East Coast Offense: Two Sports Betting Ideas

Two Sports Betting Ideas

I've picked every NFL game against the spread for the last 19 years, and if I've learned anything, it's that making money after paying the rake is hard. The NFL lines are usually pretty tight as it's so widely wagered on, and good information about the games is freely available to everyone. But I've been thinking about two potential areas those with better statsitical chops than me might be able to exploit.

(1) Team Volatility

One thing I've noticed is the point spread correlates almost 100 percent with the game's moneyline. That is, if the spread is 10, the implied moneyline (what you get when you average the favorite's and underdog's lines, thereby ironing out the rake) will give that team roughly an implied 83 percent chance to win. If the spread is nine, it'll be closer to 79 percent. This is virtually always the case. Put differently, you'll never see the nine-point favorite have a moneyline that gives it an implied 83-percent chance to win, while the 10-point favorite is at 79 percent.

On its face, this makes sense. The team that's favored by the larger margin is of course more likely to win. Usually. But is that always the case, as the correlation would imply?

Let's do a quick thought experiment. It's Week 11. Team A is 5-5, winning every other game, i.e., it has not had a two-game winning or losing streak. In its wins, it wins by 50. In its losses, it loses by 40. It has beaten two 8-1 teams and lost to two 1-8 ones. Team B is 2-8 heading into Week 11. It has beaten two 6-3 teams, but lost to four 3-6 ones by an average of 35 points. Vegas has the spread as Team A -10 over Team B. Should the moneyline be -550/+450, roughly commensurate with the spread, or should it be more like -300/+250, given how insanely volatile these teams have been all year?

It's impossible to say without knowing more, but the point is the projected margin of victory shouldn't be the only basis for evaluating the moneyline. Put differently, it's possible that when Vegas aligns the two it could be making a mistake in cases where teams have higher than normal volatility. The idea then would be to figure out a metric for volatility and backtest it to see that it's predictive. If so, you could bet the favorite with the points and hedge with the underdog on the moneyline. Or you could just bet the dog on the moneyline without hedging. And unlike in the example above, you'd only need one team (on either side) to exhibit abnormal volatility to make the bet.

I'm assuming you could also do the same in reverse. Bet favorites on the moneyline when both teams are exceptionally stable.

(2) Expect the Unexpected

Every once in a while there are major upsets. Out of nowhere, a 14-point underdog destroys the favorite. Sometimes, it's dumb luck - a massive turnover differential, bad weather, special teams TDs, etc. But other times, the "bad" team just flat out crushes the "good" one down in and down out. I want to focus on the latter scenario.

The guys at Massey Peabody have done great work measuring the strength of teams' weekly performances based not on outcome, but on down-by-down success rates. They measure what the final score should have been based on the quality of play, taking out the noise to the extent they can. What I'd be curious to see is where the greatest disparities occur from pre-game projection to post-game performance-based result. Let's say we draw the line at 20 points, i.e, we use the subset of games that had disparties of that size between the predicted and performance-based result. Team A was projected as a 12-point favorite but were minus-14 on the MP metrics for the game. That's a 26-point swing and would qualify for our purposes.

Assuming there were 10 examples of 20-point swings every year, we could look back over the last 10 years or so, and generate a decent-sized sample. What do these 100-odd surprise games have in common? Are they night games, early body-clock games for a west-coast team, short-week games, divisional games, etc. Are there particular conditions that portend major upsets, irrespective of the disparity in previoius play?

Our sample might be too small at 20-point disparities, so maybe we'd have to lower it to 15 to generate more results. Either way, there should be a sweet spot where the sample is large enough and the differential extreme enough to use as our data set.

Of course, it's possible most upsets simply happen at random, and we're unable to discover anything material in common between them that has predictive value. But I think it's worth looking into. We often dismiss as random things we haven't yet developed a process for explaining/predicting. Every effect has a cause, and if we can identify some causes/conditions that tend to obtain generally during upsets, we'll have an edge in predicing them. This would have applications not only in betting the moneylines but also in Survivor and even fantasy football.

Giants and Bucs Quit?

