individuals don’t have the wherewithal, knowledge, or work ethic to arrive at a selection other than their own false intuitions. They’re not only lying to themselves, but their lack of transparency, and ridiculous rhetoric is an insult anyone with even an average intelligence level.

 

It’s About “Beating the Number”
The easiest way for me to ignore one’s opinion as it pertains to sports betting is by talking about irrelevant topics such as weather, injuries, and individual player statistics. This just in, weather and injuries are all factored into the line. Nobody has better reliable sources in that regard than the odds-makers themselves. They’re quick to make necessary adjustments, and to think you are one step ahead of them is just plain naïve. My suggestion to those of you that get caught up on individual players statistics when handicapping a game is that right now you’re on the wrong betting platform, and should consider concentrating your efforts on daily fantasy sports wagering sites such as Draft Kings or Fan Duel.

 

Introduction To Algorithms
My forte has been technical handicapping over the past two decades. The description of a technical handicapper is often misconstrued as an individual that relies simply on team trends. That’s the furthest thing from the truth, and in the following paragraphs I’ll exhibit the fallacy behind that belief. Trends involve past results for specific teams, while algorithms involve all teams in a specific situation or situations.

Team trends have become relatively useless over the past 15 years or so due to free agency in professional sports, and players leaving college early to pursue a career at the next level. No longer do you see a core group of players remain with the same professional sports franchise for an extended period of time as a result of free agency. Elite college athletes staying in school for an entire four years has long been a thing of the past, and is pretty much extinct.

There’s another point of emphasis that I must make clear. Because someone is a technical handicapper, doesn’t mean that individual doesn’t look at other miscellaneous factors such as emotion, betting patterns, or head coaching trends just to name a few. My personal philosophy being, you can’t limit yourself to one specific facet of handicapping a game, and considering all pertinent factors will just increase your edge. There are no shortcuts, while hard work is always a foundation of success, and the sports handicapping industry is no different than any other in that regard.

 

What Is A Handicapping Algorithm?
First let’s begin with my personal definition of what a handicapping algorithm is. A handicapping algorithm can be defined as: one set or a finite number of unambiguous instructions that, given some set of initial conditions, can be performed in a prescribed sequence to achieve results based on a percentage of probability. Now let’s break that down in laymen’s terms as it applies to sports handicapping.

 

Parameters, Definitions, and Details
For starters, unambiguous means categorical, clear, precise, or unequivocal. Then what are unambiguous instructions given some set of initial conditions mean in regards to sports handicapping? I’m glad you asked. It can be defined as a query or set of queries with regards to a specific situation. An example of such would be the following. How does a NFL home underdog do against the spread, following a straight up underdog win by 10 points or more in their previous game? It’s very important to understand, I’m not referring to a specific team, rather asking the handicapping software for results based on all NFL teams when cast into that exact scenario. When entering that query into the software database, it indicates to me that those NFL home underdogs have gone 79-59-2 ATS (57.2%) since 1980.

I’ve now established a percentage of probability (57.2%) by using unambiguous instructions, and have set some initial set of conditions (NFL home underdogs off a SU win by 10 points or more). The key phrases from this point on now becomes a finite number of unambiguous instructions.

We’ve already determined how home underdogs of 10.0 or more do against the spread following a straight up win by 10 points or more in its previous game. Now let’s take this one step further. I’ll now ask additional queries while broadening the set of conditions. Remember the unambiguous instructions can be finite or never ending. What if the home underdog is +5.5 or less, and their previous win came versus a non-division opponent? When adding those queries into the mix, the algorithm now improves to 48-24-22 ATS (66.6%) since 1980. As a matter of fact, the underdog won 45 of those 73 (61.6%) games outright.

 

Specific NFL Example
In Week 16 of this past NFL regular season, Arizona was hosting Green Bay. At the time, I handicapped that contest. Arizona was a 4.5 point favorite, and the line eventually closed at 6.0. The Cardinals were coming off a 40-17 win at Philadelphia in their previous game. The victory improved Arizona’s record at the time to 12-2 (.857). Green Bay entered that contest having gone 3-0 SU&ATS in their previous three games, and the Packers were a favorite on each of those three occasions. That winning streak improved Green Bay’s record to 10-4 (.714). What intrigued me the most about this situation was having a red-hot team as an underdog, after covering each of their previous three games as a favorite. Then with a couple of follow up queries, my professional intuition was realized, albeit that it was a small sample size.

Any NFL home favorite of 7.5 or less, possessing a win percentage of better than .666, versus an opponent that went 3-0 SU&ATS as a favorite during its previous three games, and that opponent has a win percentage of better than .625, resulted in the home favorites going a perfect 12-0 SU&ATS since 2001. Arizona went on to win and cover the game easily by a score of 38-8, and now that NFL betting algorithm stands at 13-0 SU&ATS.

