Week 9 NFL Survivor Strategy: Which To Trust, Our Numbers Or Our Gut?

posted in NFL, NFL Survivor Pools

*UPDATE #2* Since we wrote this post, the spread for Oakland (our official pick) has moved from -9 to -7. Because of this, our projected win odds and the projected value of picking Oakland have both dropped. We now feel that Dallas is a better pick. As long as the Oakland line is -7.5 or worse and the Dallas line is -11 or better, we’d recommend picking Dallas. We will NOT be changing our official pick below, as we feel it’s basically “locked” already. So, the text of the post still says Oakland, and that’s what we will grade ourselves on this week. But our current suggestion is to choose Dallas.

Welcome to the Week 9 installment of our series of Survivor advice columns, where we use a data-driven strategy to get an edge in NFL Survivor contests. This post includes analysis based on NFL predictions from our algorithmic Team Rankings models and other relevant data.

This discussion below is also meant to complement our Survivor Predictor tool, which provides a grid view of win odds predictions by week, from now until the end of the season. This presentation of prediction data will help you quickly grasp the current week’s best Survivor pick options, and will explain our logic in making our official pick.

Week 8 Survivor Strategy Review

We just survived by the skin of our teeth last week, as our pick of the Giants had to come from behind in the fourth quarter in order to beat the Dolphins by three. Still, a win is a win.

The only notable upset of the week was St. Louis taking down the Saints, which eliminated 13% of Yahoo contestants … and made us feel pretty smug about our mini-rant in last week’s column about how the line was way too high. Besides the New Orleans loss, we came close to seeing another third of the pool eliminated, but the Ravens managed to claw back from three TDs down to top the Cardinals.

Overall, 85% of Yahoo contestants survived, which means we didn’t gain much immediate value. We did, however, conserve all of our future value by using a team that is worthless from here on out. It was a stressful week, but it turned out fine in the end.

Week 9 Survivor Decision Factors

Time for the heart of our post, the table showing the factors that influence our decision:

TeamOpponentSpreadTR OddsYahoo! Pick %Future ValNotes
Dallasvs Seattle-11.584%11.6%1best Wk12 option
Houstonvs Cleveland-10.585%27.5%1PICKED
New Orleansvs Tampa Bay-9.084%1.7%0PICKED / best this Wk?
New Englandvs NY Giants-9.083%2.3%2PICKED
Oaklandvs Denver-9.078%5.5%0now only chance
Philadelphiavs Chicago-9.072%1.4%3best Wk17 option?
Atlanta@ Indianapolis-7.071%37.0%2why so popular?
Green Bay@ San Diego-5.569%0.7%5
Kansas Cityvs Miami-4.065%8.2%1
San Francisco@ Washington-3.565%2.7%6
Tennesseevs Cincinnati-3.064%0.1%0
Pittsburghvs Baltimore-3.058%0.2%2PICKED
Arizonavs St Louis-1.567%0.3%0
Buffalovs NY Jets-1.057%0.2%4

Future Val = the number of future games where the team is expected to have win odds of 75% or greater, based on our NFL Survivor Tool.

Teams We Already Picked: Houston Texans (WIN), Pittsburgh Steelers (WIN), San Diego Chargers (WIN), Tampa Bay Buccaneers (WIN), New England Patriots (WIN), Cincinnati Bengals (WIN), New Orleans Saints (WIN), New York Giants (WIN)

Weighing the Options

Dallas Cowboys (vs Seattle Seahawks) — This is the safest pick of the week, in terms of Vegas line and TR win odds, yet they are only the third most popular. That’s likely because people are trying to save them for week 12 (when they host Miami) or already used them in Week 7 (vs the Rams). This is a great pick for pools with very few people left (under 10), where future value is less important. In a large pool, we may want to save them, but they are definitely going to be a tempting choice.

Houston Texans (vs Cleveland Browns) — The big line and high win odds are definitely plusses, but they are the second most popular team, in a week with a lot of decent options. We’ve already picked them, but even if we hadn’t, we’d pass and root for the Browns to knock out 25% of our opponents.

New Orleans Saints (vs Tampa Bay Buccaneers) — Now we’re talking. Our win odds have this as one of the safest picks of the week, and Vegas sees it just a notch below the Cowboys or Texans. Yet absolutely nobody is choosing New Orleans, probably because hardly anybody has them left. This is also their easiest remaining game, so they’re likely the best option this week, if you haven’t used them yet. Unfortunately, we already have.

New England Patriots (vs New York Giants) — The Patriots are pretty similar to the Saints, except that they could be useful in Week 13 (vs Indy) and maybe Week 16 (vs Miami) if they haven’t clinched their playoff positioning yet. So, they’re behind the Saints because of future value, and they’re behind the Cowboys and Texans in terms of the Vegas line. Not a particularly compelling pick, but not terrible. Regardless, we’ve already used them.

