Week 3 NFL Pick’em Strategy: Two Value-Driven No-Brainers On The Board

posted in NFL, NFL Pick'ems

Week 2 featured a few outstanding performances for our NFL pick’em strategies, along with one area where we took it on the chin. It’s still too early to pay much too attention to where we’re ranked at this point, but as of right now:

  • 5 of our 6 pick strategies (non-confidence points) are in the top 20% of ESPN
  • 3 of 6 are in the top 10% of ESPN
  • 2 of 6 (our Aggressive and Very Aggressive game winner picks) are in the top 1% of ESPN

So overall, we’re off to a solid start. The best news was an absolutely dominant performance by our Very Agressive and Aggressive game winner picks, which both currently sit in the 99.6th percentile of ESPN. These pick sets went 14-2 and 13-3 respectively last week, including the Patriots loss that almost nobody picked, and did amazingly well with confidence points too.

We’ve already received multiple reports of people winning “best score of the week” prizes by playing those pick sets last week, which is great. These strategies are more boom-and-bust, higher-variance approaches for larger pools, and they are fantastic options to increase your odds to win weekly prizes. One or two spectacular weeks a season is what we want those picks to deliver, and they’ve already come through early.

In other areas, our Conservative game winner picks jumped up from the 77th to the 93rd percentile of ESPN, and our Agressive and Very Aggressive against the spread picks pretty much held steady in the 88th percentile after going 8-8. In our comments, we highlighted the Eagles, Falcons, Panthers, Rams, and Seahawks all as strong value picks last week — and all five of those teams won.

The one pain point of Week 2 was our Conservative against the spread picks, which pretty much crapped the bed on ESPN, getting only 4 of 12 right. That’s an uncommonly poor result for us, but some of that performance resulted from getting victimized by slight differences between ESPN spreads and actual spreads. For example, the real Browns line was +7. ESPN decided to make it +6.5 in their contest, instead of +7.5; ESPN makes all lines .5’s so there can be no pushes. A half-point difference is usually not enough to make us change a pick to the other side, and of course, the Browns ended up losing by 7 points and we lost the ESPN pick.

The good news is if you are using our conservative picks and following our advice, you should be in a fairly small pool, and one bad week by no means kills your chances overall. I can’t remember a year where our conservative NFL picks finished lower than the 85th percentile or so on ESPN, so the advice at this point is just to hang in there.

Our Week 3 NFL Office Pool Picks

Just a friendly reminder; here’s where you can find our computer generated picks for NFL office pools:

Also, if you haven’t checked out our NFL pick’em strategy series, do yourself a favor and check it out. It’ll ensure our choices here make a lot more sense.

OK, time for our handy Game Winner table, which summarizes a few teams that look like they could be smart picks for people that need to take some risks. It’s based on our detailed pick’em analysis page.

Week 2 Value Pick Highlights: Game Winners

TeamOpponentAdj Win OddsPublic %ValueSpreadValue Indicator
San Diegovs. Atlanta62%~30%32%-3Odds-On Contrarian
Carolinavs. NY Giants58%~30%28%-2.5Odds-On Contrarian
Houstonat Denver52%~50%2%-1Odds-On Value
Miamivs. New York Jets48%~25%23%+1Low Risk Upset
Jacksonvilleat Indianapolis43%~10%33%+3High Risk Upset
Oaklandvs. Pittsburgh38%~5%33%+4High Risk Upset
Minnesotavs San Francisco29%<5%25%+6.5Long Shot Upset
Kansas Cityat New Orleans23%~5%18%+9Long Shot Upset

There are again a couple great value opportunities this week.

The first thing to point out is tonight’s game. There was a big swing in the odds for this game recently on account of some New York Giants injury news, which very quickly moved the Panthers from one-point underdogs to 2.5 point favorites. You need to jump on that. It’s still a close game that could go either way, but odds are a bunch of your opponents chose the Giants earlier in the week and won’t go back and update this pick to the new favorite. So the value is clearly all Carolina’s right now.