Reports surfaced last week that both the Giants and Bucs had packed it in. What does that mean, exactly? Usually, it means the team's not playing well, and when the previous game got out of hand, players lacked discipline and focus, causing the other team to have an easier time of it than you'd expect against even a bad NFL defense. It's also usually associated with a coach who is not only losing, but struggling with the media and considered bad by most fans. Ben McAdoo and Dirk Koetter certainly qualify. Someone like Kyle Shanahan or even Hue Jackson (for now) probably do not, despite their records.

But lacking discipline and focus for a game that's out of hand isn't the same thing as "quitting." Quitting implies the players voluntarily suspended their focus out of despair or disgust for the situation. To satisfy that standard, you'd have to know what the players were thinking, and that's obviously impossible.

This might seem like a semantic distinction, but it isn't. A frustrated player, working for a bad coach, might lose focus and make a mistake the other team exploits. But a player who's half-assing it on the field as a passive-aggressive protest of sorts is a different story altogether. The former will presumably make peace with the disappointing situation and go back to work for the sake of his teammates, his future earnings and his long-term viability as a player in the league. The latter would probably get injured or benched, and might not have made it to the NFL in the first place, as quitting during adverse circumstances tends to curtail your career early on.

For that reason, when I hear a team has quit, I'll usually ignore it. That doesn't mean the team will play well - the Giants are a dysfunctional mess under their terrible coach. But I'm expecting mental errors and poor play, not an outright in-game protest.

Roger Goodell's Payday

NFL Commissioner Goodell is negotiating his next contract, and leaving aside how he can justify asking for nearly $50 million per year given his legacy of arbitrary, incomprehensible catch rules, a soul-destroying number of penalties and stoppages, botched handling of domestic violence cases, Deflategate, lawsuits from Colin Kaepernick and potentially Jerry Jones, declining ratings and the scourge known as the Thursday night game, one item in particular for which Goodell is asking jumped out at me:

The proposal included an approximate salary of around $49.5 million per year, lifetime use of a private jet, and lifetime health insurance for his family.

If you're making $50 million a year, why would you ever buy health insurance? Health insurance is a terrible deal - as insurers are for-profit entities taking a major rake from the pool. It's like betting NFL games with your bookie, only he was taking a 20 percent rake, and when you won (got sick), you had to fill out tons of forms, read through reams of fine print and spend hours fighting to collect all the money that was owed you.

The only reason to buy it is to avert catastrophic loss if you're unlucky enough to require medical care exceeding your means. We take a terrible deal paying hundreds or even thousands of dollars per month for high-deductible/high out of pocket care that we might never use only because we don't want to risk bankrupting ourselves and our families who would probably have to kick in. Why on earth if you had enough money to bankroll a worst-case scenario health problem without even feeling it would you ever purchase health insurance? No one in his right mind would give into this extortion scheme if he didn't have to.

But here's Goodell, rich beyond belief for doing a mediocre job. This isn't Tom Brady, Aaron Rodgers or even a mediocre owner who inherited an NFL team and presumably at least has some capital at risk. This is a no-skin-in-the-game administrator asking for a preposterous sum for past growth due far more to fantasy football than him, and he wants health insurance, to boot! It's a testament both to how foolish Goodell is and simultaneously how terrible our health insurance system is that even a person in his shoes dreads it so much he made it a point in his contract.

Week 10 Observations

• I wrote on Twitter after the late games that the Rams, Eagles and Saints were the NFL's three best teams. After Sunday night, it's hard not to add the Patriots. Their defense was still somewhat shaky – Denver moved the ball pretty well - but the offense methodically dismantled the Broncos defense in Denver. And Tom Brady took some deep shots, even if they didn't connect.

• Brady's 7.8 YPA, three TDs, no picks and one sack counts as a clinic in that environment. That Mike Gillislee was a healthy scratch bodes well for Brady's fantasy stats, though it might simply have been the opponent (Denver has been very tough against the run this year.)

•Dion Lewis had 14 carries and a rushing score (as well as a return TD), while Rex Burkhead had 10. Using those two as the primary runners with James White as a third-down option gives the Patriots maximum versatility on very play. It's possible Gillislee comes back against a softer front, but the offense is better when the backs are receiving threats.

•Rob Gronkowski and Brandin Cooks tied for the team lead with 74 receiving yards, though Cooks had 11 targets and Gronk seven. Both saw looks in the end zone, though neither converted.