Oakland Raiders (vs Denver Broncos) — This one is interesting. Our models have Oakland as a tad less likely to win than any of the above teams, though Vegas puts them basically even with the Saints and Patriots. They’re also a touch more popular. However, their big plus is that they have no future value. We’re not to the point yet where we can start burning the teams we’ve been saving willy nilly, so they could be a good choice. Of course, the big question is how will Carson Palmer do now that he’s had a bye week to prep himself and learn the offense? Picking Oakland might be scary this week.

Philadelphia Eagles (vs Chicago Bears) — The Eagles are similar to the Raiders in terms of immediate value — they are less popular (good), but also slightly less likely to win according to our models (bad). The biggest difference is that the Eagles could be an option next week (vs Arizona) depending on how the public pick percentages play out, and also may be the best bet in Week 17 (vs Washington) since there’s a decent chance that game will matter to them. So we’d lean towards Oakland over Philly here.

Atlanta Falcons (@ Indianapolis Colts) — Wow, did people completely forget what happened to the Saints last week when they visited the Rams? This set up seems very similar to us, and we’re very surprised that this game is so popular. There are four teams (Cowboys, Texans, Saints, Raiders) who are better than Atlanta in every single metric we list above. STAY AWAY.

Green Bay Packers (@ San Diego Chargers) — There’s a reason under 1% of the public is picking them. They have far too much future value to burn them when they are only favored by 5.5 points.

Kansas City Chiefs (vs Miami Dolphins) — Why???? Not sure what those 8.2% are thinking. The Chiefs and all those below them are definitely getting our STAY AWAY label. There’s just no reason to take on this much risk.

Official Pick: Oakland Raiders over Denver Broncos*

[Better pick, for those who still have them available: New Orleans Saints over Tampa Bay Buccaneers. Also, please see the update at the top of the post.]

This is an interesting week. Because so much of the public is weirdly picking Atlanta, any of the teams favored by 9 or more should end up being a positive choice for us. That said, deciding which one of those we should make the official pick is pretty tough:

  • We think New Orleans is the best choice this week, but we used them earlier.
  • Dallas has the most immediate value, and should be your pick in small pools where future value is much less important. But they look like they’ll be very useful in Week 12, which is important in large pools.
  • Oakland is a nice choice because we get to save Dallas (and others), but we are assuming a bit more risk this week in return.
  • We’ve already used Houston and New England, but we’d probably pass anyway, since Houston’s public pick percentage is so high, and New England has some future value.
  • Philadelphia is very similar to Oakland, but has lower TR win odds and more future value. We may want to use them next week, depending on how the public pick percentages shape up, and they could be useful in Week 17. So, they’re a pass.
  • Beyond that, it gets pretty dicey, so no reason to consider any other teams.

It basically comes down to Dallas or Oakland. Should we take the pick with more immediate value this week (Dallas), which will force us to be riskier in Week 12? Or should we take on some more risk now, and save Dallas?

The best remaining alternative to Dallas for us in Week 12 is Atlanta. The Falcons won’t be the top choice any other week, so we can look at this in a kind of vacuum, ignoring all weeks but this one and Week 12. Without getting into the math here, we’ll just say that according to projected power ratings and Vegas lines, the drop off from Dallas to Oakland this week is less than the drop off from Dallas to Atlanta in Week 12.

Of course, we should discount future value because of uncertainty, which shrinks that difference. Plus, we’re scared of backing Carson Palmer. But the bottom line is that — for people in large pools — the numbers are telling us to pick Oakland.

Now, our gut is getting queasy just thinking about relying on Carson Palmer to carry us to a win. But our whole philosophy here is that you have to trust the numbers, and try to ignore your subjective whims. And it’s not like it’s just our models that are high on Oakland. The market has set the line at -9, and that’s a much better measure of the overall Carson Palmer / Tim Tebow / Oakland bye week impact than we can hope to come up with on our own. So, with that bit of reassurance from the invisible hand of the market, our official pick is the Oakland Raiders. This was definitely a close choice, however, so we certainly wouldn’t say Dallas (or even Philly) is a bad pick.

*UPDATE* A couple of comments below pointed out something that we overlooked here — we’ve already used Cincinnati, but if you still have them left then you can use them in Week 12 instead of saving Dallas for that week. That removes the main reason not to take Dallas. Taking Dallas this week and Cincy in Week 12 is better than Oakland/Dallas or Dallas/Atlanta. So if you haven’t used the Bengals yet, Dallas looks like the best pick for you.

*UPDATE #2* Given the line move in the Oakland game, our new suggestion is to pick Dallas. For more details, please see the note at the top of this post.

Normally, we answer a lot of questions in the comment section. This week, however, we’re going to be heading out on the road, so we likely won’t be able to get to as many questions as usual. Of course, as always, feel free to ask. We just can’t guarantee you’ll get a timely answer this week. Sorry about that!

  • http://twitter.com/gunz4sale Ray P

    I have a question, it’s just myself and another guy and he’s already played Houston. Should I play Houston or play Oakland?

  • Winter

    What about Houston in the decision write-up?

  • http://www.teamrankings.com/ David Hess

    Wow, that was fast! We covered Houston here:

    “We’ve already used Houston and New England, but we’d probably pass anyway, since Houston’s public pick percentage is so high, and New England has some future value.”

  • Emerald Murphy

    Given that Darren McFadden is nursing an injury (walking boot / cast yesterday at practice) and Carson Palmer is still only 3 weeks removed from retirement (although probably still significantly ahead of Tebow), how does McFadden’s status impact the recommendation (or is that likely already factored into the spread (haven’t seen if it moved this week). Also, if I have Houston left in a pool of 3 people remaining, how does the very small pool size change things (if at all) (used AZ,  Pitt, SD, TB, NE,
    GB, NO, NYG)

  • http://www.teamrankings.com/ David Hess

    Looks like the spread hasn’t budged:


    Honestly, running back injuries are usually overrated (look at KC without Jamaal Charles).

    As for pool size, with 3 people yet, you just want to try to survive this week and hope your opponent gets booted. Take the safest pick each week, unless there is a clear future value concern.

  • http://www.teamrankings.com/ David Hess

    Gotta run, guys. Hopefully I can answer some questions this evening…

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

    Clearly Houston.

  • Emerald Murphy

    thanks for the great columns, always a fun read each Thursday, and advice has been much appreciated, thanks for the quick response, and we should be in for some interesting weeks ahead as the picking gets thinner (for our competition hopefully)

  • Cory Fantasyteam

    Oakland raiders arent playing the dolphins. I know you know this, but you have it wrong in the official pick bold heading.

    I have 60 people left in my pool, and still have Houston left.  Im really trying to make it to week 10 since im one of the few left that hasnt used green bay, i have a good feeling about week 10 so should i play it safe this week?

  • PATS1

    Oakland Raiders over Miami Dolphins (WHAT WEEK ARE YOU IN)

  • AL

    There are still 24 players in my pool.  This is out of 83.  So there’s still about 30% still standing.  Is this still considered a “large Pool”.  It not a “Small” poll (under 10 people).

    Also it sounds like McFadden is out for Oakland.  Would Oakland still be a better choice than Dallas in my situation?

    Also, do you ever take injuries into consideration in your TR Prediction analysis?  Or is it strictly based on past stats?

  • Geddy1001

    Have 5 people left in my pool including myself.  Only myself and one other person have not picked Dallas. Is this where we go with the safest pick (Dallas), no matter of future value?  Im thinking this  pool may goto 12 weeks which is where Dallas would be the best pick BUT I guess I would have to get there first huh? :)

  • Chris


  • Chris

    Jeesh, sorry about that mess.   Was trying to point out who has been used and it came out a mess.  These are my picks so far week1 = sd /2 = det/ 3 = ten/  4 = ten/ 5 = cin(don’t ask, they were a 1 pt. dog.) / 6 = pitt/ 7 = n.o./ 8 = giants and minor atrial fibrilations.  My question is with 14 left out of 105, who would you take this week?  Again, sorry about that mess prior to. 

  • chris

    4 = tb not tenn again.  :)

  • Golfpro Triplets

    I am in a super large pool, with about 800 or so remaining, and I still have 4 entries alive (out of 6 that I started with).
    I have New Orleans avail in two of them, Hou is gone in all 4, Dallas avail in 3, Oak avail in all 4.  My intial feeling is to take the two with N.O. and two with OAK.  But since we have not seen enough large upsets, I may need NO come play-off time.  What are your thoughts on VERY large pools and saving playoff teams for week 18 and beyond.

  • MO

      “The best remaining alternative to Dallas for us in Week 12 is Atlanta.” 

    What about Cincinnati?  The model suggests they have more value than Dallas or Atlanta in week 12… Could you make a case for taking Dallas this week, and Cincinnati in week 12?

  • Ryan M.

    Hi Dave,

    13 people left in my pool, near your cutoff of 10. Every remaining player has used the Saints. All but three other guys and myself have used Dallas. Only four have used the Texans and five have used the Pats.

    I could pick either Dallas or Oakland… Leaning heavily toward the Cowboys. Would that be your choice?

  • http://www.teamrankings.com/ David Hess

    Cory, if Week 10 looks to be a week of many eliminations for your pool, it could be smart to just survive this week. I think Houston is a good choice in your case.

  • http://www.teamrankings.com/ David Hess

    Thanks. Brain fart.

  • http://www.teamrankings.com/ David Hess

    The TR predictions don’t directly take injuries into account, but they do use current Vegas lines. Those should help to indicate if there’s been a substantial change in the quality of the team.

    As for your pool size, that’s medium — not clear which way you should go, there. Honestly, it’s a toss up for you, so I’d go with whichever pick you subjectively like better, out of Dallas and Oakland.

  • http://www.teamrankings.com/ David Hess

    That’s a rough one. If you think that other guy is going to pick Dallas, and nobody will pick Oakland, I’d go Oakland. If you think the other guy WON’T pick Dallas, I’d go Dallas.

