NFL Week 4 Predictions, Rankings & Tips: The Weekly Rundown
September 25, 2014 - by Tom Federico
Here’s a rapid-fire list of weekly highlights from the vast array of NFL content we publish on TeamRankings. We’ll also share some predictions and picks for NFL Week 4 and beyond.
Just announced: The prize pool for our Week 4 NFL contest is up to $2,249. Claim your spot now.
Seattle and Denver still tops. Unsurprisingly, after playing to a tie in regulation, Seattle and Denver still top our NFL rankings entering Week 4. Partly because the Seahawks ended up winning by six in OT, which was slightly above their expectation for this game, their rating crept up a little bit to 7.5, while Denver’s rating stayed flat at 6.9. So the system currently sees #1 Seattle as about half a point better than Denver on a neutral field.
Cincinnati makes a jump. The biggest ratings story of the week is probably Cincinnati, who leapfrogged the Patriots and 49ers en route to a current #3 ranking. The Bengals’ margin of victory performance this year has been impressive (for reference, Cincy has covered the Vegas spread by at least six points in all three of their games), and the blowout of the Titans raised Cincinnati’s rating by 1.3 points. The Bengals still have a ways to go to catch Denver, though, and none of the their three wins has come against a team currently rated in our top 10. However, #5 New England looms in Week 5, after a bye this week.
San Francisco’s loss wasn’t a crushing blow. #4 San Francisco’s 9-point loss on the road to Arizona wasn’t a complete disaster, given the Cardinals’ above average rating. But the Niners are now clearly out of the top three teams in the league. It could have been worse, but the fact that San Diego again boosted its rating in Week 3, coupled with the fact that Arizona beat San Diego in Week 1, helped cushion the impact of this loss a bit for the 49ers.
New England’s win may have been worse. #5 New England was able to pull out a win, but significantly underperformed at home against terribly-rated Oakland — so much so that the Patriots’ rating actually decreased more than San Francisco’s did this week. Having yet to play a game against a team that we rate as above average, New England still has some proving to do. Week 5 against the Bengals is their chance, provided they don’t choke at Kansas City this week.
Looks like we jinxed Carolina. Carolina’s 37-19 blowout loss at home against previously #22 ranked Pittsburgh was an absolute crusher, dropping the Panthers’ rating by nearly 2 points, which is huge. Carolina dropped from #6 to #9 in the rankings, while the Steelers shot up from #22 to #17. That result really derailed Carolina’s promising start, so now we’ll see if it was a fluke or not.
Detroit did its job. We expected Detroit’s win over Green Bay, but the margin of victory was enough to help cause about a half-point shift (Detroit up, Green Bay down) in their respective ratings.
Nice win by Kansas City, but other results hurt. #22 Kansas City still is rated well below average after the upset win at Miami, but the Chiefs at least pulled themselves out of the dregs, moving up five spots from #27. However, Tennessee getting blown out by Cincinnati and Buffalo losing by 12 at home to San Diego both conspired against the positive impact of KC’s win, since those results make both KC and Miami look weaker.
Bears get better, but rank worse. In the “why ratings are fun” category, the Bears actually dropped from a #17 to #19 ranking after their 8-point road win over the Jets. The win was a plus for Chicago, improving their rating by almost half a point, buy they were leapfrogged in the rankings by several teams that registered more impressive results, including Atlanta, Pittsburgh, and the NY Giants.
Three tiers of good teams now. After three games, some fairly well separated rankings tiers have formed. First, there’s a clear duopoly at the top with Seattle and Denver. A step down from them are six teams of great to good quality (Cincinnati, San Francisco, New England, New Orleans, San Diego, and Arizona, with team quality decreasing almost linearly from the first to last team listed). Finally, another step down from those teams is a batch of six above-average teams that are all very closely rated right now (Carolina, Indianapolis, Baltimore, Detroit, Atlanta, and Green Bay).
Don’t be misled by win-loss records. At this point, we can start to identify some teams whose early record doesn’t reflect their rating. Teams with higher ratings but losing records include San Francisco (#4 with a 1-2 record), New Orleans (#6, 1-2), Indianapolis (#10, 1-2), and Green Bay (#14, 1-2). Teams with lower ratings but winning records include Philadelphia (#15, 3-0), Buffalo (#24, 2-1), Dallas (#20, 2-1), and Chicago (#19, 2-1).
What our old ratings think. Finally, it’s interesting to see how our original predictive ratings system currently sees things. (For reference, our original system does not factor in a preseason rating for each team like our new system does; it measures teams exclusively on current season results.) Here’s the current top 10 in our old rankings: #1 Cincinnati, #2 San Diego, #3 Seattle, #4 Arizona, #5 Denver, #6 Baltimore, #7 Pittsburgh, #8 San Francisco, #9 Chicago, and #10 Buffalo.
