Week 12 2019 picks
Needless to say last week wasn’t a good start. The raw algorithm plays went 1-4 (20%) and the unofficial plays went 0-5! I would say it was a week of close calls (DET), other results I’d argue were outliers (like the multiple PIT injuries), but the fact remains that picking NFL games is a VUCA endeavor.
Quick note: The unofficial plays I post in previous weeks are those I bring to my contest partners. From there we negotiate and down select to our 5 “official” plays that I use for any Tableau visualizations and performance reporting.
Raw algorithm picks:
NE -6.5 (DAL)
LAR +3 (BAL)
SEA +2 (@PHI)
CIN +6.5 (PIT)
GB +3.5 (@SF)
A good friend and contest partner (and future contributing author) made me reconsider the timing of when I pull market data to analyze the sharp/contrarian sides. Typically, I had done this in the middle of the week but as I was reminded there are late lines moves through the weekend into Sunday pre-game that can alter the perceived sharp side of the games. I will try to post my picks as close to Sunday as possible.
Sharp/Contrarian sides:
GB (@SF) - Weak - Line opened GB +3.5 and moved to +3 (EV) following 53% of the public tickets. I think books are especially keen to remove the hook (the hook is the additional 1/2 point added to a number) if there’s even the slightest market imbalance, particularly for a “public” team. A public team is one that the casual fan knows well and the public likes to bet on. Aaron “discount double-check” Rodgers and GB are a public team. In this case, they made a quarter point move off of 3.5 to 3 even money (meaning bettors can wager $100 to win $100, rather than the typical -110 odds which means you wager $110 to win $100). This tells bettors that the market hasn’t quite committed to GB +3. My algo says GB -1 (53%) but the consensus is closer to the opening line. My only hesitation with this game is that the Pythagorean W/L differential between these two teams says GB is due for regression but like SEA is off the bye. This may be setting up for a late week movement so I’ll continue to monitor.
LAR (BAL) - Moderate - Ravens have been steamed up to 3-pt favorites from an even opening line. Ticket are heavy on BAL (4:1) suggesting an overreaction to the big high-profile blowout win against HOU last week and a LAR team that hasn’t looked too dominate as of late. My algo suggests LAR has a 53% chance of winning versus the consensus at 41%. LAR at home getting a FG is good value here.
CLE (MIA) - Moderate - That TNF game last week was ugly for CLE and with suspensions for a key defensive player the public seems to have soured on CLE. 60% of the tickets are on MIA but the line has moved from CLE -9 to -10.5. I think that’s a good number. As a rule, I can’t lay NFL favorites over 10 points so I’ll pass.
PHI (SEA) - Weak - Line has been steamed up to SEA +1 off of 3:1 tickets on SEA. My algo likes SEA to win (65%) and is the #2 play this week. However there are red flags. Like the GB side, Pythagorean W/L differential seems to think SEA is due for a let down. On the other hand SEA is off a bye, which might reset the Pythag W/L expectations (TBD) and PHI is riddled with injuries to key offensive players. SEA is alluring but on the road against an opponent vying for the division lead, and getting the overwhelming number of public tickets is not a good spot. I’m leaning to pass on this matchup.
DAL (@NE) - Moderate - Ugh, it’s never good to a) bet against the Pats and b) see that your algo is on the square side of a bet between two public teams. I have NE -9 in my model so laying less than a TD at home had me licking my lips! The sharps disagree. They seem to be laying money on America’s Team. The line has moved towards DAL from 38% of tickets. Mid-week lines around key numbers of 3 and 7 sometimes set up well for some late market movement. I’ll monitor but right now leaning to pass on this game.
DEN (@BUF) - Weak - DEN opened 5.5-pt underdogs only to be bet down to 4-pt ‘dogs. Tickets are 50/50 between these two teams. My algo likes the opening number so if I were to pick a side here I’d lean BUF -4. This will be in consideration this week.
WSH (DET) - Strong - This is a classic contrarian pick. This game opened WSH +1.5 only to blow through the key number of 3 behind 4:1 tickets on DET. My algo actually has WSH as a suggested play this week (#6 on my rank) and thinks the line should be closer to the opening number. Matthew Stafford remains out. I will be plugging my nose and taking this game.
CIN (PIT) - Weak - I’m intrigued since this is already a raw algorithm play—and gives me the opposite feeling of the DAL-NE game—confirming evidence! The line has moved off of the key number of 7 to 6.5 in favor of PIT while a slight majority of tickets are on PIT. Since PIT is one of the historically public teams, I wouldn’t say this game confirms the raw algo play but it doesn’t disqualify it yet. I’ll monitor the line up to game time.
IND (HOU) - Strong - I noted this game as a strong sharp pick before kickoff this Thursday. Line moved 2 points towards IND while the public was behind HOU at a 60% clip. I didn’t officially play this game because of the conflicting signals between the algo (HOU 68%) and the available line and sharp action. IND lost but covered the 4 points, 17-20.
Here are the unofficial picks I’m taking to the table—3 home ‘dogs and Aaron Rodgers off a bye week: