This article is part of our East Coast Offense series.
Expert Failure
"In the beginner's mind there are many possibilities, but in the expert's there are few."
- Shunryu Suzuki
I took the Patriots in Survivor last week, and while I never had a strong feeling about them, I also didn't feel great about the Bills and Steelers, the two other teams I seriously considered. (The Falcons, seven-point favorites, were a lesser consideration.) Of three mediocre options, I picked one, and it turned out to be wrong. Nothing to beat myself up over, right? It's how Survivor goes sometimes, and you have to let it go.
But that's not the entire story. The truth is the team I instinctively liked the most and for whom I picked the largest margin of victory in my Beating the Book column last week was the Rams, 24 - 6 over the Colts. A reader asked me on Twitter why I didn't pick the Rams, given that score, and I told him the predicted scores are really just off-the-top-of-my-head hunches, and I pick Survivor according to the numbers. By which I meant (1) The probability of victory; and (2) The distribution of ownership. You can read here to understand the entire framework. While those factors are without doubt the correct ones on which to base a pick, I actually wrote a piece for the 2017 RotoWire magazine questioning the way we measure those two factors.
For (1) using the Vegas moneylines as a proxy for win probability is dicey. While Vegas sets fairly accurate lines generally, that doesn't mean there isn't massive variance as to their accuracy in particular instances. The Patriots were favored by 8.5 on Thursday night and lost by 15. Even if there's noise in that final margin, there's a strong case to be made that the Patriots are not quite as good as we thought pre-Week 1, and the Chiefs (with a healthy Eric Berry) were better than we assumed. Averages are not that useful when dealing with particular cases - for example if you need to be taller than five feet in order not to fall out of a rollercoaster, the amusement park would have a lot of blood on its hands if it simply maintained a 5-foot-5 average over the course of the year.
So I knew the fallacy of subbing in Vegas' numbers (a combination of algorithms and public perception) for the actual probability a team won or lost. Even so, because the Rams were merely 3.5-point favorites - and not even close to the 8.5-ish numbers of the Patriots, Steelers and Bills - I didn't consider them seriously, despite commenters in the column picking them, and they were the team I picked to win by the widest margin. Had the Rams been seven or even six-point favorites, perhaps I would have picked them, but beneath that threshold, they didn't make the cut in my "expert" opinion.
This is absurd, as whether Vegas says the Rams are 3.5 or 10-point favorites has no obvious impact on how they'll perform. Using Vegas' moneylines (or spreads which correlate tightly with them) as a proxy for probability then is at best imprecise and at worst lazy and cowardly - instead of relying on my own estimations and risking the error, I'm outsourcing them to the often-incorrect marketplace as a way to bolster the legitimacy of my endeavor, i.e., cover my ass. The Survivor column acknowledges this to an extent - the probabilities I assign to each of my picks differ slightly from Vegas' every week, but even then, it's only within a certain narrow range.
What I missed - and that I'll try not to miss next time - is when my opinion differs substantially from Vegas', I need to be willing to go way outside the permissible bounds of disagreement with the market. This might increase the probability of errors - it will certainly increase the probability I look stupid - but it's also the only way to stay true to my observations so that if and when I lose it won't be because I took bad advice from Vegas, but from myself. In the end, you either have to trust yourself or know your limitations. If the best I can do is point you to the Vegas' numbers and Officefootballpools.com ownership stats, I should ask our tech guys to build an automated tool and get out of the game. Maybe that day is coming at some point, but for 2017 I still want in.
Disaster Week
I can't remember a Week 1 where so many highly drafted players had terrible games. Let's go through the top two rounds (per NFFC ADP) pick by pick.
1.1 David Johnson - 23 yards rushing, six catches for 68 yards, two fumbles, one lost, out for 8-12 weeks with a wrist injury.