Biggest differences between our new and old ratings. So the biggest differences are Buffalo (#24 new ratings, #10 old ratings), Chicago (#19 new, #9 old), Pittsburgh (#17 new, #7 old) all ranking much higher in our older system, plus San Francisco and New Orleans lose the “benefit of the doubt” bestowed upon them by a good preseason rating. We’ll see how things look after more games are played. It’s worth noting that Buffalo and Chicago are currently tied for third in the NFL in turnover margin, at +1.3 per game, after only being +0.2 and +0.3 last year.
Seattle the new Super Bowl favorites. Seattle’s OT win over Denver, with a small assist from San Francisco’s loss to Arizona, helped the Seahawks edge the Broncos as the most likely Super Bowl champions. Seattle now has roughly 16% odds to win it all.
Teams with the highest NFL champion odds. The seven most likely 2015 Super Bowl champions are currently Seattle (16%, up from 14% last week), Denver (14%, down from 16%), Cincinnati (12%, up from 8%), New England (8%, down from 10%), and a virtual three-way tie between San Diego, San Francisco, and Arizona (all at 5%).
AFC division winner projections. Our current projected division winners in the AFC are New England (66% AFC East, vs. 58% last week, thanks to the rest of the division going 0-3), Cincinnati (63% AFC North, vs 61% last week, with a strong win being partially offset by Pittsburgh and Baltimore wins), Houston (51% AFC South, vs. 63% last week, thanks to a loss plus a strong road win by Indianapolis), and Denver (61% AFC West, vs. 73% last week, falling again after a loss plus a strong San Diego win). It’s interesting to see how New England’s division winner odds improved this week due to losses by the Jets, Dolphins, and Bills, but the Pats’ NFL champion odds decreased after a weak showing against the Raiders.
Other AFC division winner threats. Outside of the current favorites, Indianapolis is now the biggest division winner challenger in the AFC, with 36% odds to win the AFC South now that the Titans have faded over the past two weeks. San Diego is close behind at 35% odds to win the AFC west, up from 25% from last week.
NFC division winner projections. Our current projected division winners in the NFC are Philadelphia (44% NFC East, vs. 43% last week, with our models remaining relatively pessimistic about the Eagles), Detroit (43% NFC North, and back into division favorite status after beating Green Bay), Atlanta (35% NFC South, and taking the top spot thanks to a blowout win plus a blowout loss by Carolina), and Seattle (52% NFC West, vs. 44% last week, after a win plus another San Francisco loss).
Other NFC division winner threats. Outside of the current favorites, the relatively wide open NFC has several other division threats, including Dallas (27% to win the NFC East), Chicago and Green Bay (25% and 22% to win the NFC North, respectively), and now Arizona (30.4% to win the NFC West). As things stand today, San Francisco projects more like a very strong Wild Card contender than as an NFC West division winner.
NFC South is totally up for grabs. The NFC South deserves special mention as being completely up for grabs between Atlanta (35% odds to win the division), Carolina (33%), and New Orleans (31%). New Orleans is currently one game back in the standings compared to the other two teams, but the Saints still have a higher predictive rating than Atlanta or Carolina.
Playoff odds madness in the NFC West. The parity in the NFC West has resulted in a crazy situation regarding playoff odds. As one would assume, Arizona’s defeat of San Francisco made a big impact on postseason implications for those two teams. We still rate the 49ers as the better team, but being two games up in the standings (and having a head-to-head win in hand) is huge for the Cardinals; Arizona’s playoff odds have now increased to 62%, compared to about 44% for the 49ers. And despite San Francisco having the fifth-best chance to win the Super Bowl this year, right now we actually don’t expect the Niners to make the playoffs! But if the Seahawks and/or Cardinals stumble, and San Francisco does make the postseason, the 49ers are a good enough team that they’d be expected to pose a serious threat to win it all. Hence the 5% NFL champion odds. Wow, fun.
Don’t count these teams out based on their record. Teams with good playoff odds but not so great win-loss records right now include Indianapolis (1-2, but a 45% chance to make the playoffs), New Orleans (1-2, 45%), and the aforementioned San Francisco (1-2, 44%, so not technically “expected” but still a solid chance).
Don’t count these teams in based on their record. Teams with good records but not so great playoff odds right now include Buffalo (2-1, but only a 27% chance to make the playoffs), Dallas (2-1, 36%), and Chicago (2-1, 36%).
