NFL Point Spread Picks Week 1
Pick 1: Minnesota -3.5
Pick 2: Jets -3
Pick 3: New Orleans -7.5
Whew! Almost didn't even make for Sunday's games. As always, these picks are unofficial as we start counting, usually, on week 5. There are a couple of reasons why this year's first week picks are posted so late in the week and I'm hoping to improve it as the season progresses.
1. How to account for strength of schedule in the forecasts
2. Injury impact on result of games (the only solution here would start with player level data which I don't have)
I'm first going to work on the first few I mentioned. Now being part of a tech space in Chicago, it is easier to recruit a partner in crime so we can really improve this. We update our software every week! And here, I haven't touched or revised the algorithm in two years.
We're glad to have been able to post these so that our friends at thepredictiontracker.com can continue to track our success throughout the year.
Again, these are unofficial picks to allow the model a few weeks to capture some of this year's data.
How to read the table:
Pick 2: Jets -3
Pick 3: New Orleans -7.5
Whew! Almost didn't even make for Sunday's games. As always, these picks are unofficial as we start counting, usually, on week 5. There are a couple of reasons why this year's first week picks are posted so late in the week and I'm hoping to improve it as the season progresses.
- Like everyone, automated emails keep me busy
- There are a few factors in the algorithm that I'm working on:
- weighting - some people call it the "forgetting factor"
- thresholding - measurements taken to decrease the impact of high margin games
- defensive and offensive impact on forecasts
- family - I love my kids and prefer spending time with them. Which then is taking a larger toll on my nights
1. How to account for strength of schedule in the forecasts
2. Injury impact on result of games (the only solution here would start with player level data which I don't have)
I'm first going to work on the first few I mentioned. Now being part of a tech space in Chicago, it is easier to recruit a partner in crime so we can really improve this. We update our software every week! And here, I haven't touched or revised the algorithm in two years.
We're glad to have been able to post these so that our friends at thepredictiontracker.com can continue to track our success throughout the year.
Again, these are unofficial picks to allow the model a few weeks to capture some of this year's data.
| Game | Vegas Line | Estimate | Prediction-Vegas | Confidence |
|---|---|---|---|---|
| JACKSONVILLE @ MINNESOTA | -3.5 | -8.4 | -4.9 | 59% |
| BUFFALO @ NY JETS | -3 | -6.1 | -3.1 | 58% |
| WASHINGTON @ NEW ORLEANS | -7.5 | -4.0 | 3.5 | 58% |
| ST LOUIS @ DETROIT | -8.5 | -3.2 | 5.3 | 56% |
| CAROLINA @ TAMPA BAY | 2.5 | -0.6 | -3.1 | 56% |
| SAN FRANCISCO @ GREEN BAY | -4.5 | -9.3 | -4.8 | 56% |
| PHILADELPHIA @ CLEVELAND | 9.5 | 8.9 | -0.6 | 55% |
| MIAMI @ HOUSTON | -13 | -2.2 | 10.8 | 54% |
| INDIANAPOLIS @ CHICAGO | -10 | -3 | 7 | 53% |
| SEATTLE @ ARIZONA | 2.5 | -10.7 | -13.2 | 47% |
| NEW ENGLAND @ TENNESSEE | 5 | -6.9 | -11.9 | 46% |
| CINCINNATI @ BALTIMORE | -7 | -19.3 | -12.3 | 46% |
| ATLANTA @ KANSAS CITY | 2.5 | 10.1 | 7.6 | 42% |
| SAN DIEGO @ OAKLAND | -1 | 9.2 | 10.2 | 41% |
| PITTSBURGH @ DENVER | -1.5 | 15.1 | 16.6 | 41% |
How to read the table:
- Vegas Line: A NEGATIVE number implies the spread favors the HOME team
- Estimate: NFL pickles spread estimate
- Pred-Vegas: Subtraction of the previous two. POSITIVE implies VISITING team will cover spread.
- Confidence: The probability that the spread pick is on the correct side.
Comments
The Washington @ New Orleans pred-vegas is +3.5...wouldn't that mean the pick should have been Washington?