  • http://www.teamrankings.com/ David Hess

    Ooh, that is leaning towards a small pool. I think I’d still go Oakland, but it’s very close. Oakland, Dallas, Philly.

  • http://www.teamrankings.com/ David Hess

    Yikes, playoffs complicate things immensely. I think the Saints become much less attractive, as they stand a good chance to host a playoff game. Oakland looks much better then, as well. Plus, in large pools, staying away from the most popular teams is good.

    I think you should save NO in at least one pool. I’d probably go NO, DAL, OAK, OAK. But that is COMPLETELY off the cuff. No spreadsheet in front of me to do actual math. Sorry.

  • http://www.teamrankings.com/ David Hess

    Yes, we should have mentioned that. We already used Cincy, but if you still have them available in Week 12, Dallas becomes a great pick this week.

  • http://www.teamrankings.com/ David Hess

    Yeah, I think so. You are in a pretty small pool, and it sounds like Dallas won’t be popular. Like we said above, DAL/OAK is very close this week. If you strongly favor DAL, for good reasons it seems, I’d go with them.

  • Killerzoe

    Well..80 left…looks like the current top choices in the pool so far are Houston, Dallas Atlanta….i used Oakland and the Saints and Dallas…of the teams left..im leaning towards Houston, Atlanta and Philly?Houston..is my gut..even though majority so far is leaning in that direction…any thoughts?

  • http://www.teamrankings.com/ David Hess

    I’d lean Philly over Houston, but it’s very close. The most important thing is to stay away from Atlanta.

  • http://www.teamrankings.com/ David Hess

    Oh, wait. Are you saying Houston is the top choice in your pool (HOU, DAL, ATL)? If so, then definitely Philly over Houston. Gotta root for the upset there.

  • Marco

    David, thanks for your posts. Don’t let those bastards who yelled at your last week get you down. It’s just a blog, no one forced them to follow what you said. I just wanted to say thanks for the analysis but I disagree with your pick for Oakland. It’s a very risky play because the QB situation is unknown and McFadden is hurt. I hate to jump on this Tebow fan wagon but it seems like that guy plays hard at the very end and I think they have an offense that can drive the ball down at least into FG position. In regards to the Raiders, they have no passing so no matter how good their RBs are, they may not even score this weekend.

  • http://www.teamrankings.com/ David Hess

    Marco — Yeah, if I was going based on what my eyes see, there would be no way I’d be picking the Raiders. But a valuable lesson was reinforced for me a few weeks ago. KC was playing at SD after losing Jamaal Charles, and I thought they were going to get destroyed. I thought our models were crazy for making KC +13.5 a 3-star pick. I mean, I was thinking SD was LOCK CITY. Of course, KC covered and almost sent it to OT. The point is, it reminded me that the market and our models can process WAY more information than I can (or any other single person). So if they both agree that Oakland is a reasonable pick, I’m going to swallow my pride and trust the numbers.

    That said, this week the numbers say that it’s SUPER close between Oakland and Dallas. There is absolutely nothing wrong with taking Dallas (or even Philly), if you feel strongly that the Raiders are bad news.

  • MJair23

    I’m in a double elimination survivor pool (i.e. you can lose once and you’re still alive…you still can only use every team only once win or loss.  Know that if the pool ends up running the whole season no losses beats anyone with one loss automaticaly). Right now there are 66 people with no losses and 285 with one loss (351 total “still alive”).  48 of the 66 people with no losses have already used Dallas (not sure about how many of the 285 people with one loss have used him, but let’s assume it’s a similar percentage as the people with one loss).  Taking Dallas becomes a lot more appealing under this circumstance no?  I have not used Oakland or Houston yet, so have them available. thoughts?

  • http://twitter.com/drumzan drumzan

    Why even consider Philly?  They’ve had ONE good game this year and that was last week.  Don’t forget last year Chicago beat Philly.  Chicago’s O-line is playing much better.  I just don’t see it.  And I also don’t see Oakland being a good pick.  I watched the entire Oakland-Cleveland game because I picked the Raiders in my survivor pool that week.  After Jason Campbell was knocked out of the game, and without Run DMC, the Raiders stunk and almost gave up a huge lead in the 4th quarter…to Cleveland!  I just don’t think 3 weeks is enough prep for Palmer.  As bad of a team as Denver is and as good of special teams and defense that Oakland has…I wouldn’t touch this game with a ten foot pole.

  • Steelers GO!!!

    In short, in one pool I have already won. Now I just wanna see if I can survive as long as possible.
    My picks:
    HOU, PIT, SDG, TAM, NWE, GNB, NOR, BALI am now in another pool, with every team available. Your recommendations? 

  • Justin Geoghegan

    I’m in a small pool with seven left, BUT it’s a 3-strike pool. Four opponents have 2 strikes and 2 opponents have 1 strike (I have 0 strikes, thanks to your ongoing advice.)