With San Diego at home, and the memory of the Falcons’ win on Monday Night Football fresh in many peoples’ minds, the Chargers are also no-brainer. Atlanta is very over-picked.

Getting into riskier picks for midsize and large pools, taking Miami vs. the Jets looks good. The game’s a tossup, Miami’s at home, yet only 1 in 4 people are picking them.

We’re not overly enthused by either of the high-risk upsets this week, with Jacksonville on the road and Oakland having under 40% win odds. There is significant value on both teams though, so if you feel strongly on either upset, you’ve got a case for picking it.

Neither of the long shot upsets looks worth playing, given safer alternatives that are only a whisker more popular. If you want to take a few educated gambles in a very large pool, which is usually your best strategy, focus on Jacksonville and Oakland instead.

Week 3 Against The Spread Pick’em Strategy

A lot of point spread based pick’em contests publish the games and associated point spreads at the beginning of the week, but give you until the end of the week (or close to it) to submit your picks. This means a great strategy is to look for a game where the official spread in your contest is different from the current spread offered at most sports books, and take the “free points.” In general, the edge you get from those free points is going to be stronger than any lean you have on who is favored to cover, especially if the difference is more than one point.

Below are the biggest line movements from earlier in the week. Highlights can also be found on on the right side of our NFL odds page.

Point Spread Movement Highlights

TeamOpponentOpening LineCurrent LineMovement
Carolina PanthersNew York Giants+1-2.53.5
Miami DolphinsNew York Jets+3+12
Oakland RaidersPittsburgh Steelers+5+3.51.5

The other major source of value in ATS pick’ems is finding teams that a huge majority of your opponents are picking, and then taking the other side. Most Vegas point spreads are pretty efficient, so if 85% of your opponents are picking a certain team to beat the spread, that team is probably incredibly overvalued. And that means you’ve got roughly a 50/50 shot of pulling ahead of a huge chunk of your opponents by picking against them, no matter what your personal opinion is of that game.

Here are five teams where the lines for the Yahoo! ATS pick’em are the within half a point of those found at Pinnacle sports book on Wednesday afternoon, yet under 20% of the public is selecting them to cover:

Point Spread Pick Imbalance Highlights

TeamOpponentPublic Pick%TR Cover OddsCurrent Line Spread
Seattlevs. Green Bay~20%48.2%+3
Oaklandvs. Pittsburgh~15%55.0%+3.5
Minnesotavs. San Francisco~15%49.7%+6.5
Tennesseevs. Detroit~15%50.6%+3.5
Jacksonvilleat Indianapolis~20%54.8%+3

If the lines in your pick’em contest are the same as those shown above, and if you think your competitors will be picking teams in roughly the same manner as the general public (e.g., in line with the Yahoo! contest picking trends), we’d recommend playing all five of these teams. Four of them are the classic home underdog situation.

For advice on other games in your spread-based pick’em pool this week, check out our NFL office pool spread picks page, or our NFL ATS picks page.

As always, please feel free to ask questions in the comment section below. We’ll do our best to answer as many as possible, and we look forward to a fun and interactive college football weekly pick’em advice column this season.

  • TheTinDoor

    Your Point Spread Pick Imbalance highlights are from LAST week’s games.

  • http://www.teamrankings.com TeamRankings.com

    Yikes, forgot to put in the code for the new table. Fixed. Good catch, thanks!

  • Chris

    It also says Pittsburgh Pirates lol

  • Dan

    Pittsburgh “Pirates” under Point Spread Movement Highlights? C’MON MAAAAAN!

  • http://www.teamrankings.com TeamRankings.com

    Hahaha what the hell! Dang, think I need to get some more sleep tonight. Fixed. I wonder what that spread would have actually been though…..???

  • http://www.teamrankings.com TeamRankings.com

    Can’t talk right now, I’m headed out to see the Celtics – Yankees game.

