NFL Week 8 Survivor Strategy: Injuries Wreak Havoc On The Possibilites

Welcome to the eighth installment of our weekly series of posts exploring data-driven strategies for getting an edge in NFL Survivor contests. This post includes analysis for NFL Week 8, based on NFL predictions from our algorithmic Team Rankings models and other relevant data.

This analysis 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, while also analyzing future schedule considerations when you are torn between two or three picks.

Everyone can play around with a limited version of the tool to see what it’s all about, but to unlock the full functionality you’ll need at least a Pro Package subscription to Team Rankings.

Week 7 Recap

Week 7 projected to be a relatively easy week as far as survivor pools go, with a couple really big spreads and high win odds. Of course, when things seem too easy, that’s when the football gods typically lay the smack down.

Our official pick for Week 7, the Baltimore Ravens, managed to survive (barely) against a Buffalo Bills team that came out of nowhere to force the Ravens into overtime. The 34 points the Bills put up was the most scored on the Ravens defense all season.

But that wasn’t the most ridiculous outcome of the week. Those of you who picked the New Orleans Saints or Denver Broncos have your own stories to tell. On that note, we now observe a moment of silence for Hank Moody, a frequent commenter, who joined the Who Dat nation just in time for Drew Brees’ four interceptions.

How rare was Cleveland’s dismantling of New Orleans? In our historical database there are 270 instances of a 13.5 or higher spread. In those 270 games, the underdog won only 35 times (13%). You can’t ask for much better than 87% win odds with an NFL Survivor pick.

As for the Broncos, their offense has been flying high all season, but it was the Raiders that piled on an unbelievable 59 points. How rare was that blowout? Winning outright as 7 point dogs is not outrageous; such a thing has happened 449 times before as far back as our records go.

But winning by more than 20 points? That has happened 23 times. By more than 30 points? 5 times. By more than 40 points? That was the first time that has happened in recent history.

All told, Week 7 slimmed the numbers down considerably in most pools. New Orleans (29.5%!), Denver (9.7%) and San Francisco (2.1%) combined with the other losers to knock out ~45% of remaining players. That means that if your pool started with 100 people, there are now probably only around 12 left.

If you’re still alive, you can almost taste it at this point…

NFL Week 8 Survivor Strategy

Teams we already picked: New York Giants, Green Bay Packers, New England Patriots, New Orleans Saints, Cincinnati Bengals/Indianapolis Colts, Chicago Bears, Baltimore Ravens

Teams with Week 8 byes: Atlanta, Baltimore, Chicago, Cleveland, NY Giants, Philadelphia

Week 8 introduces a wrinkle that we’ve largely been able to avoid thus far: injuries. As we’ll see, a few of our top choices have some significant injury concerns and we’ll need to decide how that impacts our picking strategy. First let’s review the short list based on our model win odds projections:

1) Buffalo at Kansas City
2) Jacksonville at Dallas
3) Minnesota at New England

Buffalo at Kansas City (Team Rankings Pick: Kansas City)

We joked last week that Buffalo needed more than rest over the bye week to get things on track. Given the way they played against a very good Baltimore team, perhaps we underestimated the effects of a little R&R? Regardless, the Bills are still 0-6 and on the dog end of the biggest spread of the week at 7.5. We have to think their performance against Baltimore dropped that spread a bit lower than it might have been.

The Bills travel to Kansas City where the Chiefs are 3-0 (4-2 overall). Our models make Kansas City the consensus favorite of the week with win odds between 74% and 85%. We’ll cover this a bit later, but the Chiefs have an easy stretch of games coming up. They don’t play another top 10 team, according to our power rankings, until Week 16 against Tennessee. So is this the week when we should use the Chiefs?

Jacksonville at Dallas (Team Rankings Pick: Dallas)

Several of our frequent commenters mentioned the Cowboys as a team they were eyeing for a Week 8 pick. The hope was that after a big home win against their in-division rivals, the Giants, they’d beat up on a Jacksonville team and get things back on track. Alas, not only did they lose to the Giants, but they also lost starting QB Tony Romo for 6-8 weeks with a collarbone injury.

We mention this often, but our models don’t explicitly attempt to adjust for recent injuries beyond any adjustments that are already “baked into” the current point spread. That’s usually a decent indicator, but still, recent injuries to key personnel do warrant some caution regarding our model projections.

In this game, our models are giving Dallas a consistent 69%-72% win odds. Ultimately, it comes down to how close to Tony Romo’s production you think backup Jon Kitna will be able to get.