    Would you still recommend Dallas in this small pool, or go with Oakland since there’s a chance we might still be around in Week 12?

  • Steelers GO!!!

    Long story short, I have won one bracket. Now I just want to survive as long as possible.

    I just started another one. Everybody will have every team viable now. 

  • Wile

    The analysis as always is very helpful. I have a question on the numbers. Your answer can wait until you get back because it applies to all the analysis. I get putting off Dallas until week 12 and not taking Houston because of how popular the pick is. My question is on Oakland. When you are doing the calculations, I think it is based on the team’s past performance. Given that, the sample size of Tim Tebow as a starter is small and of Carson Palmer as a Raider is almost nonexistent. I get that the spread will take some of this into account but in cases like this would it be useful to compare quarterback stats or some other offensive metrics? Maybe I just don’t understand the models well enough.

    I am going to go with you and pick Oakland. You’ve gotten me this far.


  • Thaddeus


    I’ve been following your blog for most of the season, but haven’t felt the need to post until now.  This week is a tough call.   I’m in a very large pool with 500+ remaining (out of 20,000).  Starting on week 14, 2 picks will be needed per week.

    With ~50% of the remaining entries in, the pick percentages for this week are:
    Texans 39%
    Falcons 19%
    Cowboys 13%
    Raiders 10%
    Chiefs 7%
    Eagles 3%
    Saints 3%
    Patriots 1%

    I’ve used: TB, NO, NYG, NE, SD, Balt, Pitt (contest started week 2)

    I guess I have three viable options: Dallas, Oakland, Philly.  (I obviously have Houston available, but they are being over picked.) As you can see in my pool, the pick %ages between the Dallas & Oakland are much closer than Yahoo.  Addionally, I have Cincy left so I have a viable week 12 alternate.  It doesn’t look like Dallas will be a good option any week but this one and #12.  Philly could potentially be used 14, 15, or 17 (when I need to make 2 picks/week).  Oakland is obviously worthless if not picked this week.

    So, my question is does having Cincy & the closer %s tip the balance to Dallas?


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

    Wile (and others) — it’s Tom from TeamRankings here. I’m going to chime in on one topic quickly, but leave the more detailed answers to David, who is much more of an expert of the survivor math than I.

    There’s clearly some reservation out there in comment land about taking Oakland this week, which I totally understand. I think David made it clear in the post that personally, it sort of makes our stomach turn giving Oakland the nod this week for bigger pools.

    With that said, pretty much all the analysis we do here at TR is based on trusting the numbers from the most reliable long term predictors. For us, those are the Vegas lines, and our own algorithmic models — which, as David said, process a heck of a lot more information than any human football-watching brain ever could.

    When people don’t agree with a pick of ours, we tend to hear a lot of logic like, “How could you ever pick team X? Did you see how crappy their new QB played last game? Did you see how they folded in the 4th quarter last week?” etc.

    I’ll come right out and say that while I enjoy watching NFL as much as the next guy, I believe that basing your handicapping strategy on opinions you form by personally watching games (and more specifically, on how the TV network chooses to present the game, what angles they show you, etc.) is incredibly risky.

    Sure, there are probably a few people out there who have literally watched every single NFL game this year and can process all that info into some sort of minor edge on some games each week. But I would argue that the MINIMUM requirement to succeed consistently at that approach to handicapping would be to watch every single NFL game without exception (and probably multiple times each), and spend a whole lot of time adjusting stats and team ratings based on what you felt were really lucky plays, times when the defense just gave up, etc. And hardly anybody is doing that. 

    What we know is that Oakland is -9 / -350 on the money line (close to 75% implied win odds adjusting for some vig) based on everything the market currently knows about McFadden’s likelihood to play, whether Carson Palmer really has anything left in his tank and can get integrated into the Oakland offense, etc. They’re also about 78% to win by our models. So if you disagree with our analysis, you’re taking a position that says, “I know more than thousands of bettors laying real money on this game do about Oakland’s chance to win this game, and I also know more than a computer that takes into account much more data than I ever could process in my head does.”

    Which is fine — that’s what makes sports fun, being opinionated and defending your opinions. But that’s not what we do here.

    Regarding our models, they do not explicitly adjust for recent injuries, QB changes, etc. It pretty much impossible for a data-driven system to predict with a high degree of confidence how a QB like Palmer is going to perform in his second game with the Oakland offense. So we have two options in terms of model design — either insert some sort of subjective logic / assumptions based on how we THINK he is going to perform (maybe based on some likely irrelevant performance stats of how he did with the Bengals in recent years), or just do nothing, and rely on the Vegas odds (which are inputs into our models) to assess the likely impact. We opt for option B — we trust Vegas odds far more than our subjective opinions.

    Because Vegas odds are only one of many inputs into our models, though, this does sometimes mean that our models probably underestimate the impact of recent major injuries, personnel changes etc. But here we have a case where Oakland’s Vegas implied odds AND our model odds are both very close, so we don’t have much reason to question them. If our models said Oakland 80% win odds but the Vegas implied odds said Oakland 55%, that would give us pause. But that’s not the case here.