  • Yunier

    I have a request if it isn’t that hard and at hand. What were the picks performance for last year for the 3 different strategies “Conservative”, “Aggressive”, and “Very Aggressive”? It isn’t on the site anymore and I’m trying to see which one is better for a custom league I’m playing on. Thanks in advance.

  • http://www.teamrankings.com TeamRankings.com

    This should be in our database somewhere, but unfortunately not “at hand” right now as our main DB guy is on vacation. So it’s likely going to be a couple weeks before we can get to this. We should just have that data still on the site, but we built the whole office pool section of our site pretty quickly and didn’t really design for showing historical years’ results, etc.

    What are you looking for exactly? Overall record at the end of the season? Record week by week? Just win/loss, or the confidence points assigned to each pick too? Can’t promise anything since I have no idea how this data is stored and accessed, but if what you need isn’t too complex to get, happy to look into it. This blog post contains a general overview of where we were heading into the last week of the season last year:


  • Average Joe

    Hi Tom, I’m having a heated debate with my buddies and I was hoping your expertise could help us settle it. As I’m sure you are aware most pick ’em leagues lock in their spreads days prior to kickoff. My pick’em locked the spread in at Panthers +1. Prior to kickoff the spread had shifted a full 3.5pts to the Panthers -2.5. My argument is that there is a ton of value built into the panthers +1, therefore an average joe like myself who picks winners 50% of the time on a good day cannot pass up on taking on taking the Panthers +1 (mind you we pick only pick 5 games a week) . My friends seem to believe this is a suckers mentality and the line shift does not equate into an advantage. I’m writing this as the Giants are stomping out the Panthers 23-0, but I stand by my claims. Am I insane and or does my argument have some weight?

  • Average Joe

    I just read in your article “A lot of point spread based pick’em contests publish the games and
    associated point spreads at the beginning of the week, but give you
    until the end of the week (or close to it) to submit your picks. This
    means a great strategy is to look for a game where the official spread
    in your contest is different from the current spread offered at most
    sports books, and take the “free points.” In general, the edge you get
    from those free points is going to be stronger than any lean you have on
    who is favored to cover, especially if the difference is more than one
    point.” maybe you could further explain this?

  • Yunier

    I’m just looking for the records weeks by week for each pick group, that’s it. And only for the NFL not college. Thanks!

  • EmJayBee83

    Hey Tom, why was there such a large swing in Carolina vs. NYG pickem (conservative: from -1 to +11) in the course of a couple of days? Having followed you guys for the past year-plus, most “close game that could go either way” are not given 11 or better confidence (or am I mistaken?). I think this would be the kind of non-normal case that might give some interesting insight on the process over-all.

  • http://www.teamrankings.com TeamRankings.com

    That was a weird case indeed. Well, the big reason was a major move in CAR’s point spread on Wednesday (I believe), where they went from 1 point underdogs to 2.5 point favorites. Our models therefore adjusted their win odds up by a decent amount. If you look at our win predictions page now (http://www.teamrankings.com/nfl-win-picks/) that game is/was our 10th most confident winner of the week, which would equate to 7 confidence points in a strategy that ignored pick value and only focused on win odds.

    I didn’t know where they were before the line change, but based on your comment that they started the week at 1 confidence in our office pool picks, we must have been projecting the game as a toss-up.

    So why 11 confidence and not 7, after the line move? I agree that seems a little large of a risk for the conservative strategy, but it had to be driven by value. Check out Carolina on the detailed analysis page (http://www.teamrankings.com/nfl/office-pool-picks/game-winner/detailed-analysis/). When the office pool picks froze on Thu morning, Carolina was both (a) an odds-on favorite to win and (b) had huge value vs. the public (57% win odds as of Thu yet only 22% of the public is picking them).

    A situation like that is rare for a favorite, and exploiting value opportunities is at the heart of our pick’em strategies. You usually only see value like that, especially in NFL pick’ems, on underdogs that are much riskier picks, so our strategy loves a contrarian pick like that. The other example this week is San Diego, and you can see where they rank this week across the three office pool strategies. And the more aggressive you get, the bigger the bets we start placing on the top value picks of the week.