1.2 Le'Veon Bell - 32 yards rushing, three catches for 15 yards
1.3 Antonio Brown - 11 catches for 182 yards
1.4 Julio Jones - four catches for 66 yards
1.5 Odell Beckham - Out
1.6 Mike Evans - Game Postponed
1.7 LeSean McCoy - 110 yards rushing, five catches for 49 yards
1.8 A.J. Green - five catches for 74 yards
1.9 Melvin Gordon - 54 rushing yards, five catches for 25 yards and one TD
1.10 Jordy Nelson - seven catches for 79 yards and one TD
1.11 DeVonta Freeman - 37 yards rushing and one TD, two catches for two yards
1.12 Michael Thomas - five catches for 45 yards
2.1 Ezekiel Elliott - 104 yards rushing, five catches for 36 yards
2.2 Jay Ajayi - Game Postponed
2.3 DeMarco Murray - 44 yards rushing, two catches for 16 yards
2.4 Brandin Cooks - three catches for 88 yards
2.5 Jordan Howard - 52 yards rushing and one TD, three catches for 14 yards
2.6 Todd Gurley - 40 yards rushing and one TD, five catches for 56 yards
2.7 Doug Baldwin - four catches for 63 yards, one rush for minus three
2.8 Amari Cooper - five catches, 62 yards and a TD
2.9 Dez Bryant - two catches for 43 yards
2.10 Rob Gronkowski - two catches for 33 yards
2.11 Aaron Rodgers - 311 passing yards, 21 rushing yards, one TD pass, one INT
2.12 T.Y. Hilton - three catches for 57 yards
Keep in mind these are NFFC averages for the last couple weeks, so Kareem Hunt's ADP is 50, dragged down by his slot before Spencer Ware was out. Still, it's mind boggling there is not a single player with an ADP in the top two rounds with 100 yards from scrimmage and a TD.
The only players in Week 1 to accomplish that feat were Hunt, Leonard Fournette, Austin Hooper, Tyreek Hill and Tarik Cohen.
Contrast that with last year where six WR had at least 100 yards and a TD, four of whom (Green, Cooks, Brown, Cooper) were top-two round picks, and six running backs (David Johnson, C.J. Anderson, DeAngelo Williams, Danny Woodhead, Spencer Ware, Theo Riddick), including one first rounder (Johnson) did it.
So if you're aghast at the way your fantasy team performed, you're not alone.
Week 1 Observations
• The Giants offense without Odell Beckham is a joke. It's basically the Colts without Andrew Luck or the Texans with Tom Savage. Checking the ball down to small-play targets like Shane Vereen and Sterling Shepard was all they had. Paul Perkins went nowhere, the offensive line couldn't block, Brandon Marshall couldn't get open, and Eli Manning couldn't make a play. But at least Beckham wasn't "distracting" them. I've said it before – Eli Manning would be Matt Hasselbeck had the Giants not drafted Beckham three years ago.
• The Cowboys defense is probably bad, but they drew a laughably easy matchup. Don't read into it.
•Ezekiel Elliott looked good, but the Giants defense – though stout – got tired of being on the field the entire game. Dak Prescott also looked good, but had to do very little.
•Aaron Rodgers is a monster. Trying to stop him from running out the clock late in games is a nightmare. The Seahawks defense played great all game, but Rodgers' ability to scramble, his zip, accuracy, rapport with his receivers and decision making was too much.
•Jordy Nelson is no longer a big-play threat, but he's still Rodgers' clear No. 1 and catches everything.
• Maybe Ty Montgomery really is the Packers' three-down back. He wasn't efficient, but he'll have easier matchups than the Seahawks going forward.
• Seattle's style of offense whereby the QB runs for his life and throws on the move is high degree of difficulty – even for Russell Wilson.
•Paul Richardson looks like Seattle's No. 2 (seven targets to Tyler Lockett's three.)
•Jimmy Graham had a terrible game, dropping a third-down conversion that would have kept the Seahawks alive. Still, he had seven targets, at least one of which was in the end zone.
•Chris Carson was the only competent running back. And there's no compelling reason for Seattle to stick with Eddie Lacy or Thomas Rawls.
•Cam Newton looked awfully rusty, missing wide-open receivers and would-be touchdowns. I wouldn't worry about him long-term though.