    Finally, the argument that I think some people are making is that “Yeah, that may be true about the Vegas implied win odds, but there is just so much uncertainty about that game.” In other words, there’s an implication that the Vegas lines are likely to be less efficient or trustworthy for this Oakland game than for other games this week without so many personnel question marks.

    That’s a legitimate question to raise, and probably deserves more study if there are enough similar situations in recent history. How accurate HAVE Vegas spreads been, in fact, when home Team A had a new QB who was approximately as good as Carson Palmer and their star RB was out, and away Team B had a new QB that was approximately as good as Tim Tebow — and Team A was pretty good before these QB/RB issues and Team B was generally pretty bad before its QB issue?

    As you can tell, that’s a pretty darn complicated scenario, and there’s almost certainly not enough historical data to evaluate it objectively, using hard data, with any level of confidence. But I can tell you we have done some research into other types of situations — e.g. early season games — where conventional wisdom among bettors is that lines are less efficient than usual. And we’ve found that wisdom to be largely untrue. Vegas lines have recently done pretty much just as well at predicting games over the first few weeks of the season than they have done in later weeks, for example.

    So we’re back to the same old question…do we go with the market and the computer’s numbers, or do we doubt them based on our subjective assessment of what’s going on? And our answer is, as always, the former.

    Sorry for the novella here, but hopefully it gives some more info on our approach. Personally, every time my mind drifts toward fear about how the Raiders could just totally suck now, I also realize I need to remind myself that this is a team that should have lost to Miami, who absolutely sucks, and got completely smoked by Detroit at home. And they’re playing on the road against a team whose original starting QB (Jason Campbell) is not going to be confused for Aaron Rodgers any time soon. McFadden is good, but one player does not a team make (except perhaps a star QB).

    But of course, if anyone out there is going to pick Oakland and then blame everything that’s wrong in their life on Team Rankings if they lose, then DEFINITELY DON’T PICK OAKLAND and just go with your gut! :-)


  • chris

    Thanks Dave,  I’ll roll with the “Carson Show” this week.  Any reccomendations on my future value?  Haven’t used Hou, GB, NE, SF, Buff, Bears, Cowboys.  Any reccs on how and when to use them?

  • Wile

    Thanks for the detailed response. As I said I am going with Oakland because your analysis has been excellent all season. I am numbers person and have great confidence in your models. My post was to get a better understanding of the inputs and metics used to measure performance. Win or lose – it is still better to have system otherwise as you point out it is just a subjective call.

    Thanks for the explanation.


  • Killerzoe

    Philly huh? hmmm..what about the 49ers? 

  • Killerzoe

    1Texans24.2%82Falcons18.2%63Saints18.2%64Cowboys12.1%45Eagles9.1%36Patriots6.1%27Chiefs6.1%28Raiders6.1%2here is the breakdown…..so far….33 of 80 are in…

  • http://www.teamrankings.com/ David Hess

    Thaddeus — Yep, you nailed the analysis there. Having Cincy available for Week 12 means you don’t need to worry nearly so much about Dallas’s future value. And the pick % between the two teams is pretty close, while Dallas has better win odds — so Dallas is definitely the better pick in terms of immediate value this week. I’d definitely go Dallas.

  • http://www.teamrankings.com/ David Hess

    Hey guys, waiting for my flight to board (San Diego –> Cincinnati), so I’ll try to get through a few of your questions while I sit here in the airport…

  • http://www.teamrankings.com/ David Hess

    Chris — I don’t have a ton of time now to look into a Q like this, but I can tell you that what I’d be doing is checking out our Survivor Tool: http://www.teamrankings.com/nfl/tools/survivor-predictor/

    I use that page sort each week and see who the best picks are, and try and plan a path that lets me pick each of those teams when they have 80%+ odds if possible.

  • http://www.teamrankings.com/ David Hess

    Great essay, Tom! This does a good job of laying out what we think pretty much any time somebody asks “What do you think about this game?” or “What about this situation, should that impact our pick?”

  • http://www.teamrankings.com/ David Hess

    49ers seem like they have too much future value to use this week.

  • Steelers GO!!!

    I would like help with this.

  • http://www.teamrankings.com/ David Hess

    MJ, do you have 1 loss or no losses? If you have no losses, I think you should basically treat this like a normal Survivor pool with 66 people left. Chances are good that at least somebody from that group will make it to the end without a loss.

    In that case, it sounds like 18 of 66 (27%) have the option of using Dallas. It seems like ending up with the same fraction as Yahoo (11%) seems pretty likely, so not sure the analysis is much different than what I laid out in the article. Tough choice between Oakland and Dallas. One question — do you have Cincy left? If so, then you don’t need to save Dallas for Week 12, and I’m use the Cowboys this week.

    If you already have a loss, on the other hand, you need to be taking more risks and hoping that the popular teams get upset. In that case, I think I would go with Oakland and root for a Dallas loss.