Vegas lines seem to say that Kitna will be able to run the Cowboys offense pretty well. A power ratings based prediction with Romo healthy would be Cowboys -8 or -9, and the line for this game with Romo out is now Cowboys -6.5, the second highest of the week. So Romo is probably worth 2-3 points more per game than Kitna, according to the lines. Do you agree?

Dallas’ opponent, Jacksonville, complicates things even further. They’ve been an incredibly sporadic team so far. They have a quality victory against Indianapolis, but all four of their losses were blowouts. This probably isn’t a very good team (their current #27 ranking in our overall power ratings feels about right), but they are still a team that can win a football game on “any given Sunday.”

Minnesota at New England (Team Rankings Pick: New England)

NOTE: New England is not an option for our official pick because we picked them in Week 3. However, we will consider them here since we know some of you haven’t been following our picks every single week.

This is another game where a potential QB injury changes the dynamic. While Brett Favre certainly hasn’t been lighting things up this season, there is no doubt the Vikings are a different team with him at the helm compared to Tavaris Jackson. As of Tuesday night, it’s unclear whether Favre’s consecutive game streak will end or, if he plays, how effective he will be.

New England boasts a 5-1 record that warrants an asterisk. Mainly, that they’ve been on the receiving end of some good fortune, particularly the last two weeks when they’ve beat Baltimore and San Diego by 3 points each, the former in overtime.

Taking all of this into account, Vegas opened the line at New England by 4.5 and has since moved to 5.5. Our models give New England win odds between 66% and 76%, but those figures may be at least slightly understated if Favre is out.

Other games that we are not discussing in detail:

Green Bay at NY Jets (Team Rankings Pick: NY Jets)
Given the hype with which both of these teams came into the season, we were surprised to see this matchup show up so high on our win odds projections. The fact is that Green Bay just hasn’t been playing well and as a result the Jets are being given 66%-72% win odds, along with a Vegas spread of -6. The game is in New York, but unless you are really stuck, we feel like the Jets are a little too dangerous for a survivor pool pick this week.

Houston at Indianapolis (Team Rankings Pick: Indianapolis)
This is a revenge game for Indianapolis, who lost to Houston the first week of the season. Our models are giving the Colts 60%-67% to win and Vegas has the line at -5.5, but this feels like another game where the numbers don’t always project how close these teams play one another.

Carolina at St. Louis (Team Rankings Pick: St. Louis)
Carolina finally squeaked out a victory last week against a very disappointing San Francisco team. As for St. Louis, it seems strange to say, but at 3-4 they are probably exceeding expectations. It’s also interesting to note that with the exception of a blowout loss to Detroit, St. Louis’ other 3 losses have been by a combined 7 points. St. Louis has been pretty good at home as well, winning 3 in a row after an opening week loss to Arizona. With all that said, St. Louis is only favored by 3 at home, meaning that Vegas thinks this will be a relatively close game, and it’s not exactly a safe bet to pin your survivor pool hopes on the shoulders of the Rams.

Future Schedule Considerations

Our Survivor Predictor tool provides week-by-week win odds for every NFL team, allowing you to sort by week and see how the schedule breaks down for your top picks.

What is the Survivor Predictor telling us this week, if we just look at win odds?

  • Not surprisingly, the game against Buffalo represents Kansas City’s highest odds for the rest of the season. With that said of their remaining opponents only Tennessee (#5), Seattle (#14) and Oakland (#19) crack the top 20 of our power ratings.
  • On the flip side, Dallas has a tough upcoming schedule with the exception of a Week 11 matchup against Detroit (at home) and a Week 16 game at Arizona
  • New England has 3 away games of interest left: Cleveland (Week 9), Detroit (Week 12), Buffalo (Week 16)

Future Pick Considerations

Last week we started projecting our picks for the rest of the season. Its a useful exercise to make sure that we don’t use a team too early that we absolutely need later on, as long as we have another good alternative. This week we made a few updates using the current projected win odds from the Survivor Predictor:

  • Week 9: Pittsburgh (at CIN) or Atlanta (vs. TAM) or New Orleans (at CAR)
  • Week 10: NY Jets (away vs. Cleveland)
  • Week 11: Tennessee (home vs. Washington) or Cincinnati (home vs. Buffalo)
  • Week 12: Pittsburgh (away vs. BUF) or Denver (home vs. STL)
  • Week 13: Minnesota (home vs. BUF) or Tennessee (home vs. JAC) or Seattle (home vs. CAR)
  • Week 14: Atlanta (away vs. Carolina) or Washington (home vs. Tampa Bay)
  • Week 15: Miami (home vs. Buffalo)
  • Week 16 & 17: Too early to tell, based on potential resting starters

