- Optimal
- Posts
- Optimal | "Outlier Hunter" Model Explained

# Optimal | "Outlier Hunter" Model Explained

## Finding favorable odds to profit long-term.

Our goal as +EV sports bettors is to precisely calculate the win percentage of any given bet. If we do this, we can find favorable odds and **profit long-term**.

If you read our post on expected value, you'll recall there is no way to know with 100% certainty what the exact win percentage of any given bet is. The best we can do is create models to calculate the win percentages. The better our models, the more profitable our bets will be.

This post will discuss the Optimal+ “Outlier Hunter” model.

## First, what are “No-Vig Odds”?

To understand the "Outlier Hunter" model, we must first be familiar with the term "No-Vig Odds" (aka Devig Odds).

"No-Vig Odds" are odds with no built-in advantage for the sportsbook, also known as juice. It's a theoretical betting line representing the true odds of a sporting event without any markup by the sportsbook to ensure a profit.

Before we get into the math, let's go through some no-vig examples. Let's demonstrate this with the Timberwolves vs Grizzlies game.

The no-vig odds for the spread with both sides at -110 would be **+7 +100 for the Timberwolves** and **-7 +100 for the Grizzlies**.

The no-vig odds for the moneyline of +225 / -265 would be **+236 for the Timberwolves** and **-236 for the Grizzlies**.

There's a mathematical formula for calculating the no-vig odds. If you're not a fan of math, skip over the next section.

To calculate, take the implied probability of the bet and divide it by the implied probability of all outcomes.

For the spread, the implied probability of -110 is 52.4% on both sides.

```
No-vig win percentage = 52.4% / (52.4% + 52.4%)
No-vig win percentage = 52.4% / 104.8%
No-vig win percentage = 50%
50% is equivalent to +100 in American odds.
```

For the moneyline, the implied probability of +225 is 30.8%, and the implied probability of -265 is 72.6%.

```
Timberwolves:
No-vig win percentage = 30.8% / (30.8% + 72.6%)
No-vig win percentage = 30.8% / 103.4%
No-vig win percentage = 29.78%
29.78% is equivalent to +236 in American odds.
Grizzlies:
No-vig win percentage = 72.6% / (30.8% + 72.6%)
No-vig win percentage = 72.6% / 103.4%
No-vig win percentage = 70.21%
70.21% is equivalent to -236 in American odds.
```

Using the moneyline example, if you find Timberwolves +237 or better, you've discovered a +EV opportunity. Likewise, if you find Grizzlies -235 or better, you've discovered a +EV opportunity.

## What is the “Outlier Hunter” model?

Now that we know what "no-vig odds" are, we can dive into the Outlier Hunter model.

**For the sake of explanation, the Outlier Hunter model is an average of the no-vig odds for a group of sportsbooks offering odds for a given line**. *In reality, the model layers proprietary algorithms we won't discuss, given that they're Optimal's secret sauce. Examples of these custom tweaks are weighting sportsbooks based on their sharpness, timing the odds offered vs. the event's start time, types of wagers, and more.*

Below is an example from the Optimal app: An alternate spread of -9.5 / +9.5 for the Celtics vs. Pistons game.

The Outlier Hunter model calculated that the "Fair Odds" for Boston -9.5 were -197 and for Detroit +9.5 were +197. Given this, if you can find Boston -9.5 -196 or better or Detroit +9.5 +198 or better, you have found a +EV opportunity.

BetMGM offered **Detroit +9.5 +220** in this example, creating a juicy +EV opportunity.

By using no-vig odds, bettors can compare the offered odds of a wager to the expected probabilities and determine if a particular bet has a positive expected value. Understanding and calculating the no-vig odds is valuable for serious sports bettors looking to gain an edge over the sportsbook.

If you haven't already, you can download Optimal here:

And make sure to join our Discord!