Team Opponent Date Line SU ATS
San Francisco 43 Miami 21 12/16/2001 -3.5 W W
San Francisco 13 Philadelphia 3 12/22/2001 -3.0 W W
New Orleans 35 San Francisco 27 10/20/2002 -1.0 W W
Tampa Bay 34 Atlanta 10 12/8/2002 -3.5 W W
New England 24 Baltimore 3 11/28/2004 -7.0 W W
Dallas 37 Green Bay 27 11/29/2007 -7.5 W W
Pittsburgh 20 Dallas 13 12/7/2008 -4.0 W W
New Orleans 48 NY Giants 27 10/18/2009 -3.0 W W
Minnesota 34 Dallas 3 1/17/2010 -2.5 W W
Baltimore 16 San Francisco 6 11/24/2011 -3.5 W W
New England 30 New Orleans 27 10/13/2013 -2.5 W W
Denver New England 1/19/2014 -5.0 W W
Arizona Green Bay 12/27/2015 -6.0 W W

 

College Football Example
One of my more successful college football betting algorithms involves a home favorite of 10.0 or more, following an upset win as an away underdog of 10.0 or more. Those exact set of conditions resulted in the home favorite going 55-30 ATS (64.7%) since 1994. If those home teams were facing an opponent playing with revenge, the algorithm improved to 35-14 ATS (71.4%) during that same time span. What was interesting about this algorithm is that it defies sports handicapping-101. The popular school of thought is, when a team is coming off a huge upset win, they’ll be in for a letdown in their following game. These results clearly disprove that notion, and this exact situation can be used for future college football picks.

Team Opponent Date Line SU ATS
Maryland 31 Wake Forest 7 9/24/1994 -12.0 W W
Colorado St. 47 UTEP 9 10/15/1994 -23.5 W W
Auburn 31 Arkansas 14 10/29/2014 -11.5 W W
Kansas 34 Iowa St. 7 10/14/1995 -17.0 W W
Air Force 34 Hawaii 7 10/26/1996 -35.0 W L
Clemson 40 NC State 17 11/16/1996 -10.0 W W
Fresno St. 46 UNLV 28 10/25/1997 -11.5 W W
Boston College 41 Rutgers 14 9/12/1998 -22.5 W W
Syracuse 70 Rutgers 14 9/19/1998 -41.0 W W
Navy 33 Rutgers 36 11/7/1998 -10.5 L L
Illinois 29 Northwestern 7 11/20/1999 -13.5 W W
UAB 47 UL-Lafayette 2 9/30/2000 -19.0 W W
Virginia 31 Duke 10 9/29/2001 -17.5 W W
Boise St. 49 Nevada 7 10/27/2001 -22.0 W W
Syracuse 24 West Virginia 13 11/10/2001 -14.5 W L
San Jose St. 58 UTEP 24 9/28/2002 -17.0 W W
Wake Forest 36 Duke 10 10/12/2002 -12.0 W W
Pittsburgh 29 Temple 22 11/9/2002 -16.5 W L
Ole Miss 55 Arkansas St. 0 10/11/2003 -26.0 W W
Tennessee 59 Mississippi St. 21 11/15/2003 -24.0 W W
South Florida 41 East Carolina 17 11/13/2004 -13.0 W W
Oregon 45 Washington 21 10/15/2005 -16.5 W W
Northern Illinois 42 Western Michigan 7 11/23/2005 -11.0 W W
Western Michigan 41 Temple 7 9/23/2006 -30.0 W W
Colorado St. 28 UNLV 7 10/7/2006 -17.0 W W
Central Michigan 47 Army 23 10/13/2007 -14.0 W W
Utah 23 San Diego St. 7 10/13/2007 -14.5 W W
Oregon St. 23 Stanford 6 10/27/2007 -14.0 W W
Western Michigan 16 Temple 3 11/24/2007 -14.0 W L
Boise St. 38 Louisiana Tech 3 10/1/2008 -24.0 W W
Houston 45 UAB 20 10/9/2008 -17.5 W W
Oklahoma St. 34 Baylor 6 10/18/2008 -17.0 W W
UTEP 36 SMU 10 11/15/2008 -14.0 W W
Northern Illinois 31 Idaho 34 9/26/2009 -14.5 L L
Texas Tech 30 Texas A&M 52 10/24/2009 -22.5 L L
North Carolina 19 Duke 6 11/7/2009 10.0 W W
UCLA 42 Washington St. 28 10/2/2010 -23.5 W L
Marshall 28 Memphis 13 11/13/2010 -16.0 W L
Iowa St. 13 Kansas 10 11/5/2011 -14.5 W L
Northwestern 28 Rice 6 11/12/2011 -17.0 W W
TCU 34 Colorado St. 10 11/19/2011 -34.0 W L
Kansas St. 56 Kansas 16 10/6/2012 -25.0 W W
Kent St. 35 Akron 24 11/3/2012 -19.0 W L
Notre Dame 29 Pittsburgh 26 11/3/2012 -16.5 W L
Navy 51 Delaware 7 9/14/2013 -17.0 W W
Auburn 45 Florida Atlantic 10 10/26/2013 -23.0 W W
Vanderbilt 22 Kentucky 6 11/16/2013 -11.0 W W
Ole Miss 27 Vanderbilt 16 9/26/2015 -27.0 W L
Arizona St. 48 Colorado 23 10/10/2015 -15.0 W W

 

comment here