Line Shop Until You Drop

When Bo Nix ran in a 4 yard touchdown with two minutes left in the Broncos game Sunday afternoon, cutting the Seahawks lead to 6, Broncos bettors made some noise! Some cheered wildly, and others merely sighed with relief.

This is because some sportsbooks offered a betting line of Broncos +6.5, others offered Broncos +6, so the final score of 26-20 created a win for some bettors and only a push (or tie) for others — ouch. 

In today’s newsletter, I’ll talk about the importance of “line shopping” — having access to multiple sportsbooks in order to compare odds and bet at the best price. I’ll start with the basics of line shopping 101, and then we’ll get into an advanced lesson on how to make tough choices between two lines most optimally. Finally, as always, we’ll make some bets using what we learned.

Line Shopping 101

If you’re reading this newsletter, you probably know intuitively that you’ll win more with a better line, but a small edge is more powerful than you think.

In Nate Silver’s new book, “On the Edge”, he quantifies the value of improving your bets against the line in detail. For example, a 2023 NBA bettor who wagered $100 on every game, and got just one extra point on the closing line, would have made over $7000!

Let’s get into the basic concepts now.

There are two ways one line can be more desirable than another for a bettor.

1. The “vig” (short for vigorish), is lower. That means a smaller bet could be made and still win you the same amount of money! Take a look at this example for Friday’s Kansas vs UNLV NCAA Football matchup.

Betting $115 on UNLV +7 at BetMGM would win me $100, but I’d only need to bet $108 on DraftKings to win the same $100. It’s a no brainer to chose the book that’s offering us a lower investment to win.

2. The point spread is more favorable. That means the favorite doesn’t need to win by as much, or the underdog has more room to cover, in order for us to win our bet. Take a look at this example for Thursday’s Bills vs. Dolphins game.

A bet on the Dolphins will win the same amount at both sportsbooks, but if the Dolphins win by 2, you’ll lose a bet with Fanatics that you could have easily won by placing your bet with Caesars!

We’ve covered what line shopping is (comparing lines amongst books), and why you should line shop (win bets more frequently, and/or win more money when you do). Let’s talk about how to do this:

First, you need to have access to more than one sportsbook.

I compare lines between 8 different sportsbooks, but to start, I implore you to open accounts with at least two. You’ll gain the most value going from a single book to two and creating the ability to compare. Going from six books to seven has value, but is less critical because you already have an array of lines to shop.

If you’re inspired to open new accounts, Juice Reel’s links will get you the best offers, so check them out here.

Second, you need a fast, easy place to compare lines and choose where to bet. The Juice Reel app is fantastic for this activity. Here is a Juice Reel screenshot of all the lines I have access to in New York for the UNLV vs. Kansas game mentioned above.

Lines can move quickly, and clicking any of those lines in the Juice Reel app immediately takes me into the listed sportsbook to place my bet, so I don’t miss out on value.

This is critical if you’ll be making “live” or “in-game” bets when lines move up and down every play, and it’s almost impossible to compare between many books manually.

One last pro-tip: sometimes you’ll find cases where one book has a “rogue” line while the others are operating in unison. The Bills vs. Dolphins example we mentioned earlier fits this pattern — take a look:

In this case, if you want to bet on the Bills, I’d act fast and bet with Fanatics. Whereas a Dolphins bettor has 6 sportsbooks offering the same line, the Fanatics line is a full-point valuable outlier.

Because most sportsbooks operate like sheep, the line is likely to move toward the group and you could be stuck with +2 if not +1.5 if you delay.

Line Shopping 201: Advanced Tactics

We’ve covered the easy cases, where one line is obviously more attractive than another. Now let’s move on to a harder problem: how to make the best bet when you can choose between a better point spread with higher vig, or a worse point spread that allows us to win more money.

If we wanted to bet on UNLV in their football game against Kansas reviewed above, Juice Reel shows three unique pairings of point spread and vig worth considering. To be considered, a book has to offer the best vig for a particular line. Here they are:

Depending which line we choose, we’ll create a unique risk/return profile. In particular, we will see varied outcomes of our bet (win/lose/push) in the unique case where Kansas wins by exactly 7 points.

To decide which line to choose, the key data point we need to estimate is: how often will Kansas win by exactly 7. Once we do this, we can assess the expected value of a bet on each line, and make a smart decision.

Football is unique among popular sports for betting because points are commonly scored in bunches of sevens and threes. This leads to a margin of victory that is not a balanced distribution, but instead clusters on key numbers (3, 7, 10, 14, etc.).

You can see this visually depicted in a graph of NFL outcomes here:

I’ll explain the expected value math behind all this in the appendix (as always), but for now, let me share the key resource for actually making the right bet; it’s called a “push chart”.

A push chart estimates how much vig a bettor should be willing to pay in order to move a half-point off of any point spread number (in either direction). So when the chart says ±7 next to a line of 4, that indicates you would be indifferent between lines of -4 (-110), and -3.5 (-117).

You’ll find slight variations of these charts for NFL vs. College football (college has slightly less clustering). You can also see that the price we’ll pay to move off a number is higher in more common outcomes, corresponding with the taller green bars above. Here is the chart I recommend using:

Push Chart for Football

Returning to our example, because we value buying on or off the 7 at 15 points of vig (bolded above), we would not move down from DraftKings’ UNLV +7 (-108) to FanDuel’s UNLV +6.5 (-102). In that shift, we only gain 6 points of vig (the difference between 108 and 102) and we require more than 15 for it to be a better value.

By contrast, we would grab the extra half point BallyBet offers and choose +7.5 (-121). Since this half point is also worth 15 points of vig, and we’re only paying 13 (the difference between 121 and 108), this is the bet we should make.

If you want to read more about how rule changes in the NFL have impacted margin of victory, this piece by Covers.com goes DEEP.

Bets of the Week

I’ll stick with the game we’ve been talking about throughout this newsletter, and I will be taking UNLV +7.5, at -121 vig or better, for one unit. While I don’t follow these teams closely, I like the price we’re able to pay to get an extra half point and win on a one TD victory by Kansas, and Juice Reel shows more smart money hitting the Rebels than it does on the Jayhawks (more on this type of research next week).

Next week, we’ll dive into how winning bettors assess the inflow of money onto different sides of a bet, and how Juice Reel allows you to get much smarter about using this data to find a betting edge. Thank you for reading — see you then!

Appendix

In our example above, we compared the following three lines in the Kansas vs UNLV matchup:

To assess the relative merits of a bet to win $100 on UNLV at each line, we’d set up the potential scenarios and outcomes as follows: 

To gauge expected value (EV), we use the following formula:

If around 10% of games are decided by exactly 7 (see NFL Margin of Victory chart above), but that margin is achieved by the favorite four times as often as the underdog, we’d ascribe an 8% chance of Kansas winning by exactly 7. We’ll assume the other 92% of outcomes are split equally between winning scenarios and losing scenarios (giving the oddsmakers credit for being good).

So the EV of the Ballybet wager is as follows:

Winning amount of $100 multiplied by 54% probability of winning (8% of Kansas by 7 plus 46% of all other outcomes) = $54.00

Losing amount of $121 multiplied by 46% probability of losing = $55.66

Expected value of -$1.66

Applying the same formula to the other two options, while maintaing the assumption that 8% of outcomes have Kansas win by 7 produces the following comparison table:

While all EV calculations show up negative because the house applies a vig, the BallyBet offer is clearly the best value.

If we assessed Kansas winning by exactly 7 to be a very unlikely outcome, you could use the same formula to see that you’d be better off with a different line.