# Tips for NFL picks: Where Vegas NFL lines and win totals disagree on the 2018 season

When it comes to figuring out best bets in the NFL season win totals market, it's easy to get caught up in the hype for or against particular teams. But that can also have a snowball effect on the betting market as everyone looks to get in on this year's sexiest names.

One thing I find instructive is to take the posted lines for Weeks 1-16 by CG Technology, translate them into a win expectancy for each team in each game, and add it all up to see how close those game lines come to the odds posted in the current win total market. Of course, there are no lines for Week 17 as those depend largely on which teams are still playing for something, but by doing the math for the season's first 16 weeks and comparing all the divisional matchups from Week 17 to their lines earlier in the year, we can get a pretty good estimation.

So how do game-line win expectancy totals compare to the posted win total lines? It turns out, pretty closely!

Want more picks? At SportsLine, I've posted my play for the Hall of Fame Game, as well as analysis of the Super Bowl betting lines and five win totals I like.

Out of the win expectancy totals for 32 NFL teams, 24 fall within a half-game of their win totals in the betting market. And for most of the outliers, the juice is on the side where the math would tell you to expect. We'll get to analysis in a minute, but first, here's our data, which uses the CG Technology game lines to calculate win expectancy and the Westgate win totals.

TeamWin ExpTotalOverUnderDiff
CAR7.9449120-140-1.056
NYG6.3637-135115-0.637
KC7.8958.5100-120-0.605
WAS6.4727110-130-0.528
ATL8.9839.5110-130-0.517
LAR9.49210100-120-0.508
SEA7.6128105-125-0.388
OAK7.6218-125105-0.379
NYJ5.6936110-130-0.307
IND6.2096.5-130110-0.291
BUF6.2536.5140-160-0.247
MIA6.2656.5100-120-0.235
CIN6.7857120-140-0.215
NO9.2859.5-120100-0.215
DET7.3017.5-130110-0.199
TB6.326.5100-120-0.180
DAL8.3458.5100-120-0.155
PIT10.38510.5105-125-0.115
CHI6.3856.5-150130-0.115
TEN7.9878-110-110-0.013
CLE5.4935.5-150130-0.007
MIN9.99510-110-110-0.005
DEN7.0077-1401200.007
GB10.03210120-1400.032
SF8.5348.5-1201000.034
LAC9.6969.5-105-1150.196
PHI10.72910.5110-1300.229
JAC9.3489-1201000.348
ARI5.9485.5-1451250.448
NE11.45711-1601400.457
BAL8.7818-1401200.781
HOU9.3878.5-1401200.887

Notes:

• I made my best guess on the Week 17 lines based on the earlier divisional matchup for each game and what each team had to play for in this simulation in the final week.
• The difference column doesn't take into account the juice for each line. So the Panthers win total is calculated at nine wins despite the Under being -140.
• The win total market is perennially overinflated (people love to bet Overs). Adding up the win total numbers above gives you 259.5 wins in a 256-game season (and it'll likely keep rising). Our win expectancy column adds up to 256 wins.

Now let's take a look at our four biggest outliers and what their numbers mean for how you should approach the betting market.

### Carolina Panthers

• Game-line win expectancy: 7.944 wins
• Win total: 9 wins (+120 Over, -140 Under)
• Difference: -1.056 wins

The win total market values the Panthers somewhere between 8.5 and nine wins. But the game lines for Weeks 1-16 has them a bit weaker than that. Just look at Week 1, where the Panthers aren't getting a full three points against the Cowboys, a team with an 8.5-win line and slight juice on the Under.

Think you're a football expert? It's time to prove it. Compete against friends, family and co-workers or put your football knowledge to the test against the NFL Today cast. Play 2018 Football Pick 'Em now!

If you're a believer in the Panthers, it makes more sense to take some of those more questionable game lines in a market where they're perceived as a weaker team. CAR -2.5 vs. DAL in Week 1 is a good example, but you could also look to CAR -4 vs. CIN in Week 3, CAR +2.5 at DET in Week 11 (added bonus: Panthers play on Thursday in Week 10, Lions play on Thursday in Week 12) or CAR -3 vs. SEA in Week 12. If instead you think the win expectancy number is closer to the Panthers' true value, it makes a lot of sense to jump on Under 9 wins, even at -140.

