2009 NFL Point Spread Picks Week 9

Pick 1: San Francisco -4 incorrect NFL point spread pick -$85
Pick 2: Green Bay -10 incorrect NFL point spread pick -$85
Pick 3: NY Giants -5 incorrect NFL point spread pick -$85
Pick 4: Baltimore -3 incorrect NFL point spread pick -$85

As we enter the second half of the NFL season, we understand that there will be good days and bad. There are many factors that are unaccounted for by statistical models, hence the concept of variability and confidence. The goal is to create a winning strategy using Statistics as a guide and Discipline as the driver. Although I would like to maximize the ATS success percentage (currently at 70.4%) for the year, the main goal of this blog is to increase the growth percentage (currently at 61.3%), that is the the real key performance indicator to measure success.

Thank you for the great feedback you've given so far, we've incorporated at least one new thing, instructions on reading the table below. Thanks again. If you haven't yet, please complete these few questions so we can keep improving NFLpickles.

Going all even this week in terms of bankroll used per game. No underdogs this week. Sorry for the short descriptions, need to "get back to work".

TEN @ SF $85 - Almost bailed out since because of Vince Young. I should've bailed out of this game last week. Estimate is above and beyond what Young can do. 49ers by 10.

GB @ TB $85 - The NFL statistical model backing up the tomato cans of the NFL theory. GB by 2 TDs.

SD @ NYG $85 - New York underrated because of its 3 game losing streak.

BAL @ CIN $85 - Too much D by the Ravens.

Without further adieu, here are the NFLpickles free point spread estimates for Week 9 of 2009.









































































































GameVegas LineEstimatePrediction-VegasConfidence
TENNESSEE @ SAN FRANCISCO-4-15.8-11.867%
GREEN BAY @ TAMPA BAY1015.15.164%
SAN DIEGO @ NY GIANTS-5-7.9-2.963%
BALTIMORE @ CINCINNATI35.22.259%
DETROIT @ SEATTLE-10-11.4-1.453%
MIAMI @ NEW ENGLAND-11-12.2-1.253%
KANSAS CITY @ JACKSONVILLE-6.5-9.4-2.952%
PITTSBURGH @ DENVER3-7.1-10.151%
HOUSTON @ INDIANAPOLIS-9-9.2-0.250%
CAROLINA @ NEW ORLEANS-13.5-121.550%
WASHINGTON @ ATLANTA-10-12.2-2.248%
ARIZONA @ CHICAGO-36.99.946%
DALLAS @ PHILADELPHIA-3-2.11.145%


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.

29 comments:

Joel said...

My model suggests:
Buccaneers +10
Chargers +5

Why? I dunno, my model did terrible last week on getting 3 of all the games correct! Was it a fluke week? Maybe... It is the first year i'm doing this and I need to make some changes, especially some type of strength of schedule stat to weight the predictions a bit. My model would have still did poorly even with the week 8 data entered now, so in a way, that is a good thing since the predictions seem to be stable.

As for your survey, I opened the window and while waiting for it to load i was surfing the web, and then closed the browser. So it says I already took it.

Anyway, good luck this week!

Jaime said...

Joel, great job last week. Your model suggests just those two picks? Is it a coincidence that they go against mine? Should I be worried now that your model is doing terrible?

Joel, try to take the survey again, it should work now, although it can be slow.

Joel said...

Forgot to put my other picks this week:

Texans +9.5
Redskins +10
Buccaneers +10
Chargers +5
Broncos +3

All underdogs for some reason this week.

My overall this year is 58-57-1
My Bets this year is 28-24-1

So not good. I've made some significant changes after week 6 though.

I've been researching neural networks, and testing them in R, I have not considered them yet for using them for predictions as of right now. I have a hard time using something that you have no idea what is going on inside of it.

So might tough it out with what i have this year, cause there is going to be a lot of code cleanup once it is over and I have time haha.

DDW said...

Jaimie,

My picks vs yours:

SF -4
GB -10
SD +5
CIN +3

Again I won't bet on a game SD is involved in, but my numbers point to SD covering. I like your pick with BAL -3 better than mine, but again this is what my numbers point to.

Good luck this week!

HappyBreathNet said...

You’re early this week. (I won’t run my model until this evening). This gives me the chance to put my gut to the test (of course, my model often goes against my gut). Regarding your picks I (not my model) agree on SF-4, GB-10, and SD+5. I wouldn’t touch the BAL vs CIN game but if I had to I’d probably take CIN.

The games I (not my model) consider most appealing are MIA +11, IND -9, GB -10, and SD +5. I’ll run my model tonight and update my (my model) real picks at that time. I’ll compute BOBB lines tomorrow (Brian publishes his probabilities on Thursdays).

- Happy

Mike d. said...

Nice work this year Jaimie. Things have been working out well.

I'm a little surprised by the confidence in the Giants pick.

Are they underrated because of their 3 game losing streak to decent opponents or were they overrated because of their 5 game win streak against extremely weak opponents?

I don't really like anything about that game.

dtBy said...

I had some trouble answering the survey questions. Specifically, I don't know if the fact that I include a look at your picks as a reason-check on my own model's selections means I'm "using your picks."

To resolve this dissonance, I've decided that I'll make a bet or two based on your picks this week on my way out of town.

So, extra-good luck this week!

dtBy said...

Oh, I'm definitely skipping the Packers, tho.

Jaime said...

Why? You don't believe in the tomato can theory? Let me know which game you do end up using.

dtBy said...

I see the tomato can theory work week after week.. and yet I still can't buy into it completely.