    That whole situation was a little weird with the late line change, which may well have just been public overreaction to NYG’s injury news. The public tends to vastly overestimate the impact of injuries in the NFL, in my opinion, so I was a little suspicious of that line move myself. But my job is to be objective and help our users play the odds. These office pool strategies are based on automated algorithms designed to optimize picks mathematically and take subjectivity completely out of the equation. I’m 100% certain we could refine the logic, and maybe this CAR case indeed provided a situation we should look at again. But still, directionally, I can totally see why our pick’em models really liked this game.

    Obviously it ended up not even being close to a 1 point game or a 2.5 point game, so it didn’t much matter. Win some, lose some…

  • http://www.teamrankings.com TeamRankings.com

    OK, thanks for the clarification. I’ll see if we can get this after next week.

  • http://www.teamrankings.com TeamRankings.com

    Well, let’s put it this way. We’re in the LVH SuperContest ($1,500 buy in ATS pick’em contest, make 5 picks a week) and the Carolina spread was frozen at Pick’em. That game quickly became one of our five plays of the week.

    Now do I feel bad or like a “sucker” after CAR got blown out? Sure, I was pissed. I started rationalizing it (see my response to EmJayBee above) — “Damn, what an idiot, that line move was totally just driven by square public bettors overreacting to NYG injury news and the sports books trying to capitalize on it,” etc. It’s hard not to have 20/20 hindsight after a blowout like that.

    But the bottom line is, points are points. How does a line shift NOT equate to an advantage? So if I offered your friends Packers -10 and Packers -7, they’d gladly take Packers -10 because they think the Packers are so awesome? I’d love to book some bets for these guys! :-)

    That’s an exaggeration obviously, I’m assuming your buddies are saying if you didn’t like Carolina before, why would you like them now, even if you are getting some free points.

    My sense here is that most people are fairly imprecise in their pick-making. It’s “OK, I really like the Packers in this game, the Cowboys in this one. I don’t have a good feeling on this Jets/Miami game, so I’m going to pass.” Etc. So if that’s the context, I can see why they think free points on a game or side they originally didn’t like can seem like a trap. If you think Miami is going to get beaten, an extra point or two isn’t likely to make a difference, right?

    Um, yes, it could.

    We obviously take a more precise approach. For instance, our models gave Carolina a 52% chance to cover -2.5 this week. That’s below the break-even point of 52.4% winners, so it was a one star rating (i.e. expected unprofitable bet). At -2.5, we would not have picked Carolina for the SuperContest.

    At Pick or +1 in a contest, though, Carolina’s odds to cover those numbers frozen earlier in the week are clearly higher. Based on the latest information, the market is valuing them at -2.5, period.

    We’ve developed a model for calculating these things (oops, guess I just let the cat out of the bag, but we intend to launch this on the site in the next few months), and getting 2.5 free points, it moved Carolina’s cover odds up to 56%, which is a great pick, and we took them as a result. It put Carolina in our top 5 cover odds teams based on the contest’s lines.

    Does that difference really matter? Depends on your time frame. Both 52% and 56% percent are close to 50/50 — you’re going to lose almost half of those picks either way. In a contest where you pick 5 games a week for 17 weeks, you would expect a 4% advantage on one game a week to add up to an extra 0.68 wins by the end of the season. So it’s not likely to make a huge difference. But over the long term, it most certainly will. In the long term, a 52% winner is a losing gambler; a 56% winner makes a great profit.

    Regarding the comment in the article, what we’re saying is that point spreads are pretty efficient overall. Even if you’re right in thinking that the Giants are the more likely team to cover the spread, in reality, that probably means that Carolina has a 47% or a 45% chance to cover, or something around there. Add the impact of getting three free points to that percentage, though, and all of a sudden Carolina should be the one with more than 50% odds to cover.

    Get some smarter friends, man! Only kidding.

  • EmJayBee83

    Thanks for the explanation. I was right, it was interesting (at least to me ;)).