I probably fall in the latter group, as expecting the Panthers to get to 10 wins seems a bit much -- I think the Saints and Falcons are a step ahead in that division.

### Houston Texans

• Game-line win expectancy: 9.387 wins
• Win total: 8.5 wins (-140 Over, +120 Under)
• Difference: +0.887 wins

The Texans are the reverse case of the Panthers, a team that sees its Over juiced in the win total market and even higher expectations in the game-line market. Those that buy into a great season for the Texans should hop on Over 8.5 wins. If you're not a fan of paying the juice, the Texans are even odds to make the playoffs, and in what appears to be a weaker AFC, nine wins may well be enough to get them in.

I'm pessimistic on this team, thanks in part to what I feel will be a poor offensive line, so I'd be looking to fade them in the game-line market, where they're perceived as being close to a 9.5-win team and edging out the Jaguars for the best number in their division. Lines I like include NE -6.5 vs. HOU in Week 1, IND +4 vs. HOU in Week 4, DAL +3.5 at HOU in Week 5, JAC -3 vs. HOU in Week 7 and PHI -6 vs. HOU in Week 16.

### Baltimore Ravens

• Game-line win expectancy: 8.781 wins
• Win total: 8 wins (-140 Over, +120 Under)
• Difference: +0.781 wins

I think the Ravens are one of the AFC's sleepers this year, and the win expectancy market seems to agree. While their win total is marked at only eight wins, their game lines have them closer to being a nine-win team. Since I'm high on Baltimore, I agree that there's plenty of value to playing Over 8 wins.

If you're on the other side of Baltimore and think they're no better than an eight-win team, here's a few game lines worth checking out: NO PK at BAL in Week 7 (with Saints coming out of their bye), CAR -2 vs. BAL in Week 8, ATL -3.5 vs. BAL in Week 13 (with Falcons playing on Thursday in Week 12), KC -2 vs. BAL in Week 14 and LAC -4.5 vs. BAL in Week 16.

### New York Giants

• Game-line win expectancy: 6.363 wins
• Win total: 7 wins (-135 Over, +115 Under)
• Difference: -0.637 wins

With the Giants we have an outlier among outliers. Not only do the win-expectancy and win-total numbers disagree, but the win total is juiced the wrong way. That is, if you trust the win expectancy numbers, and in this case I do, as I like the chances of the Giants finishing with six or less wins. Sign me up for getting Under 7 +115.

If you think the Giants belong in that 7-to-7.5 win range or better, here are a few game-lines to consider: NYG +3.5 vs. JAC in Week 1, NYG +7 at HOU in Week 3, NYG +6 at ATL in Week 7 (this game is on Monday, with the Giants playing on Thursday the week before), NYG +6.5 at SF in Week 10 (another Monday night game, this one right after their bye) and NYG +1.5 vs. TEN in Week 15.

### Kansas City Chiefs

• Game-line win expectancy: 7.895 wins
• Win total: 8.5 wins (+100 Over, -120 Under)
• Difference: -0.605 wins

Like the Ravens, here's a team I expect to beat expectations. The Chiefs have had at least nine wins and a point differential of +70 or better in each of Andy Reid's five seasons with the team, and even with the transition in QB, I see no reason to doubt him.

I like going Over 8.5 wins on the Chiefs, and I really like taking them to win the division, where they were +300 at Westgate last week. But the game-line market has them even more undervalued than the win total market, so this is a team I expect to wind up with a great ATS record against these opening lines. Here's the best of the bunch to me: KC -1 vs. SF in Week 3 (they're -1.5 against the Jags at home in Week 5; is SF better than JAC now?), KC +1.5 at DEN in Week 4, KC -1.5 at CLE in Week 9, KC +2.5 at OAK in Week 13 (coming off a Thanksgiving bye) and KC -1 vs. LAC in Week 15 (on a Thursday).

Note: The initial version of this story had a miscalculation on the Ravens line, which has since been corrected.

Our Latest Stories