Week 5 Football Recap (10/8/12)
Monday, October 8 at 11:00 PM ET
For the weekly football review blog throughout the season, I am stealing a page from our weekly football podcasts (while also preparing for said podcasts) and will identify three things that stood out to me in college football and the NFL from the weekend as well as a play or two from the Live ScoreCaster in game projected results that had the most notable bearing on a pivotal game.
NFL Week 5 Thoughts:
Dog Days… The public LOVES favorites. Last week, according to SportsInsights.com, NOT one underdog was wagered upon more often than the favorite in its game. The top four most selected teams against-the-spread were Green Bay (-7), Baltimore (-5), Chicago (-5.5) and Cincinnati (-3). Those picks went 1-3 ATS with only Chicago covering. It’s not just the public that tends to love favorites, many pros (or “sharps”) are on board (or at least they were last week).
Of the 745 participants in the LVH Supercontest (admittedly not all sharps), who each need to select the five teams they feel are most likely to cover the spread, Cincinnati, Pittsburgh (-3.5), New Orleans (-3.5) and Baltimore (-5) – all favorites – were the most selected teams. Those teams also went 1-3 ATS.
This market phenomenon that inflates lines and provides value in underdogs is more the rule than the exception in today’s game (and it is just as true this season in college football). By nature, people like to bet on favorites because we like to bet on teams that are also most likely to win. Winning a blowout where the straight-up and against-the-spread outcome is rarely in doubt throughout the game is also more gratifying than sweating out a game that waffles back and forth around the line. Also, for most, the belief is that it is easier to identify the good and the bad teams. Wagering on perceived good teams and against perceived bad teams sounds like a viable strategy and a much easier one to employ than trying to determine which teams are a little better or worse than perception.
The problem with that is that the wrong teams are often evaluated as “good” and “bad” and the only true blowout, (as close to) slam dunk (as we will see… in other words, 80% or so to win and 60% or so to cover) favorites to win and cover appear in games when above average, good teams take on well below average, terrible teams. For instance, most thought that, going into the season, Cleveland, St. Louis, Arizona, Miami and Minnesota were very bad teams, but their strong defenses remove them from qualifying as terrible. Tennessee and Jacksonville (and possibly Buffalo and Oakland at this point) – teams with bad offenses and defenses - are the closest teams to terrible that we see in the league right now. Any time one of these teams plays an above average team, assuming no significant injuries, a win by a touchdown or more can be expected. Likewise, when Houston, San Francisco, Atlanta and New England play teams that are below average, each should win by a touchdown or more.
In just about every other circumstance, based on what we have seen, the value is more likely to be in the underdog than the favorite. This week, for example, our strongest opinions were a combination of the expected blowouts (picking against Tennessee and against Jacksonville when each played an above average team and for Houston when it played a below average team) and undervalued underdogs in in-between games (liking Philadelphia, San Diego and Kansas City as more than a field goal dogs).
Ultimately, the most challenging (and fun) part of markets is that they are fluid. And the most valuable asset that we have is that we can account for the actual matchups across the board to uncover specific situations that may belie (or support) typical market expectations in an effort to exploit the market. Nonetheless, simply understanding psychology and human competitive nature can explain some inherent market advantages that those who can control their own biases can exploit.
RGIII, ouch… Robert Griffin III has been hit a league-high 26 times (tied with Jay Cutler) in the passing game and has run the ball 42 times (to my knowledge without sliding – I did research on this and could not find any video or record of RGIII sliding this season). In college, Griffin ran or was sacked (both count as runs in college football) 528 times in just over three seasons (he missed almost all of 2009 with a torn ACL).
I love watching Griffin play. I love that, at 6’2”, 220 lbs, he is bigger and stronger than most run-heavy quarterbacks. I love that RGIII is tough and has the mentality to get every yard he can get. I love that the numbers I review suggest that he is going to be great… if he can stay on the field. No body (not even Cam Newton’s or Tim Tebow’s) is meant or built to sustain roughly 13 of the kind of hits that RGIII takes a game.
For a player who has unequaled athletic ability for his position and almost always makes the right decision with the ball, having a reconstructed knee within the last five years and something resembling a concussion (regardless of the degree of the head injury, it already cost them significantly in a chance to win over Atlanta) in the first five games of his NFL career, Griffin’s propensity to take hits is already starting to look like a potentially major deterrent to his value.
Fourth and a foot… This is not rocket science or even something on the analytical level to what we do with our simulations. On fourth down and a foot, all that matters is picking up a foot or more. Both Cam Newtown with the Carolina Panthers and Peyton Manning with the Denver Broncos took unnecessary chances by throwing the ball (ultimately) unsuccessfully on fourth down and less than a yard. Philadelphia faced fourth down and about a foot twice and, both times, unnecessarily (though ultimately successfully) started the play four yards behind the line of scrimmage with handoffs to LeSean McCoy.