• Apparently Martellus Bennett (recently rumored for season-ending surgery) was just fine, catching all three of his targets for 38 yards. It was a nice trolling of the Packers who released him.

•Brock Osweiler wasn't terrible, though he threw a bad pick in garbage time just as Cris Collinsworth was making that very point. The problem with Osweiler-level QBs is there's no margin for error. They never make an unexpected play or scramble out of trouble. With Cam Newton or Tyrod Taylor there's something to offset a mistake. With Alex Smith there are so few mistakes. Even Josh McCown and Ryan Fitzpatrick types have some play-making ability. Osweiler stands in the pocket and offers nothing but the possibility of throwing an accurate pass. So even when plays reasonably well, as he did for most of Sunday night's game, it doesn't offer much hope.

• The Broncos are a bad offense with a three-way committee at running back. Pass.

•Emmanuel Sanders is plainly healthy and had a big game on 11 targets. Demaryius Thomas had eight of his own and a touchdown. The rest of the team combined for 13. The tree is still narrow.

• Why did Vance Joseph punt on 4th-and-6, near mid-field, down 18, with less than 17 minutes left?

•Eli Manning had 7.4 YPA and two TD passes, but he lost a fumble during a sack wherein he rolled the ball forward on the ground like he was at a bowling alley. It's time to audition Geno Smith, who was set to start for the Jets before getting punched in the face during the preseason in 2015. It's highly likely Smith is bad, but the team has six games to find out, and there's nothing to be gained by sticking with Manning's carcass. Of course, this will never happen, but it's a no-brainer to move on from Manning at this point.

•Orleans Darkwa ran well on 14 carries and also caught two passes for 18 yards. He's nothing special, but he's decisive, hits the hole and runs hard.

•Sterling Shepard had a big game with 11 catches and 142 yards on 13 targets. Evan Engram had on 31 yards on nine targets, but caught another TD. Like the Broncos, the Giants have a narrow tree.

• Incredibly, backup TE, Garrett Celek, filling in for the injured George Kittle, caught a TD against the Giants, running their streak of TDs allowed to opposing TEs to nine.

•Carlos Hyde went 17-for-98, but Matt Breida got the 33-yard TD run. Neither was a factor in the passing game.

•Adrian Clayborn had six sacks of Dak Prescott. It's one thing to have a great offensive line when everyone's healthy, but depth is also important.

•Dak Prescott didn't go much while getting dropped eight times and fumbling twice, but he managed 42 rushing yards and a TD.

•Dez Bryant caught four of eight targets for 39 yards. He still looks good after the catch, but he's not beating anyone down the field.

•Alfred Morris had 11 carries for 53 yards, Rod Smith had three for 14 and four catches for 15 more. Darren McFadden was a non-factor.

• Another modest day for Julio Jones – 57 yards on eight targets. Austin Hooper scored a TD and caught all six of his targets for 49 yards.

•Devonta Freeman left early with a concussion, and Tevin Coleman was passable – 20 for 83 and a score.

•Jared Goff had a quiet first half before exploding in the second. Robert Woods is clearly the team's No. 1 receiver and flashed some big-play ability en route to a 10-8-171-2 line. Sammy Watkins caught a TD, but saw only three targets. Cooper Kupp caught six of seven targets for 47 yards.

•Todd Gurley is fantasy's No. 1 player. Even though he had only 11 carries and didn't score, he managed six catches and 136 yards from scrimmage. When you're a workhorse back not dependent on TDs or even carries in a given game, you're a monster.

•Tom Savage's play can only be described as barbarous. Two picks, two fumbles and three sacks while getting only 6.1 YPA. DeAndre Hopkins caught seven of 14 targets for 111 yards, but he's the only functional member of the offense.

•Lamar Miller ran well but had only 11 carries and one target. The Texans figure to be playing from behind a good deal, limiting the rushing upside. At least Miller doesn't have much competition from D'Onta Foreman who had only 18 yards on seven carries.

• At least the Chargers covered the spread. Even that was in doubt when A.J. Bouye nearly took a Philip Rivers pass back to the house in overtime. Not even the Lions can touch the Chargers' legacy in finding ways to lose games they had essentially salted away.

• The Chargers even partially blocked Josh Lambo's overtime field goal, but it barely scraped through. On the previous snap, the Chargers committed a delay of game penalty that moved the attempt up five yards.