As of right now, we’re not changing too much, but a couple points of interest:

  • We still think Weeks 9 and 12 are shaping up to be tough decisions coming down to when to use Pittsburgh
  • The injury to Brett Favre puts Week 13 back in play a bit; luckily Tenneesee and even Seattle look like they might be options

What The Public Thinks

(Thursday AM Note: The Yahoo! pick distributions have changed substantially as of Thursday morning. Please see our commentary in the conclusion about what we think this means.)

As of Tuesday evening, it appears that the ESPN and Yahoo! users are once again diverging, with Yahoo! players going heavily for the public favorite Kansas City and ESPN players evenly split between Dallas and Kansas City:

1. Kansas City Chiefs (home) over Buffalo Bills (23.3% of ESPN users | 51.9% of Yahoo! users)
2. Dallas Cowboys (home) over Jacksonville Jaguars (23.1% of ESPN users | 19.5% of Yahoo! users )
3. Washington Redskins (away) over Detroit Lions (13.9% of ESPN users | 5.8% of Yahoo! users)
4. Carolina Panthers (away) over St. Louis Rams (11.7% of ESPN users | 0.4% of Yahoo! users*)

* 7.2% of Yahoo! users are taking Carolina’s opponent St. Louis.

For whatever reason Yahoo! users seem more concerned with Tony Romo’s injury then ESPN players. Beyond that, positions 3 and 4 are pretty intriguing:

  • Washington is a team that we didn’t even consider on our short list above. That is because our models are actually giving the edge to Detroit at home with 52%-60% win odds. Even if you don’t think that will happen, we don’t see how a Washington pick is justified this week.
  • As for Carolina and St. Louis, we don’t understand why so many ESPN users are picking Carolina. It’s rare for us to see such a high percentage of the public picking a team that our models completely disagree with and that Vegas isn’t favoring. And rarer still for our other public set of data, from Yahoo!, to completely disagree.

Our Week 8 Survivor Pool Pick: Kansas City Chiefs over Buffalo Bills

For the second week in a row we are picking against the Buffalo Bills. We hope Kansas City can take care of business and make us sweat a bit less this week. In making this pick, we are assuming a fairly balanced pick distribution, like ESPN’s current numbers.

Of all the risks we could take this week, betting that the Bills’ performance against the Ravens last week was more of an anomaly than a meaningful indicator seems the safest bet of all. The Bills do look like an improving team, and you could also say they’re due to win a game. But we think the odds are stacked against them.

As for Dallas, the judgement isn’t that Jon Kitna is so horrible that we don’t think he can run the Dallas offense. In fact, as far as backups go, at least he has some experience under his belt. The issue is more that we know the models we rely on just aren’t going to be able to properly gauge the impact of the injury and we don’t see the risk as being worthwhile.

Bottomline: if the pick distribution in your pool ends up being more balanced like ESPN, then there’s little value in making a contrarian pick this week. The Chiefs have significantly higher win odds than any other team.

However, if 70%+ of your pool is picking either Kansas City or Dallas, as the Yahoo! numbers indicate, then you’ve got a chance to take some additional risk and increase your overall odds to win your pool … if you can stomach it. We’re guessing most people have already taken the Patriots – but if they are still available to you, they might be a nice play against Minnesota and either an injured Brett Favre or Tavaris Jackson. Beyond that, the Jets, Colts, and Rams all have win odds that are close enough to Kansas City (65% vs. 75%) and nearly equal to Dallas.

We have differing personal opinions over here at TR as to which of those teams would be best as a contrarian play, but all of them merit some level of consideration if 70%+ of your pool is picking Dallas or KC. Fire away in the comments and let us know what you think…

Thursday AM Update: In checking the Yahoo! numbers this morning, things have changed significantly:

Kansas City – 58.1% of the public
Dallas – 8.4%
St. Louis – 7.8%
NY Jets – 5.5%
New England – 5.1%

If the distribution in your pool looks more like this then there is a strong incentive to pick a team other than Kansas City. The next 4 options all have roughly the same win odds and no value advantage based on the public picking, so it ultimately comes down to who you are most comfortable with based on your analysis.