• Christian McCaffery got a lot of work, 13 carries, seven targets, five catches. The broadcast kept trying to make the point he was a Steph Curry of sorts, making everyone else better because of his unique skill set that required the defense to play differently, but it seemed like a stretch.
•Kelvin Benjamin had only five targets and one catch. Newton spreads the ball around like a poor man's Drew Brees.
•Pierre Garcon is a decent bet to be top-10 in targets and catches this year.
• The Colts should be a pick 'em against the Jets right now on a neutral field. If Indy goes 0-4, does it even bother bringing Luck back this year? If so, to what end?
•Marlon Mack looks like Frank Gore's backup, though Robert Turbin got some work near the goal line.
•Todd Gurley was inefficient as ever, but scored a TD and caught five passes for 56 yards. The matchups won't get easier, either.
•Jared Goff looked sharp and accurate, Sammy Watkins healthy and Cooper Kupp and Robert Woods competent, albeit against a soft opponent.
• So much for my theory that Hurricane Harvey would inspire the Texans to play well. That narrative worked for the Saints after Katrina, but maybe it works only after female hurricanes.
• The Texans quarterback play in the first half was uncivilized, barbaric – "savage" if you will. It's amazing Bill O'Brien started Savage over the more mobile, improvisational DeShaun Watson behind that awful offensive line.
• If Allen Robinson is out for the year, Marqise Lee and Allen Hurns should be useful. Lee did nothing Sunday, but the Jaguars didn't need much from their passing game.
• Maybe the offseason hype about the Jaguars defense was legitimate, but we'll need to see them do it against a better opponent.
•Leonard Fournette didn't blow anyone away, but he was the team's workhorse with 26 carries and three targets.
•Nelson Agholor had a big game, but take away his 58-yard TD on a busted play, and he was 5-for-28 on six targets.
•Zach Ertz producing before Week 16 is a revelation. He's the team's No. 1 target right now.
•Alshon Jeffery had three catches on seven targets for 38 yards. The Eagles like to spread around, so he's in Kelvin Benjamin territory for me.
•Terrelle Pryor dropped a potential TD, but he had six catches on 11 targets. There's no reason not to be bullish on him.
•Marshawn Lynch looked surprisingly spry for a 31-year old that had been away from the game for a year. We'll see how he holds up as the year goes on, but so far so good.
•Amari Cooper finally got the red-zone looks of which he had been so unfairly deprived, but converted only one out of four.
•Corey Davis led the team with 10 targets, but it was a fairly even split between him, Rishard Matthews and Delanie Walker (nine each.) Eric Decker (three) was the odd man out in Week 1.
•Golden Tate is easy money in PPR – 12 targets, 10 catches, 107 yards, but Kenny Golladay made the big play and scored two TDs. While Marvin Jones caught a TD, he was again an afterthought with only two targets.
•Larry Fitzgerald was up to his old tricks – 13 targets for 74 yards. John Brown had nine targets and only caught four for 32, and you have to worry about Carson Palmer (5.6 YPA, three picks.)
•David Johnson was ineffective as a rusher, but still made plays in the passing game. Unfortunately, he sprained his wrist and might miss significant time. Some scrub – Andre Ellington or Kerwynn Williams – is likely to fill in, though the team could always bring in Chris Johnson's carcass. When was the last time the No. 1 overall player by ADP got seriously injured in Week 1? Think about how excited people were to draw the No. 1 pick too. What a disaster.
•Antonio Brown caught all 11 of his targets for 182 yards, while Martavis Bryant caught two of six for 14. It's hard to carve out a big role when the guy on the other side has a massive one. Still, let's reserve judgment until after a home game where the Steelers pass offense is typically far better.
•DeShone Kizer looked okay to me, given the limited snaps I watched and that it was his first NFL game.
• Jeff Erickson warned me about the Bengals offensive line, and I blew it off and drafted Dalton everywhere. Four picks, five sacks and a fumble later, I'm thinking maybe I should not have.
• Javoris Allen had 21 carries and could be the team's primary pass catcher now that Danny Woodhead is out. Incidentally, who thought it was a good idea to draft a 32-year old coming off an ACL tear and missing time with another preseason injury? Woodhead's absence is good for Terrance West too.
