Many fantasy football formats in the NFL playoffs ask you to pick a team before the postseason starts and then keep that team throughout the playoffs. An important factor in building your team is predicting how many games each team – and therefore its players – will get to play. The ideal player is on a team that has to play in the first round but makes it all the way to the Super Bowl.

Now you probably want to make your own predictions, but it is always handy to have a sanity check on your thinking. The oddsmakers in Vegas provide us one way to check your forecasting. They have set the odds that each team will win the Super Bowl. I used those numbers to build this table: 

Team

Vegas Odds

Implied Probability
of Winning
Super Bowl

Normalized Probability
of Winning
Super Bowl

New England Patriots

+$160 (8 to 5)

38.5%

33.8%

Dallas Cowboys

+$375 (15 to 4)

21.1%

18.5%

Pittsburgh Steelers

+$900 (9 to 1)

10.0%

8.8%

Atlanta Falcons

+$950 (19 to 2)

9.5%

8.4%

Green Bay Packers

+$950 (19 to 2)

9.5%

8.4%

Kansas City Chiefs

+$1,000 (10 to 1)

9.1%

8.0%

Seattle Seahawks

+$1,400 (14 to 1)

6.7%

5.9%

New York Giants

+$1,800 (18 to 1)

5.3%

4.6%

Houston Texans

+$9,000 (90 to 1)

1.1%

1.0%

Miami Dolphins

+$9,000 (90 to 1)

1.1%

1.0%

Detroit Lions

+$9,000 (90 to 1)

1.1%

1.0%

Oakland Raiders

+$11,500 (115 to 1)

0.9%

0.8%

In the odds column, the number after the plus sign tells you the payout per $100 bet as well as the odds of that payout, for example, if you bet $100 on the Pats and they win the Super Bowl, you'll get $160 back, or an odds payout of 8 to 5.

In the "Implied Probability" column, I converted those odds to a percentage chance of winning the Super Bowl for each team (implied probability for NE = 5/(5+8) = 38.5%). However, the problem with this is that there is 113% chance of some team winning it all (this is the house edge). To convert that to a truer probability, I normalized those numbers by dividing them by 1.13. That gives us the chances that each team will win the Super Bowl which sum to 100%, which is a bit more realistic.

Next, I used those probabilities to predict how many games each team will play. I did that by assuming that the Super Bowl probabilities are applicable on a game-by-game basis. There are some obvious problems in this, because some matchups will be more favorable for some teams than others. However, the bookies have tired to account for this, as well as home field, so it may not be as wild an assumption as it first appears.

Anyhow, let's take Miami vs. Pittsburgh. The Steelers have an 8.8% chance of winning the title, the Dolphins just 1.0%. Using these two numbers, we can infer that PIT has a 90% chance of beating MIA (=8.8/(8.8+1.0). I did this for every game in the first round, then carried those probabilities into the second round. For example, if MIA upsets PIT, it then plays NE. NE has a 97% chance of beating MIA—but of course, there is a low probability of this happening. I took battled through calculating not just the winners but the conditional probabilities of each matchup occurring all the way to the finals (my probability and stats class was a long time ago, so what a sophomore math major could do in minutes took me considerably longer). Here's the expected games for each team, using the probabilities set by Vegas: 

Team

Normalized Probability
of Winning
Super Bowl

Expected Games

New England Patriots

33.8%

2.8

Pittsburgh Steelers

8.8%

2.5

Seattle Seahawks

5.9%

2.3

Dallas Cowboys

18.5%

2.3

Green Bay Packers

8.4%

1.9

Atlanta Falcons

8.4%

1.8

Kansas City Chiefs

8.0%

1.6

Houston Texans

1.0%

1.6

New York Giants

4.6%

1.5

Oakland Raiders

0.8%

1.5

Detroit Lions

1.0%

1.1

Miami Dolphins

1.0%

1.1

Some interesting things emerge. Of course the favorites, NE, has the highest number of expected games. But the Steelers, thanks to a favorable matchup are #2 in this category, even though the Cowboys are more than twice as likely to win the championship. Seattle is a good bet to play as many games as Dallas, as well. Since the Cowboys are starting a round later, their 2.3 expected games takes them deeper into the playoffs than Seattle. But unless you get bonus points for your guys who make the Super Bowl, Seahawk players will get as many chances to score for you as Cowboys.

The Chiefs, thanks to a likely matchup with the Steelers in the 2nd round, are projected for the fewest games of any team with a bye and essentially the same number of games as the Texans or Raiders, who have a far lower chance of winning the Super Bowl.

And the Giants, who many think have the best chance of blocking a Cowboy SB run, have to get past a hot Packers team and therefore also have a the same (roughly) expected games total as Houston and Oakland as well. It's only because those teams are playing each other that they have any chance to advance. Personally, I think calling the Raiders-Texan game right is one of the keys to fantasy success this post-season.

You may be able to beat Vegas on a small sample like this and therefore better project the wins and losses in each game. This will give you a different expectation for games for the guys you draft. But it won't hurt to use these numbers as a second opinion before picking your team.