I may be as big of a proponent as there is for passing at high frequency that even extends beyond the extremes we currently see. Assuming an above average offensive line and quarterback, the expected positive result of a pass almost doubles the expected positive value of a run (both have similar expectations for negative yardage, while the gap in the expected turnover rate is ever shrinking). I am also a big proponent of aggressive play-calling that includes going for it on fourth down more often than we have typically seen in football.
BUT, there is maximum expected value and then there is game theory. Finding the result that maximizes the expected yardage gained on a play is not always the play most likely to increase a team’s chances of winning. On fourth down, all that matters is extending the drive. An unsuccessful conversion attempt gives the opponent the ball. As a play-caller my only concern on fourth down would be to call the play that gives my team the most likely chance of gaining the needed yards. On fourth-and-20, I’m doing everything I can to throw at least a 20 yard pass. On fourth-and-a-foot, I’m spreading the field horizontally as much as I possibly can, putting my quarterback under center and working on a silent count to get my quarterback the ball for a QB sneak at a time that only the quarterback and center know. I would call this play every single time. If, and only if, a team leaves the options on the outside completely wide open, would I even consider a pass.
College Football Week 6 Thoughts:
Who is #2?… Seriously. Who is the second best team in FBS? For a sport whose popularity is largely built around major upsets, college football has been burdened by parity amongst many very good teams vying for the chance to take on one elite team. With Alabama, which has been #1 in our Power Rankings since August of 2011, such a clear top team, a narrative for much of the season thus far has been deciphering which team is next on that list (though there are no guarantees that the second best team will play for the national championship – especially because a one-loss team right now may be the second best team in the country).
Someone has to be in that spot in our updated (10/7) College Football Power Rankings. South Carolina is there now. And, while the Gamecocks are undefeated, have a great deal of talent, have played a relatively tough schedule and just beat our previous #2 team – Georgia – by 28 points there are ten other teams for which I could make the argument for and against being considered the second best team in the country:
South Carolina: Good – Ranked third in the country in points/yard margin (+0.5) while playing a much tougher schedule than the top two teams. Bad – Three sophomore offensive linemen have led to bottom 20 ranking in FBS (out of 124 teams) in sacks allowed per pass play.
Oregon: Good – Winning games over FBS teams by average margin of 32.3 points. Bad – Well below average penalties-per-play.
West Virginia: Good – Our top ranked offense. Bad – Our 49th ranked defense.
LSU: Good – Top two team all of 2011 and still a defense that ranks second only to Alabama. Bad – Scored just six points in a loss at Florida.
Oklahoma: Good – Experienced star quarterback and offensive weapons. Bad – Only one strong performance in 2012 (@ Texas Tech in Week 6).
USC: Good – See Oklahoma but even more experienced and more elite. Bad – Limited depth has led to catastrophic issues when injuries have arisen.
Notre Dame: Good – Allowing just 7.8 points-per-game against 19th ranked FBS schedule. Bad - FBS average yards-per-play (5.7 yards-per-play as compared to FBS average of 5.6).
Ohio State: Good – Undefeated obvious class of the Big Ten. Bad – Motivation?
Kansas State: Good – Second best points-per-play margin in FBS (and beat Oklahoma at Oklahoma). Bad – Despite several standout transfers and a strong quarterback, there is a major talent gap between Kansas State and everyone else on this list.
Florida State: Good – Averaging 4.9 yards-per-pass more than opponents. Bad – Apparently, the Seminoles are not great at fourth quarter, fourth down conversion defense on the road.
SEC WTF?… On more than one occasion over the last 14 months, I have built the case for the relative success and/or lack thereof picking games straight-up and against-the-spread in college football around the level of consistency on a roster. Essentially, my point is that there is/should be a direct inverse correlation between the accuracy of the Predictalator and the gap between the best and the worst significant player on each roster (the smaller the gap the higher the accuracy and vice versa). We have seen that tendency strengthen with elite talent when the teams we are considering take on the trustworthiness of the NFL. Furthermore, as our home field advantage analysis suggests, SEC teams all have similar, mostly minimal, home field advantages.
Last season, we hit 65% against-the-spread picking every SEC game, which was noticeably better than our performance in most other conferences (for what was admittedly a down year for the Predictalator). This season, we are hitting 58% against-the-spread picking every non-SEC game in FBS, yet are only hitting 44% ATS in SEC games. It’s a relatively low sample size and I don’t presume that any of the three numbers I just cited accurately reflects our true performance (which we would expect to match our confidence), but I think it does tell us something about this year in the SEC.