• The Bengals look like a three-headed committee at RB right now, though it won't matter if they can't block anyone.
• I wasn't worried about Jordan Howard – even after Tarik Cohen exploded onto the scene yesterday – until Howard dropped a potential game-winning TD on the team's final series. In Howard's favor is Cohen's build (5-6,179) makes him unlikely ever to see a heavy workload. Even Maurice Jones Drew was more like 5-6, 195. Still, Cohen could be a 100-carry, 50-catch player, highly valuable in PPR.
• Like last year Matt Ryan spread the ball around to everyone, bad news for Julio Jones whose ceiling depends on Ryan reverting to 2015 form. Austin Hooper saw only two targets, but made the most of them dismissing a hapless Bears DB with a straight-arm en route to an 88-yard TD.
• The Jets-Bills went pretty much to form, with Mike Tolbert vulturing LeSean McCoy's TD and Charles Clay being Tyrod Taylor's first look.
• There's no point in the Jets trotting out Josh McCown's rotted corpse. He's bad, and he's got nothing to offer them going forward. Whether it's Christian Hackenberg, or unknown as-yet-to-be-signed free agent, they have to make a switch.
• It's pretty sad Kevin White is likely out for the year with a broken collarbone. Looks like he's taken the Yatil Green career path.
•Sam Bradford looked awfully sharp, completing his usual high percentage of passes (27-for-32, 84%), but these were not dinks and dunks. Bradford had 10.8 YPA en route to 346 yards and three TDs. Granted a home game against the Saints is as easy as it gets, but Bradford was zipping around the ball with pinpoint accuracy. It didn't hurt that Dalvin Cook's running got Bradford in favorable down-and-distance situations, though one could argue Bradford's passing set up Cook too. Either way both will be better tested in Week 2 at Pittsburgh.
• Speaking of Cook, he had 127 yards on just 22 carries (5.8 YPC), and another 10 yards on three receptions. Latavius Murray and Jerick McKinnon had five carries combined.
•Adam Thielen caught nine of 10 targets for 157 yards for a whopping 15.7 YPT. It's hard to tell if he's just good at getting open or whether the Saints are that bad in coverage, but Thielen was often wide open.
•Stefon Diggs caught 7-of-8 targets for 93 yards and two TDs. He and Thielen are WR1 and 1a for the Vikings, though Thielen tends to run deeper routes.
• Three of the five 100-yard rushers in Week 1 (Cook, Leonard Fournette and Kareem Hunt) were rookies.
• The Saints running back committee totaled 20 carries, six for Adrian Peterson and Mark Ingram, seven for Alvin Kamara, none especially effective.
• The Saints receiver by committee totaled 37 targets, eight for Michael Thomas six for Coby Fleener and Kamara and five for Ted Ginn and Ingram.
• The Vikings defense is tough, but Michael Thomas (5 catches for 45 yards) didn't have a reception longer than 14 yards.
• Vikings kicker Kai Forbath made all three of his field-goal attempts in what was quite obviously a revenge game.
• It wasn't pretty, but Melvin Gordon is a reliable workhorse with 18 carries and six targets. Efficiency is less important than health here.
•Hunter Henry, who I termed a "can't fail" player if healthy this preseason didn't get a chance to fail Monday night – he didn't receive a single target and reportedly ran only six routes. It's possible Denver's ferocious pass rush caused the Chargers to have the tight ends block more (Antonio Gates had only three targets), but it's hard to spin this positively for Hunter.
•Keenan Allen caught five of 10 targets for 35 yards and a score. I'm not an Allen fan because I think he's Golden Tate without the track record of being healthy, but at Denver is about the worst matchup for a No. 1 WR you can have.
•Trevor Siemian looked competent and just missed a second rushing TD late in the game.
•A.J. Derby looked nimble on his three catches.
• Benny Fowler caught two TDs, but it was on only four targets. Don't read anything into it. Emmanuel Sanders and Demaryius Thomas should still see the bulk of the work.
For more on these topics, check out our podcast.