  • http://www.teamrankings.com/ David Hess

    drumzan — See Tom’s long response above. One thing I’d note, though, is that looking at how a team did in the game where they lost their QB is not usually going to be all that useful, as they haven’t had time to game plan around his absence yet. Ditto for the game against KC, when they only had a couple days to practice with Palmer.

  • http://www.teamrankings.com/ David Hess

    Nice, you haven’t used Cincinnati, that means you can go ahead and use Dallas this week, and then use Cincy in Week 12.

  • http://www.teamrankings.com/ David Hess

    Justin — If you make it to then end with no strikes, and everybody else has 1 or more, do you win? If so, I’d totally ignore public pick %, and just plan the ideal route through the rest of the year, with help from our Survivor Tool:

    My guess is that Oakland would be the ideal pick in this situation, but not sure.

  • http://www.teamrankings.com/ David Hess

    OK, time to board. Good luck!

  • Steelers GO!!!

    Thanks. What about the pool where everyone has every team available? 

  • MakaveliTheDon

    38 people left in my pool. 11/38 have already picked HOU, 8/28 have already picked OAK, 0/38 have picked ATL. These are my 3 best picks for this week as I have already selected AZ, DET, PITT, GB, NO, NJ, DAL, NYG thus far through 8 weeks. Considering this, would HOU be my best pick for the week? I am assuming this since the least amount of people can pick HOU this week, because they have the biggest point spread, and because they seem safer playing at home against a non-division rival than OAK does playing at home against a division rival with their top running back McFadden out and a starting qb who was on the couch a few weeks ago and sucked his last few seasons in cincy and in his last game with them. Appreciate your opinion in regard to my pool breakdown. Thanks!!

  • MakaveliTheDon

    meant 8/38 have already picked OAK, I mistakenly wrote 8/28

  • Daniel

    While I agree with the pick on here as I also thought to save Dallas for week 12 I am a little worried this might be strike 3 for me(McFadden is out so Palmer better have learned things!)  If not then I will be in great shape going forward as almost everyone has already picked Dallas in my pool and those who haven’t will probably do so this week so in week 12 I have a good shot at being one of the only ones with that pick. 

    Starting next week there is a Thursday night game!  If I’m still alive I will have to get my pick in on Wednesday nights from here on out with Thursday games.  Will you at all be posting these picks earlier in the week due to Thursday games or no?  Thanks

  • Chris

    Hey David.

    I understand that vegas listing the Raiders as 9 point favourites is a big factor in the recommendation this week. The line has dropped to 7 points, and threatens to drop further. How low does it have to go before Dallas overtakes it as this week’s best pick?

    Thanks for your time.

  • Jay

    hey TR guys!

    been following you guys pretty closely this season. i’m in a pool where you can pick any team each week, just need one winning pick of the 32 teams, so the trick is picking your spots to go against the public. the pool started with 576 people and there are only 69 left (thanks to your giants pick last week!). any suggestions for me this week.


  • http://www.teamrankings.com/ David Hess

    Chris — I think this is enough of a move already. It was super close before, so with a line of 7, I’d recommend Dallas. I’ll add a note in the post.

  • http://www.teamrankings.com/ David Hess

    Daniel — Yes, we will most likely post the article on Wednesday, due to the Thursday games.

  • http://www.teamrankings.com/ David Hess

    IMPORTANT NOTE: Due to the line change in the Oakland game (from -9 to -7), we now feel that Dallas is a better pick than Oakland. The two were very close before, and this line move pushed us away from Oakland. Our official pick will remain Oakland, but our new most up-to-date recommendation is to pick Dallas.

  • JMB

    For all you Oak worriers due to the McFadden absence, relax.  There’s a reason Oak has 2nd best rushing attack in the NFL and that’s because of a good OL and RB depth.  Remember Michael Bush?  Yeah..him.  The guy who has topped 100 yards pretty much every time he’s subbed for DMac.  Remember, Den is porous on both sides of the ball and has a very undersized front 4.  Oak will pound the run game and will ask Palmer to do very little.  This will not be the same team that faceplanted against KC.  The NFL is about match-ups and Oak owns this one versus Den.  Sure, Tebow may make some plays, but let’s not talk ourselves into a Den upset here.  

    Now, given Teamrankings relies (and rightfully so) on models and vegas odds, I would agree that the Dallas pick is a “safer”pick now that McFadden is out.  But if for some reason you are stuck with Oak this week, don’t sweat it…you’ll be ok.

    I say “safer” because as you all know, there are no locks (remember, Dal plays the same Sea team who went to NY and upset the G men when they were the biggest odds-on favorites for the week).  

    Good luck!  

  • Deedog44

    Hopefully Oakland will still come through since I took them.  I mean after Tom’s compelling statement, I had no choice!  His words left me in tears….ok…I’m exaggerating.  But his argument was well-written. 