You can probably count THESE teams out, though. At this point, our season simulations are giving five teams less than a 10% chance of making the playoffs: Jacksonville (now down to a 1 in 1,000 chance to make the postseason), Oakland, Tampa Bay, St. Louis, and ladies and gentlemen, please welcome the New York Jets to our list this week. And congrats to Kansas City for playing their way off the list with the upset win at Miami.
We’re all equals, again, in Week 3. 11-5 was the magic record last week for picking NFL game winners. The public (i.e. the ESPN user consensus picks), Vegas spreads, and our game winner prediction models were all 69% accurate. We liked the NY Giants over Houston and the Lions over Green Bay, which the public got wrong. But public opinion beat us on Chicago over the New York Jets and San Diego over Buffalo.
Some really cool pick’em performance data (still) coming soon. This year, we’ve begun collecting extensive data from users of our Football Pick’em Pools product, asking how they are currently doing in their pick’em pools while using our pick recommendations. We’re going to start publishing this data in periodic blog posts, so stay tuned. Unfortunately we didn’t get a chance to start this up last week.
The overrated favorites strike back. After two weeks of upset city in the NFL, the four most underrated picks from our Week 3 Pick’em Data Grid — defined as teams with the biggest difference between win odds and public picking percentage — all lost: the New York Jets (actually a Vegas favorite over Chicago), Tampa Bay, St. Louis, and Tennessee. The Jets were a great contrarian play, having around 55% win odds yet only 22% picked by the public, but untimely turnovers derailed New York. The next tier of underrated teams did much better, though. The Giants (vs. Houston) were #5 on the list, Pittsburgh (vs. Carolina) was #6, and Arizona (vs. San Francisco) was #10.
Week 4 strategy for season prize pools. For Week 4, your best strategy in NFL pick’em pools that only have season prizes is to remain conservative, and not go too crazy picking upsets. One game to watch out for is Green Bay vs. Chicago. It projects to be a close contest, but Green Bay is currently the Vegas favorite yet is only 44% picked. So, given the current data, it’s smarter to stick with the Packers.
Week 4 strategy for weekly prize pools. Pools with weekly prizes call for a different strategy, one that needs to give much more weight to how your opponents are likely to pick this week. Green Bay over Chicago currently looks like a no-brainer pick in these pools. In addition, in bigger weekly prize pools it almost always pays to pick some high-value upsets as well. On that front, two underdogs with decent shots to win are being severely underrated right now: Minnesota (42% win odds, yet only 8% picked at home against Atlanta) and the Jets again (close to a 50/50 winner, yet only 16% picked at home against Detroit). It’s pretty clear that recency bias is creating solid value in those two games. Both the Vikings and Jets are on losing skids (the Jets having just played a poor game on Monday Night Football), and Atlanta and Detroit are coming off big wins.
Week 4 strategy for YOUR pool. Just remember, the optimal strategy for your specific pool, including how many favorites vs. upsets you should pick this week, depends on a lot of factors: the number of entries in your pool, the rules, the scoring system, the payout structure. That’s why we developed our Football Pick’em Pools product to apply our analytics to your specific pool.
The first ho-hum week. After two weeks of big upsets and elimination chaos, Week 3 was a snoozer. The most popular team to lose, Miami, was only picked by 1% of survivor players nationwide, and about 98% of entries still alive going into Week 3 made it on to Week 4.
Another close one, though. The most popular pick, New England, a team that we advised against picking if you only had one or two entries in a survivor pool, barely scraped by the Raiders at home after having what looked like a game-tying Oakland touchdown late in the 4th quarter overturned by a penalty. A Patriots loss would have eliminated about 50% of surviving entries.
Our against-the-grain Week 3 survivor picks both survived. New England’s popularity and future value were huge negatives last week, so the profit-maximizing decision was to save the Patriots for later and take one of two slightly riskier teams: Philadelphia over Washington (our preferred pick early in the week) or Indianapolis over Jacksonville (our preferred pick toward the end of the week, after movement in the betting lines against Philly). The Eagles gave us a scare but ultimately triumphed, while Indianapolis crushed the Jags on the road. So even though hardly anyone was eliminated last week, we saved the Patriots while half our opponents used them, which is big.
Don’t pick this team in Week 4! In Week 4, as is often the case, there is at least one popular pick you need to avoid if your goal is to maximize your chance to win your survivor pool. That team would be San Diego. At roughly 60% picked, the Chargers are even more popular than the Patriots were last week, are no safer than the Pats were against Oakland, and have plenty of future value. If you’ve only got one or two entries in a survivor pool, you will definitely increase your expected value by saving San Diego for later. In fact, at least 10 other teams currently rate higher than San Diego on our survivor pick report card this week. Go Jaguars!
