Amos: 2016 NFL Predictor

Amos: 2016 NFL Predictor

Category : Blog , Sports

Update: Follow along with our 2018 NFL predictions here.

Update: I’ve moved the 2016 AMOS model to a more permanent page, along with a revamp of the dashboard. Please keep up with the season HERE.

For those just joining, Amos is a statistical model created to predict the outcomes of each NFL game. Amos takes into account 224 different data points to compute three different probabilities for each game.

First, Amos calculates the probability of each team winning. The dashboard below then displays the team which has the greatest probability of winning. Second, given the spread that has been assigned to a particular game, Amos calculates the probability of that team covering the given spread. Finally, given the Over/Under assigned to a particular game, Amos predicts the probability of both teams’ combined scores to break that threshold.

Additionally, Amos then forecasts the remainder of the season and calculates the most probable ending record for each team.

Amos also has peers within the NFL prediction field. While there are a number of sources for game predictions, I have identified Microsoft’s Bing Predicts (Bing), ESPN’s Football Power Index (FPI) and Nate Silver’s Elo (Elo) as benchmarks for comparison to Amos this year. The selection is due to the statistical approach of each of these predictions methods, which provides similar, but not exact, grounds for comparison.

As a summarization, the below chart plots all four models’ confidence by game. Dots plotted below 50% represent an away favorite and dots plotted above 50% represent a home favorite.

The farther a dot is away from the center, 50%, the more confident a given model is of its predicted favorite.

Based on historical data and current team data, Amos also computes probabilities of over/under and spreads being covered.

Given the over/under assigned to a matchup, Amos computes the probability that both teams’ combined score will push or cover the over/under. For example, an over/under of 41 has been assigned to a game and Amos has computed a probability of 56%. This means that if the same match up could be played an infinite amount of times, 56% of the time the teams’ combined score will be equal to or greater than 41. To be able to evaluate Amos, if Amos’ assigned probability is above 50% and the teams’ combined score is equal to or greater than the assigned over/under, Amos will be given credit for being ‘correct’. The same is true if Amos’ assigned probability is below 50% and the teams’ combined score is less than or equal to the assigned over/under.

Spreads are slightly more complex. For consistency within my modeling techniques, a negative spread represents a home team favorite and a positive spread represents an away team favorite. Given this, Amos computes the probability that the favored team will cover the spread assigned to the game. For example, a spread of 3 has been assigned to a game and Amos has computed a probability of 39%. This means that if the match could be played an infinite amount of times, only 39% of the time the away team would win by 3 or more points. If 3 is swapped out with a -3 in our previous example, then the interpretation would change to only 39% of the time the home team would win by 3 or more points.

Have thoughts on the predictions? See a missing game or an incorrect score? Leave a comment below or send us an email at

See Amos’ predictions from the 2015 NFL season:

Week 1: 11-5

Week 2: 9-7

Week 3: 11-5

Week 4: 9-6

Week 5: 10-4

Week 6: 10-4

Week 7: 9-5

Week 8: 10-4

Week 9: 7-6

HALFTIME REPORT & Week 10: 5-9

Week 11: 10-4

Week 13: 11-5

Week 14: 9-7

Week 15: 14-2

Week 16: 7-9

Week 17: 9-7



September 7, 2016 at 11:04 am

I know the first NFL week is tough, but against the grain you are. I guess I should ask are those picks SU or ATS?


    September 7, 2016 at 1:11 pm

    These are currently straight up. Wanted to get these posted so viewers could digest. Will be posting ATS and Over/Under probabilities either later today or tomorrow for everyone.

    -Trevor L. Bischoff


September 8, 2016 at 8:48 am

Hi Trevor, where are the predictions? I don’t see them in the post above, maybe I’m not looking in the right place?


    September 8, 2016 at 2:46 pm


    Was the dashboard finally able to load for you?

    -Trevor L. Bischoff


September 8, 2016 at 2:27 pm

Wait, am I reading this correct? Amos is 92% sure Chicago is going to beat Houston? And 77% sure Jax is going to beat Green Bay?