In any case, I just booked the 49ers for a small wager. Not for lack of confidence.. I'll just not be here to cash the ticket. Hopefully, my friend taking me to the airport will find something to do with the $210!

Jaime said...

What a great friend you are.

dtBy said...

I don't know how great a friend... it's his only compensation for the ride to the airport. What's the expectation?

Jaime said...

Ride to the airport: $30
Expected return: $210*.67 + (-$110)*.33 - $30 = $75 :)

HappyBreathNet said...

Jaime,

I get your point, but I think I have the math a little different. Here's what I show for a single event expectation calculator (using MS Excel). Please correct me if I've got this wrong.

Cell A1
w (stands for wager amount)

Cell B1
210

Cell A2
p (stands for probability of winning the wager)

Cell B2
0.67

Cell A3
B (stands for book. Here I'm meaning the wager required to win $100. For example -110 would be 110).

Cell B3
110

Cell A4
e (stands for expectation)

cell B4
=B2*B1*(B3+100)/B3-B1
this evaluates to $58.61

cell A5
t (stands for target expectation)

cell B5
30 (cover the cab fare)

cell A6
tw (target wager, amount you must bet to achieve target expectation)

cell B6
=B5/B4*B1
evaluates to $107.49

I read this as you need to wager $107.49 on a 67% probability even to achieve a $30 positive expecation in a 10% book. You can create this tool yourself, it's fun to play with. Set your probability to 52.38% and you'll see the expectation go to zero (this is the breakeven point for a 10% book).

dtBy said...

Well, I didn't mean for my little joke to make for a big discussion. However, I think the right way to think about this (from my driver's perspective) is:

67% chance he collects a ticket worth $210. 33% chance he doesn't cash the ticket and misses out on the $30 I would have given him for the ride to the airport.

E = .67*210 + .33*(-30) = 140.70 - 9.90 = 130.80

On the other hand, from MY perspective I have a 0% chance of cashing the ticket and a 100% chance of losing the $110 I paid for it.

E = 0*210 + 1.0*(-110) = 0 - 110 = -110

HappyBreathNet said...

Apparently I need to read a little more carefully - my bad. I also misread Jaime's picks (I thought he had SD +5, but he has NYG -5).

My bets this week are:

SD +4.5 (I wish I could've got +5) for three units

Under 43 in the Det vs SEA for three units

WAS +10 vs Atlanta for one bet unit

JAC -6.5 vs KC for one bet unit

Over 47 in the NYG vs SD for one bet unit

- Happy

Jaime said...

Not one person likes my Giants pick! Should I reconsider? Not a chance, I'll show to pay some respect to the pickle. :)

Joel said...

I find that in Charger games with Norv Turner seem to be inconsistent. It is a well built team, but you never know what is going to show up.

ldu said...

San Francisco will not beat the Titans this week... the NFC West is such a week division and Vince Young is what Tennessee.....

Id pick the Packers -10.5
the Seahawks -10.5
Philadelphia -3.5
Pittsburgh -3.5

Those are the 4 locks of the week but I provide analysis on all games

ive picked tennessee san fran as my upset of the week on my daily blog where I am 64-51-1... thats money in the bank folks!!

HappyBreathNet said...

The BOBB model didn't like your Giants pick either. I think he actually has SD favored by 0.3 points. I've got NYG by 0.9 (so I'm taking SD).

Way to stick to your guns though.

Jaime said...

Eli I need you! Get some tips from your brother.

HappyBreathNet said...

Wow. Rough week (me too). We'll bounce back though.

Jaime said...

Ouch! That is ok though. Every week we learn something new.

Nice week Joel.

Joel said...

Well I did alright this week, I'd upgrade that to good if the broncos cover.
This week for you Jamie was like last week few weeks for myself, and it just plain old sucks. However, if this were easy, then there would be no bookies right?

Mike d. said...

Yikes! Rough week. Well, a couple things I learned this week:

1. My suspicions about the Giants seem to be correct. They are overrated based on their 5-0 start to junk teams. They haven't beat a half decent team all year.
2. Green Bay is terrible. Well their O-Line and Defence are at least. 5 sacks to the worst pass rushing team in the NFL?
3. Cincy is the real deal. What's up with former Bears RBs tearing it up after leaving Chicago?

Let's hope for a bounce back week next week.

Brian said...
This comment has been removed by the author.
Brian said...

Adding to Mike d's things I learned,

1. A large spread against the Dolphins = money in the bank.
Good running game and great defense yields a close game almost every time.

2. GB will lose to any team with a pass rush. (i.e. next week to the cowboys)

3. Don't bet against any team with a new starting qb for the week no matter how terrible you think he is
(vinnie young, derek anderson, kevin kolb, dante culpeppter)

4. Titans will upset some decent teams in the upcoming weeks.(texans, arizona, colts)

Brian said...

My early picks for week 10

Saints over Rams (-14)
Assuming the Saints can score some points early it'll take the it'll take Steven Jackson out of the game

Denver over Washington (-4.5)
I don't think Portis will play, which means absolutely no running game. Denver will eat Campbell.

Bengals over Steelers (+6.5) possibly buy half a point
Cedric Benson looks unstoppable rushing over 100 yds against very good defenses.

Eagles over Chargers
I think there will be a favorable line (+3). And assuming westbrook is back.

Cowboys over GB (+1)?
Pass rush.

HappyBreathNet said...

In a way this bad week increases the accuracy of your claims. If you look at your Expectation Value Probability Distribution for your 2009 picks, it put your actual results back at the center of the distribution. Your expected win, based on your projected confidence from each post, is 18.8 out of 31 wagers this season. Your actual result of 19 wins is right in line with expectations.

Keep up the good work Jaime!