2017 Week 7 NFL Point Spread Picks
Pick 1: Kansas City -3
Pick 2: NY Giants +5.5
Pick 3: Pittsburgh -5
Finally got my machine learning NFL spread models to work on Python and export the info into HTML! Exciting for a geek like me. Still working on a few formating details (like the number of decimals) but wanted to get it out there so you can see the results from all the games. Of course, I do recommend on betting the top games based on my percent confidence in the model calculation. Some of my models have been improved to take advantage of the tools now available that weren't there before. This gives me optimism to finish this season way high.
The 2017 ATS Standings are now posted and available from the top navigation bar of the page.
Ok so this week we again stay at 15% of bankroll split into 3 games gives us a bet of $520 per NFL football game. Remember that historically my models have performed much better by the end of the season and especially the playoffs which is why some years we finish below 50% throughout the year but positive on bankroll. Good luck! and yes, stay away from the Packers game...and please share your thoughts on which is your sure pick of the week, mine is KC (5-0 ATS this year, killing it!).
How to read the table:
Pick 2: NY Giants +5.5
Pick 3: Pittsburgh -5
Finally got my machine learning NFL spread models to work on Python and export the info into HTML! Exciting for a geek like me. Still working on a few formating details (like the number of decimals) but wanted to get it out there so you can see the results from all the games. Of course, I do recommend on betting the top games based on my percent confidence in the model calculation. Some of my models have been improved to take advantage of the tools now available that weren't there before. This gives me optimism to finish this season way high.
The 2017 ATS Standings are now posted and available from the top navigation bar of the page.
Ok so this week we again stay at 15% of bankroll split into 3 games gives us a bet of $520 per NFL football game. Remember that historically my models have performed much better by the end of the season and especially the playoffs which is why some years we finish below 50% throughout the year but positive on bankroll. Good luck! and yes, stay away from the Packers game...and please share your thoughts on which is your sure pick of the week, mine is KC (5-0 ATS this year, killing it!).
Game | Vegas Line | Estimate | Prediction-Vegas | Confidence |
---|---|---|---|---|
KANSAS CITY @ OAKLAND | 3.00 | 16.50 | 13.50 | 63.2 |
SEATTLE @ NY GIANTS | 5.50 | 4.37 | -1.13 | 59.1 |
CINCINNATI @ PITTSBURGH | -5.00 | -6.17 | -1.17 | 58.9 |
TAMPA BAY @ BUFFALO | -3.00 | -4.56 | -1.56 | 56.9 |
ARIZONA @ LA RAMS(LONDON) | -3.50 | -9.44 | -5.94 | 54.9 |
DALLAS @ SAN FRANCISCO | 6.00 | 9.30 | 3.30 | 54.3 |
BALTIMORE @ MINNESOTA | -5.50 | -9.04 | -3.54 | 54.2 |
TENNESSEE @ CLEVELAND | 5.50 | 7.26 | 1.76 | 53.7 |
CAROLINA @ CHICAGO | 3.50 | 5.00 | 1.50 | 53.7 |
NY JETS @ MIAMI | -3.00 | -5.85 | -2.85 | 53.5 |
ATLANTA @ NEW ENGLAND | -3.50 | 3.50 | 7.00 | 50.9 |
WASHINGTON @ PHILADELPHIA | -5.00 | -1.56 | 3.44 | 49.0 |
JACKSONVILLE @ INDIANAPOLIS | 3.00 | 12.99 | 9.99 | 47.1 |
DENVER @ LA CHARGERS | -1.50 | -0.11 | 1.39 | 42.7 |
NEW ORLEANS @ GREEN BAY | 5.50 | 11.85 | 6.35 | NA |
How to read the table:
- Vegas Line: A NEGATIVE number implies the point spread favors the HOME team
- Estimate: NFL Pickles' point spread prediction
- Pred-Vegas: Subtraction: POSITIVE implies VISITING team will cover point spread.
- Confidence: The probability that the point spread pick is on the correct side.
Comments
How to interpret the confidence number. Seems like the implied KC line puts KC chances above 90% for the win.
Wouldn't you normally tilt your bankroll heavily to that game instead of splitting the 15% amongst the 3 games?
I'd think Kelly would have one using a large percentage of their bankroll on KC moneyline if you really think KC has 70-90 chance of winning game? Thoughts?
""Of course, I do recommend on betting the top games based on my percent confidence in the model calculation. ""
As for the percent calculation, it’s situational not about how big the difference is. Meaning if the model with this difference at this spread, how have they historically performed. Hope that makes sense.
Minn -5
Carolina -3
Steelers -4
Giants +3.5