•Austin Ekeler outplayed Melvin Gordon but lost a key fumble late in the fourth quarter while the Chargers will trying to close out the game. Gordon had eight targets to Ekeler's five, but Ekeler caught all five for 77 yards and two scores. And Gordon had 16 carries to Ekeler's 10, but Ekeler had 42 yards to his 27. Expect Ekeler to continue to see run on passing downs at the very least.

• The defenses totally shut down both Rivers and Blake Bortles. And the Chargers held Leonard Fournette to 33 yards on 17 carries. Marqise Lee scored a TD and a two-point conversion, while Allen Hurns had 70 yards on nine targets, but was seen on crutches after the game.

•Case Keenum lit up the Redskins for three quarters before slowing down and throwing two picks late. Still, he had 10.5 YPA and two TDs. I don't see the Vikings turning to the unimpressive Teddy Bridgewater in the near term, and I'm not even sure why it's up for discussion.

•Adam Thielen had 166 yards and scored his second TD in two games. He gets a ton of targets and produces every week. Consider him a top-10 WR. Stefon Diggs had a long TD, but only five targets.

•Latavius Murray saw more carries and had more yards than Jerick McKinnon. Murray also scored a short TD, though McKinnon had two catches for 15 yards. McKinnon is probably the better back, but this is an even split.

•Vernon Davis and Jamison Crowder saw 11 targets and 76 yards each. Davis is a top-10 TE so long as Jordan Reed is out.

•Kirk Cousins played passably, but still can't get his outside receivers involved. Former practice squad player Maurice Harris made an Odell Beckham-esque one-handed 50-yard TD catch, but saw only two targets.

•Samaje Perine led the team with 35 rushing yards after Rob Kelley got hurt. Chris Thompson had three catches for 41 yards, but has largely come back to earth. Kelley could be out for the year, so Perine might have another opportunity yet.

• Despite a terrible game last week, Doug Martin saw 20 more carries, but managed only 51 yards. DeSean Jackson and rookie Chris Godwin each had 10 targets in Mike Evans' absence.

•Robby Anderson keeps making plays for the Jets. He's a top-20 WR at this point.

• For some reason the Jets don't like using Bilal Powell in the passing game. With Matt Forte out, Elijah McGuire got seven targets and Powell only two.

• I get the feeling Andrew Luck has played his last snap as a Colt. With Jacoby Brissett playing well – 9.3 YPA, two TDs, one pick against the Steelers – and owner Jim Irsay questioning Luck's willingness to play, I imagine Luck will be moved for draft picks this offseason. Luck can't be happy in Indy, and teams would line up to offer first-round picks and more if Luck's shoulder checks out this offseason.

•Ben Roethlisberger played passably for a road game, and one in which Le'Veon Bell – 26 for 80 – couldn't get going. Juju Smith-Schuster made the big plays and caught a TD, while Antonio Brown had a modest 7-3-47 line. When you can't trust Antonio Brown, the WR position is really hurting.

• Chester Rodgers led the team with a 6-6-104-1 line. The rest of the team's receivers struggled against a tough Pittsburgh secondary, except for Donte Moncrief who had one target that resulted in a 60-yard TD. That T.Y. Hilton was a game-time decision with a groin injury didn't help.

•Marcus Mariota gets by, but there's more upside to his game than he's shown. The 51 rushing yards were nice, but he's got to pair them with bigger plays in the passing game. That could happen with Corey Davis (10 targets) getting more involved. Rishard Matthews and Delanie Walker are steady, especially for 2017's unpredictable environment.

•Derrick Henry and DeMarco Murray are still in a 50/50 timeshare on the ground, though Murray scored all three TDs and caught all four passes.

•Andy Dalton played passably, but hasn't lit anyone up, and that was the case even with A.J. Green breaking a short throw for a 70-yard TD late. Brandon LaFell led the team in targets for some reason.

•Joe Mixon scored another TD, but got only nine carries.

• It's amazing the Browns blew the cover despite getting 12.5 points.

•Matthew Stafford went for 249 yards and three TDs, almost all of it in the second half. He took four sacks and threw a pick, though.

•Golden Tate was the only Lions receiver with more than three targets. He's a top-10 wideout too. After a huge game last week, Marvin Jones had one catch for 22 yards. Ameer Abdullah saw only 11 carries, but ran well, while Theo Riddick went four for 34 and caught a short TD.