NFL Week 4 Betting Picks & Game Predictions Update
Over/Under picks register a second great week. After a stellar Week 1 and a mediocre Week 2, our playable NFL totals picks went 5-2 against Sunday lines, with 3-star picks going 2-0 (Oakland/New England Under 47 was the top play, followed by San Diego/Buffalo Under 45). Playable totals picks are now 14-7 through Week 3.
Spread picks were losers. After two hot weeks of NFL spread picks, our “regression is coming” warning proved prescient last week, as we went 6-10 ATS overall and 1-3 on 3-star rated spread picks. We caught a tough break with our top rated play, Minnesota +10.5 at New Orleans (the Saints won by 11), but Tennessee +7 and Tampa Bay +6.5 both got smoked. All playable spread picks went 4-4, though, thanks to a 3-1 record on 2-star picks, so it wasn’t a total disaster. Playable spread picks against Sunday lines are now 16-10 on the year.
Expect weekly betting picks results to be volatile. In case you’re wondering, a 6-10 weekly performance for our NFL spread picks is absolutely going to happen over the course of an NFL season, and will almost certainly happen again this year (or worse). With a sample size of 13-16 games a week, and many picks being close to 50/50 propositions, the variance of week to week NFL spread pick results (and over/under results as well) will be high. For example, 2012 was a profitable year overall for our playable NFL spread picks. It also featured a 4-10 week, a 5-11 week, and two 6-10 weeks. Those weeks were offset by a 10-2-1 week, and an 11-4 week, two 9-5 weeks, etc.
Game winner picks still behind the public. As mentioned in the pick’em section above, both our models and the public went 11-5 this week picking game winners. So we’re still lagging the ESPN user consensus this season in terms of game winner accuracy, by three games, on account of the public’s freakishly stupendous performance picking upsets in Week 1.
Fewer Week 3 upsets than expected. Overall, out of the 12 Week 3 games listed on our Most Likely Upsets page, we expected four or five upsets to happen based on our win odds. Only two happened. One of them, however, was our most likely upset in games where the Vegas spread was at least three points: Kansas City over Miami.
Week 4 playable NFL betting picks. As of posting time, we had 7 playable spread picks, 5 playable totals picks, and 2 two-star money line value picks for NFL Week 4, which is a smaller number of playable picks than last week. We made a few refinements to our NFL prediction logic early this week, and those changes are likely to at least slightly reduce the number of playable NFL picks going forward.
Don’t call them the Week 4 locks of the week. The top confidence NFL Week 4 game winner picks from our models are currently San Diego over Jacksonville (83%), Pittsburgh over Tampa Bay (73%), and Indianapolis over Tennessee (72%).
Most likely Week 4 upsets. The most likely upset of the week is currently the New York Jets over Detroit, with a 51% chance of happening (the Jets are underdogs in Vegas, but our models like them slightly better). The most likely upset of a Vegas favorite of more than 3 points is the New York Giants over Washington, with a 38% chance of happening.
Sellout! Our one-day NFL Week 3 fantasy football tournament on FanDuel sold out, attracting the maximum of 100 entries by Saturday. Thanks to everyone that played! As a result of the demand, we are doubling the size of the Week 4 contests to 200 entries and a guaranteed $2,249 prize pool!
captainjack777 edges teamrankings in a thriller. We thought we had our first ever FanDuel contest victory in hand in Week 3, but huge props go to tournament winner captainjack777, who brilliantly drafted both Andrew Luck and Jeremy Maclin. Here’s the full dominating lineup. Entering the Monday night game, teamrankings was leading the 100-entry field, but captainjack got 2 TD’s from Chicago tight end Martellus Bennett to leapfrog us, 146.6 to 145.82. Well played, captainjack. We’ll be contacting you soon so you can claim your free premium Football Season Pass to TeamRankings, in addition to your cash prize from FanDuel.
New FanDuel page and extra bonus prizes announced. As our weekly private tournament grows, we launched a new page on the site (link at the top of this section) that will have a direct link to our weekly private tournaments, as well as some season-to-date stats on top performers. More importantly, we’ve announced a slew of new extra bonus prizes from TeamRankings, including a lifetime subscription to TR and Amazon gift cards!
Come play with us. We want you to get in on the Week 4 action with us. 100 entries will win cash this week, and it’s your chance to beat the TR Nerds. Claim your spot now before they run out.
If you liked this post, please share it. Thank you! TwitterFacebook