The SEC this year is a microcosm of all of major college football as outlined above. With one elite team and several other really good teams, there are so many true “coin-flip” games among the teams behind Alabama. Games could go either way. Florida @ Texas A&M, LSU @ Florida, Georgia @ South Carolina, Florida @ Tennessee, Arkansas @ Auburn, Missouri @ Vanderbilt and other notable SEC matchups looked like they could have gone either way entering the game. And while many of those games may have ended up lopsided, few would have been any more or less surprised had the exact opposite outcome occurred in each of those games.
The WAC! Not so whack... As we head into WAC in-conference play, we need to give credit to a conference that I contended before the season was the worst in FBS history. In non-conference games, the WAC, which only includes seven teams – UT-San Antonio, Idaho, Louisiana Tech, San Jose State, Utah State, Texas State and New Mexico State – went 20-14 straight-up and 15-9 against-the-spread (in FBS games).
Louisiana Tech and UTSA are both 5-0 and Utah State and San Jose State both have four wins. The conference has wins over BCS-AQ schools: Illinois, Virginia and Utah, has lost to Stanford and Wisconsin by a combined five points and has defeated Houston twice.
The most impressive thing about the conference to me has been the fact that they have performed well, despite still lacking star power. With the Sun-Belt, MAC, Mountain West and C-USA also sporting some teams that are far exceeding expectations, I’m not sure that I can make the argument that the WAC is not the worst conference in FBS. However, there are no players at critical skill positions like Kolton Browning (Louisiana-Monroe), Benny Cunningham (MTSU), Dri Archer (Kent State), Tyler Tettleton (Ohio), Zac Dysert (Miami, OH), Robbie Rouse and Derek Carr (Fresno State), Trey Watts (Tulsa) or Rakeem Cato (Marshall) on WAC rosters who can be the best player on the field against a BCS-AQ team.
Instead, the WAC has some of the best rising coaches in college football. Sonny Dykes leads a Louisiana Tech offense that is one of the top five offenses in the country. Mike MacIntyre leads a tremendous and balanced San Jose State defense. Gary Andersen has built a football factory at Utah State that has taken lesser talent and started turning it into future NFL athletes. Larry Coker may not be young or “up-and-coming,” but literally started and built a UTSA program that is among 16 undefeated FBS teams in the nation. Each of those coaches could soon be leading major programs.
Live ScoreCaster Play of the Week:
The Live ScoreCaster App has several notifications that can be turned on for games that will keep users up-to-date on things like quarterly scores and projections, games ending, projected lead changes and major swing plays. I love utilizing all of these, especially the latter two as it’s interesting to see which plays lead to teams being favored and ultimately winning the game (as it happens).
For the Major Swing notification, when the projected winning percentage in a game shifts by 35% or more over the course of up to two plays, the notification is activated and the user learns the play, the current score and the new projection. Most games don’t actually have these kinds of plays. And those that do often come in the form of late, game-winning field goals that are less than 65% likely to convert. But there will still be a few other plays each week that stand out for playing pivotal roles in big games.
We recently added a publicly viewable page that will show the charts from all Live ScoreCaster games for a week after each game. These are available for free to everyone. In other words, you should check out the win probability charts from Denver @ New England, Miami @ Cincinnati, Philadelphia @ Pittsburgh, Cleveland @ New York Giants (it’s interesting in the first quarter at least) and Seattle @ Carolina on our new Live History page. For this week’s highlighted play, though, I am going next level by comparing our actual projection to what we would have said if a specific (debatable) penalty would not have been called.
With 2:48 left in the third quarter, San Diego was leading Drew Brees and New Orleans by a score of 24-14. The Saints had the ball on their own 15 and were facing a 2-and-18 after a sack. At the time, San Diego was an 86% favorite to win and by an average score of 32-22. Brees dropped back to throw lofted the ball out into the middle of the field quickly as he felt pressure. Chargers linebacker Demorrio Williams caught the pass on the Saints 25 yard-line and returned the interception for a touchdown. Unfortunately, the pressure that Brees felt was literal – and illegal. Rookie Melvin Ingram was flagged for a roughing the passer penalty on the call.
Had the penalty not been called on a play that really could have gone either way, the Chargers would have been leading 31-14 and would have been 95.9% favorites to win and by an average score of 36-21. Instead, New Orleans’ drive was extended and the Saints scored a touchdown five plays later. At 24-21, New Orleans was a 35% favorite to win the game.
See the chart that tells the story of this game: New Orleans 31 – San Diego 24.