•DeShone Kizer had a good fantasy day with seven rushes for 57 yards, but he threw a third-and-goal pass well out of the end zone, managed only 6.3 YPA and gave away points at the end of the first half when he audibled to a QB draw with no timeouts for God knows what reason. It looks like his job the rest of the way, though.

•Isaiah Crowell and Duke Johnson both ran well, with Johnson also catching six passes.

• Both Aaron Jones (knee) and Ty Montgomery (ribs) left the game, so Jamaal Williams got 20 carries, but racked up only 67 yards. That was more than Jordan Howard who had only 54 on 15 carries. With Jones out 3-6 weeks, Williams should get consistent work going forward.

•Mitchell Trubisky played well with 8.5 YPA, a TD and no picks. He took five sacks, though. Brett Hundley had a similar stat line and took three sacks.

• Bears coach John Fox made arguably the worst challenge of all time. Bears running back Bennie Cunningham hit the pylon with the ball, but was ruled down short of the goal. Fox threw the flag, and replay showed Cunningham lost control of the ball before touching the pylon with it. Hence, it was ruled a fumble through the end zone for a touchback. So Fox won the challenge and cost the team a turnover and 19 yards of field position, to boot.

•Davante Adams is the only Packers receiver to own. He had an 8-5-90-1 line, while no other Packer receiver saw more than four looks.

• Recent acquisition Dontrelle Inman led the Bears in receiving, catching six of eight targets for 88 yards. He has to be considered their No. 1, for what it's worth. Josh Bellamy caught the TD, but hauled in only two of seven targets. Kendall Wrong is merely a short pass catcher.

•Drew Brees is playing well and even had a rushing TD, but the days of him getting anywhere near 5,000 yards and 30 TDs are long gone in this new offense.

•Michael Thomas got his (10-9-117), but the running backs do all the scoring in New Orleans. Ted Ginn did next to nothing.

• If you started both Saints RBs the last few weeks, you'd be crushing it. This week they went 33-237-4 on the ground, and Alvin Kamara went 5-for-32 in the air. Ingram had three of the TDs, and he's probably the favorite to lead the league in TD scoring from this point on.

• The Saints defense shut down Tyrod Taylor (3.1 YPA) entirely. LeSean McCoy ran well (8-for-49), but the game got out of hand early. Kelvin Benjamin had six targets, three catches and 42 yards, so they're at least trying to involve him.

•Cam Newton missed a throw or two, was victimized by a couple drops but still had a monster fantasy game with four TD passes, 254 passing yards and 95 yards on the ground. He's running as much as ever, and the team will get Greg Olsen back after the bye.

• Funchess was quiet for much of the game, but caught five of his six targets for 92 yards and scored twice. He looked pretty quick for such a big receiver on the screen he took to the end zone for his first score. He's worth at least as much now as Kelvin Benjamin ever was, though Olsen's return could complicate things.

•Curtis Samuel dropped a would-be TD, but caught five of seven targets for 45 yards. Newton looks his way, and though it's mostly short stuff, he did see a downfield shot on the TD drop.

• Despite scoring twice, Christian McCaffrey actually had a pedestrian game – only three catches for 27 yards on seven targets and 23 yards rushing. Jonathan Stewart came back from the dead to rush for 110 yards against a soft Miami defense.

•Jay Cutler played okay, though he missed wide open receivers at least twice. Cutler has good timing, is willing to take a hit to get a throw off, but he's usually slinging the ball with his arm rather than stepping into his throws. He didn't have a ton of time to throw, but he wasn't under assault, either. Just a lot of flinging it and hoping for the best.

•DeVante Parker saw nine targets, Kenny Stills eight and Jarvis Landry 10. Landry mostly gets short stuff, but caught a garbage-time TD.

• For some reason Damian Williams started, but Kenyan Drake had the big play, a 66-yard TD run. Each had two catches.

• After a big Week 9, Julius Thomas was quiet but for a shovel-pass TD catch at the goal line.

• For some reason, I thought Adam Gase was a good coach heading into the year. Maybe it was what he got out of Ryan Tannehill last year. But this is one of the weakest down-in, down-out teams in the league, much worse than its record.

For more on these topics, check out our podcast.