  • Anonymous

    In a pool that restarted in week 6 and having taken CIN, DAL, & NYG, it basically leaves me with NO, HOU, and OAK to choose from.  With 3 dozen players left and it continuing throughout playoffs, OAK definitely gets the lean to conserve future value even without RunDMC.  Does that outweigh the advantage that NO or HOU may have for this week though?  Gotta win to stay in right?

  • Thaddeus

    Thanks for your help.  I picked Dallas and survived.  With KC & Oak falling, looks like another 20% or so of my pool will be out. Hopefully the column continues even though it is officially out.

  • Cory Fantasyteam

    Even though your official pick didnt win, are u going to keep up the predictions for those that went in a different direction?

  • Alfredothebarber

    AND this is why you never “save” teams.

  • Killerzoe

    well…another week through….Houston with the win…17 or so knocked out…now down to 64 i believe left…..I know its early…but leaning with the Eagles for week 10…

  • JMB

    Well – TeamRankings does it again!  While I felt pretty good about Oak taking care of business, they were the riskier pick and sure enough, they faceplanted.   Great call by TR to recommend Dallas over Oak last minute.  Saved me in my pool.  Raider coaching staff gets an F for trying to force Palmer in to throwing the ball all over the yard when all they needed was to ride Bush all game.  Guy had 70 yards rushing in first half…26 the rest of the game.  Unbelievable

  • Deedog44

    Damn you Tom!  You’re too good a writer!  Wish I had gone with my gut feeling…And believe me….that gut is getting big!  I had fun looking like stinkin’ genius for 8 weeks especially when I relayed the possible Seahawks over Giants upset.  But despite the fact that I hate how much pub the guy gets….I GOT TEE-BLOWN OUT OF THE WATER!  Hopefully I can make a run in my other pool of picking the winners….

  • http://twitter.com/drumzan drumzan

    What did I tell everyone!  Don’t pick Oakland, don’t pick Philly. 

  • Mijmarti

    Are you guys going to do a post before the Thursday game?
    Love your data driven analysis!!

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

    Sorry Deedog :-) We figured a bunch of our readers would either not be able to switch their pick by the time our line change update was posted, or would just miss it. The bottom line is that at the time we originally wrote the post OAK looked barely best, and we stand by that. No game is ever a lock.

    For those who did follow our advice this year but just got knocked out, all we can say is, odds are you did a heck of a lot better than most everyone else in your pool. Actually winning a decent sized pool is always going to be a bit of a crapshoot though!

    David’s on a flight right now but in short, yes, we’re going to continue our survivor analysis and picks through the end of the year, even though our “official” pick this week lost.


  • http://www.teamrankings.com/ David Hess

    Cory — Yep, we will definitely continue the series. We will get a post up tomorrow, since there is a game on Thursday, though it may be a bit later than usual.

  • http://www.teamrankings.com/ David Hess

    drumzan — You’ve been making some pretty smart recommendations all year. Nice job.

  • http://www.teamrankings.com/ David Hess

    Mijmarti — Yep, we will get a post up Wednesday.

  • Deedog44

    No worries….Its just a game.  But I’,m a sports geek that also loves your approach to picking winners.  I’ve always loved statistics, so you’re blogs and explanations are always cool to read. 

    I’m still in a Pick-a-winner league so I’ll still be dropping by to test your picks there.  Lately, it appears the variable of chance is looming large!  I’m currently 6 points out of the lead with a field of 176.

    Continued success to your website.  I enjoy it greatly! 

    p.s.  Ever try an approach toward Fantasy football?

  • Killerzoe

    Drumzan..who do you like this week?

  • http://www.teamrankings.com/ David Hess

    Deedogg — We’ve thought about fantasy, but haven’t done much analysis there yet. The problem is that most of our current models are designed to measure teams, and fantasy is all about players. Fantasy also tends to kind of be divorced from reality in terms of what does and doesn’t matter, so it’s less interesting to us from that standpoint. But, it’s definitely something we’d like to eventually expand into.

  • http://twitter.com/drumzan drumzan

    I like the obvious pick Philly.  Beanie Wells is banged up and Skelton is starting.  I rarely if ever take the overwhelming choice, but sometimes you just have to suck it up to survive another week and hope for a few smaller upsets.  I’m still considering Miami though.  I loved their offensive play calling last week and wondered where that was all season…I’m suspecting injuries could’ve been a big factor in that.  But now they’re essentially injury free.  Miami is defending the run really well this year and Washington’s run game seems to be the only strength of the team.  Washington is banged up and doesn’t play well on the road.  BUT, Miami’s record is 1-12 in the past 13 at home.  That’s significant to me…and might be the only reason I take Philly instead.  My one recommendation is stay away from Baltimore…never ever bet against Seattle at home.

  • http://twitter.com/drumzan drumzan

    Thanks!  Your articles have very valuable recommendations for me as well.   It’s funny, in pick’em pools, I’m no better than 60%…but when it comes to picking 1-2 potential upsets each week, I tend to be a lot better at that.