    September 8, 2016 at 2:46 pm


    Yes, you are reading the dashboard correctly. This is due to the limited information Amos uses at the beginning of the year to allow flexibility moving forward, which should make more sense after the first week.

    For example, Houston’s home record last year and Chicago’s away record are identical, which Amos takes into account.

    Trevor L. Bischoff


September 8, 2016 at 8:30 pm

It loaded ok, so Amos is predicting a Denver win? Am I reading it correctly?


    September 8, 2016 at 8:49 pm


    Correct, for Week 1, Amos is has Denver with a 53% probability of winning, has a 4% probability of breaking the 43.5 Over/Under, and a 51% probability of covering the spread of -3.

    -Trevor L. Bischoff


September 9, 2016 at 11:21 am

Man, if amos gets 9 or more right this Week, I’m convinced it works.

CHI over HOU, JAX over GB, CLE over PHI, MIA over SEA, SF over LA, WSH over PIT

I would’ve picked the opposite in every single contest above.


September 9, 2016 at 12:32 pm

I agree with Neil

Niko Suave

September 10, 2016 at 5:23 pm

Keep doing what you are doing!!! Thanks!!!

Rod Starling

September 11, 2016 at 1:08 am

Thank you Trevor i appreciate your analytical conclusions and the data you present. Your input helps me to make a more educated decision


September 11, 2016 at 9:19 pm

Was the analysis of the others on the NE/AZ game done with a line of zero?

Chuck D

September 14, 2016 at 9:48 am

This would be a lot upsets Amos?

Tampa Bay
At Cleveland
New Orleans
At Los Angeles

lymon reed

September 15, 2016 at 12:00 am

love the site do you have the win probabilities for week 2 nfl


    September 15, 2016 at 8:23 am


    You can see every week of Amos’ predictions. Directly above the subtitle “Predictions by Amos”, is a selector where you can select the week you’d like to view. This changes the tables of predictions for all four models, including Amos’ betting predictors.

    -Trevor L. Bischoff

Chris B

September 22, 2016 at 12:34 pm

Do you track the Over Under results vs. the Predictor? Would be curious to see that


    September 23, 2016 at 7:08 pm


    I am currently tracking these offline, however I plan to incorporate these soon as well. Keep a look out.

    -Trevor L. Bischoff

Larry Lewis

September 22, 2016 at 5:57 pm

I’m having trouble understanding the predictions with the spread. For instance, the line for the patriots game is 2.5. What does the 39 percent mean? That they will cover it?


    September 23, 2016 at 7:09 pm


    I’ve provided some narratives and explanations above the Amos Betting Predictor. Let me know if these help.

    -Trevor L. Bischoff


September 29, 2016 at 9:02 am

Trevor, could you do an analysis when AMOS matches ELO and when AMOS differs from ELO ? Please do this for SU and then ATS.


October 6, 2016 at 2:52 pm

Hey Trevor. Love following your model against the other models. Can you give me any insight on how your model picks teams at sometimes 90%+ and how you have Cleveland winning nearly 80%?

Chris B

October 11, 2016 at 9:41 am

Hi again Trevor,
Is there any way I can look at the Predicted Over/Under by AMOS from past weeks like I can for the Win/Loss predictions?
According to my analysis of this week for O/U, you would have 7 Wins and 6 Losses if betting Under when AMOS lands on .
Whats interesting when AMOS predicted to be +- 3% within 50%, they were all losses. So if you remove those (there were three in week 5), you go to 7 wins and 3 losses (much better).

So I would love to look at prior weeks to see if that holds true or is just an anomaly.


    October 18, 2016 at 7:00 am

    Hi Chris,

    Great question, and you certainly can. The Week selector on the first dashboard controls the Week selection on the second dashboard. So just choose the week you’d like from there. Best of luck on the analysis, and feel free to share any findings.

    -Trevor L. Bischoff

Leave a Reply

What People Are Saying