Prop School

November 15, 2016

We’ve received a tons of questions about our system and while we aren’t prepared to give away all the models and methods, we do want to give an example of a filter we use when looking through prop offerings. Hope you enjoy, let us know what you think on Twitter! @ TheVegasBreaker

The first thing that should stand out to anyone looking at props is the amount of variability from book to book. Well vetted Vegas spread lines and totals differ by the slightest of margins when comparing book prices. If the Cowboys are -5 vs the Redskins, the maximum deviation you will see is a point in either direction and/or some extra juice. Vegas has these lines nailed down and they’ve been engineered by some of the greatest minds in the industry. Sportsbooks worldwide are unified on these lines as the most accurate and scientific available, there is VERY little variation. Touts will brag about “shopping” for a better line where they are lucky enough to gain one point in their favor.  Think about this, 1pt in an NFL game with a 45pt total accounts for a 2.22% variability- and that’s considered a “big” advantage?!

Props are different though, they can very wildly both in line and price. Sportsbooks usually determine their own prop offerings, or sometimes just lazily copy another popular book. Unlike spread lines, the world is definitively NOT aligned when it comes to props. In many cases books use raw averages that don’t account for all game factors and they even like to predict customer actions and force perceived “bad plays.” For example, a popular player in a primetime matchup might have their total inflated (or deflated) depending on the public betting trends. At this point the science is gone and what we are left with is a wide range of random props with a high level of variance from book to book. It’s then up to the customer to spot the lazy or inflated lines and take advantage- that’s where we come in.

Our first daily step in flagging potential props is looking at 5 different books and comparing lines/juice.  At this point we aren’t looking at stats or trends, we are simply looking for anomalies.  This is a key part of the process and can uncover high value props by itself. Let’s take a look at an example:

This year in Game 6 of the World Series the books were all over the map with a high volume of prop offerings. With two really good pitchers on the mound and full bullpens available, the numbers were lower than the typical regular season lines. Cross comparison of our books revealed three different lines for Jason Heyward:

  1. 5Dimes had Heyward (Hits, Runs, RBIs) o/u .5 @-115
  2. Local Casino Book had Heyward (Hits, Runs, RBIs) o/u 1 @-115
  3. Bovada had Heyward (Hits, Runs, RBIs) o/u 1.5 @-120 

Just looking at these raw numbers doesn’t seem to convey a big statistical difference, but think about the percentages at play. Here’s a breakdown of the outcome set:


Now consider the difference between 1 hit and 2 hits and how it applies to the UNDER. There’s about a 23% chance of Heyward getting a hit every time he bats and if he bats 4 times in a game, Prop #1 is roughly going to lose on average while Prop #2 is roughly going to Push on average. Prop #3 however, will win on average! The inverse of this can be applied to the OVER. In fact there is so much variance here that you could argue the smart play is to take both OVER Prop#1 and UNDER Prop #3 and collect on both props if Heyward hits for his average. Three separate offerings on the same prop with three very different numbers, yet all offerings are within 5pts of juice from each other!

Anomalies like this pop-up all the time and can offer you a lead to errors or general laziness from the books. I guarantee you will NOT see situations like this on the mainstream spread lines and totals because those are the lines that Vegas knows you want to bet. Touts are flipping coins trying to beat the science of Vegas by betting the most popular plays, we prefer to take the road less traveled.

For more information on how to get started (it’s all free) and a look at the books we use: