This Week in Dynasty: Week 5

Photo: Mark J. Rebilas, USA Today Sports

Hey guys and gals, welcome to This Week in Dynasty. This is a new feature this year where we're going to discuss relevant developments from around the league with a dynasty slant. Everything is fair game, from high-level strategy to nitty-gritty player evals. If you have an suggestions for topics you'd like to see covered in this space, or if you'd just like to join the conversation, feel free to let me know on Twitter at @AdamHarstad.

The Big Takeaway

Since the beginning of the season, it seems like every week I've been advising caution and recommending against reading too much into results. Now that we're a quarter of the way through the season, we're really starting to see the cream rise to the top, so this is the perfect time to revisit our preseason expectations and adjust them to better match what's actually happened on the field. Right?

The season might be at the quarter point, but four games still represents a tiny sample. It's big enough for us to draw some conclusions about the major, obvious, giant-flashing-red-light-bulb issues (that Denver offense is going to score some points, that Jordan Cameron guy might be pretty good, those Jacksonville Jaguars aren't so great at… um, offense), but for stuff that's even only slightly obvious, we should continue to heavily temper our opinions with our preseason expectations. If you'll forgive me for a moment for discussing redraft value in a dynasty column, this concept is actually pretty easy to illustrate. Here's a list of the top 12 QBs in 2012 by ADP, by performance in their team's first four games, and by performance over their team's last 12 games. Which two lists look most similar?

Through four weeks, owners who took Aaron Rodgers as the first quarterback off the board had to be devastated with his performance. Through the rest of the season, he more than justified their faith. Top-5 quarterbacks Tom Brady, Drew Brees, and Cam Newton managed to perform in the top 5 over both windows, but fourth pick Matt Stafford went from a devastating liability (QB15 through four weeks) to a mild disappointment (QB8 from weeks 5 to 17). Peyton Manning and Tony Romo underperformed their draft position through four games, but were stars the rest of the way. Early-season surprises Robert Griffin, Matt Ryan, Andrew Luck, Joe Flacco, Ryan Fitzpatrick, and Andy Dalton all performed much closer to their ADP over the last three quarters of the season (although, admittedly, Griffin was hampered by injuries over that stretch).

For the mathematically inclined among us, the relationship between these lists can be expressed as a correlation, with correlations closer to 1 meaning there's a stronger relationship and correlations closer to 0 meaning there's a weaker relationship. I went through and looked at the top 24 QBs in 2012 by preseason ADP, removing everyone who missed more than two games, and calculated the correlation for the rest of the sample between preseason ADP and game 5-16 performance, as well as between game 1-4 performance and game 5-16 performance. Also note that I'm discussing game 1-4 and game 5-16, not week 1-4 and week 5-17. For any players who had a week 4 bye (Indy and Pittsburgh, last year), I added their week 5 performance to their week 1-3 performance and compared that to everyone else's week 1-4 performance. The correlation between game 1-4 performance and game 5-16 performance was a microscopic 0.132. The correlation between preseason ADP and games 5-16 performance was a much more robust 0.575. That's a Brobdingnagian difference.

Now, if anyone is wondering whether I've cherry-picked my examples to skew the comparison in my favor… congratulations, you get a gold star. The 2012 season was an especially dramatic season for quarterbacks- if we extend our look back over the last three years combined, the correlation between game 1-4 and game 5-16 becomes .351 and the correlation between ADP and game 5-17 becomes .611. If we look at all four positions (QB, RB, WR, and TE) in 2012, the overall correlation between game 1-4 and game 5-16 is 0.471, while the overall correlation between ADP and week 5-17 performance is 0.578. So even without cherrypicking examples, the numbers broadly support a continued reliance on preseason expectations.

If all of this sounds like meaningless number soup to you, here's the key takeaway: last year, preseason ADP better predicted rest-of-the-year performance than early season performance did at three of the four positions (quarterback, running back, and wide receiver). Early season performance held more predictive power at tight end, but both correlations were tiny (0.238 for early season performance vs. 0.171 for ADP), meaning the position as a whole was just wildly unpredictable.

All told, I looked at 95 players from last season (the top 24 QBs, 36 RBs, 48 WRs, and 24 TEs by preseason ADP, minus everyone who missed more than two games). Of those 95 players, 59 finished games 5-16 ranked closer to their preseason ADP than they did to their game 1-4 ranking. An additional 5 players finished games 5-16 ranked equally distant from both their preseason ADP and their game 1-4 performance. That means just 31 players ranked more similarly in games 5-16 to games 1-4 than to preseason ADP. Where a player's performance in games 1-4 diverged from his preseason ADP, that player was almost twice as likely to finish closer to his preseason ADP than he was to finish closer to how he started.

My favorite instance of this phenomenon comes from the 2006 season. Through four games in 2006, Donovan McNabb was the #1 fantasy quarterback and Peyton Manning was #2. Neither of those were much of a surprise. The #3 quarterback, however, was Rex Grossman. #4 was Charlie Frye. #5 was David Carr. #6 was Byron Leftwich. Two thirds of the top-6 quarterbacks through four weeks would wind up losing their starting jobs by the middle of the next season. It's easy to laugh about it with the benefit of hindsight, but at the time Rex Grossman, David Carr, and Byron Leftwich were seen by many as former first round picks who were finally putting it all together, and Charlie Frye was an exciting athlete who was making plays with his legs (he scored a rushing TD in each of his first three games). Plenty of owners bought high on those four quarterbacks based on their first four weeks of the season.

What does this mean? It means there's a reason I keep advising against jumping to conclusions every week, even though we're 1/4th of the way through the season. It's not that I'm particularly conservative by nature, it's because the data supports being conservative in this instance. It also means that, as all of your fellow owners are overreacting to what is still essentially just a minuscule sample size, there's a huge opportunity for you to swoop in and take advantage of the market overreaction. Before the season, everyone was sky-high on Russell Wilson, Colin Kaepernick, Ray Rice, Stevan Ridley, David Wilson, Larry Fitzgerald, Andre Johnson, Vincent Jackson (dumpster fire notwithstanding), Mike Wallace, Hakeem Nicks, and Jason Witten. Since, most of their owners have cooled considerably. Meanwhile, Philip Rivers, Alex Smith, Jake Locker, Knowshon Moreno, Bilal Powell, Danny Woodhead, Antonio Brown, Torrey Smith, Eddie Royal, Brian Hartline, Jordan Cameron, and Julius Thomas are playing miles above expectations, and history suggests that, while you're certainly free to crown them, many (if not most) of those guys will wind up being who we thought they were.

Am I suggesting that we should just blindly ignore the first four weeks of the season? Or that all of these early surprises are sure to regress and all of these early disappointments are guaranteed to rebound? Of course not. A lot of times, hot starts reflect an understanding that wasn't baked into preseason ADP. Michael Vick, Knowshon Moreno, and Julius Thomas all had preseason ADPs that were artificially low because we weren't sure whether they'd be starters. It turns out they are, so that early ADP becomes less meaningful. In 2012, Alfred Morris was drafted under the assumption that he'd be a backup in Washington, and Stevan Ridley was drafted under the impression that he wasn't certain to start, and even if he did, that he'd be mired in a timeshare. Through four weeks of the season, it was clear that both of those assumptions were mistaken. In situations like that, maybe early season performance is more meaningful than preseason expectations.

Sometimes, too, a hot start is just an appetizer for what's to come, even absent a change in situation. Demaryius Thomas was drafted, on average, as the 18th receiver off the board last year. Through four weeks, he ranked 12th. Instead of a signal he was due to decline, that ranking served as a warning shot to the league, as Thomas turned on the jets over his final 12 games and ranked 4th down the stretch. It's possible Torrey Smith is about to dominate the league and force his way into the conversation with Demaryius Thomas as a dynasty asset. It's possible Jordan Cameron is here to stay and the "big 2" tight ends will soon be the "big 3". It's possible that David Wilson's nightmare start is just a harbinger of worse days to come, that he'll fail to get his fumbling under control, and that he'll be out of the league within two years. It's possible that Jason Witten has lost a step and we're seeing the early signs of his precipitous decline. There are a lot of possibilities at this point, and it's important that if you feel strongly about a player, you take a stand. I'm happy to take a stand on Cameron, because he's looked so sensational that it's hard for me to imagine that his performance is a fluke. It's harder for me to take a stand on Julius Thomas, who has looked more like a very good player benefiting from Peyton Manning giving the league a masterclass in quarterbacking. In the long run, though, while taking the occasional stand is necessary, it's important that we're ranking less on what's possible and more on what is probable. In all likelihood, the under-performers, as an asset class, are woefully undervalued right now, and the over-performers are priced too high. While we might lose out on individual trades in the short run, in the long run, trading guys who are studs through the first quarter of the season for guys who are disappointments is the kind of move that will generate substantial returns.

Heard Around the Water Cooler

THE EYE TEST:

Seeing praise for Robert Woods. So good.
-Joe Bussell (@NFLosophy)

Quinton Patton passed Marlon Moore and Kyle Williams on depth chart this week. That catch for zero yards was his first NFL reception.
-Matt Maiocco (@MattMaioccoCSN)

Kendall Hunter could go somewhere and thrive, a la Michael Turner, Lamont Jordan, Chester Taylor or Darren Sproles.
-John Paulsen (@4for4_John)

Jordan Cameron has that same unstoppable look in red zone as Rob Gronkowski. Just throw it in his vicinity and it's a score.
-Sigmund Bloom (@SigmundBloom)

Ben Tate is going to make a lot of money next year.
-Chase Stuart (@fbgchase)

Reggie Bush has looked like a legit superstar today.
-Patrick Daugherty (@RotoPat)

Hands to the face by OL bringing it back, but that Luck TD throw shows every bit of why he was one of the greatest prospects ever.
-Chase Stuart (@fbgchase)

Realize there's skepticism on this, but 2013 Britt <<<< 2011 Britt. No solution
-Chris Wesseling (@ChrisWesseling)

Running with Mendenhall is like punching a brick wall with your bare fist. The only thing you're hurting is yourself.
-Joe Bussell (@NFLosophy)

Michael Floyd has been terribly unlucky in TD department so far. He's due. Keep an eye on him here.
LOL - Another endzone target for Michael Floyd. That one picked off. Guy can't catch a break.
-Mike Clay (@MikeClayNFL)

Five catches for Heath Miller. Locked back in TE1 conversation.
-Mike Clay (@MikeClayNFL)

Chris Johnson runs ball down to the 4, but again pulled for Battle. This is consistent and not going to change any time soon. #FFvaluekiller
-Sigmund Bloom (@SigmundBloom)

Terrance Williams continues to impress in limited time. Makes sense why trade rumors exist RE Miles Austin
-Cian Fahey (@Cianaf)

Fantasy owners, White looks like a much safer start for next week. Cutting, planting, driving much better than past few weeks.
-Matt Waldman (@MattWaldman)

In terms of high pointers, I love Mike Williams as one, but lets not get creative here. It's Calvin+Fitz, then everyone else.
-Cian Fahey (@Cianaf)

Last night was best Ive seen Stevan Ridley run. Attacked gaps with velocity, incredibly light feet for power RB. Even made plays in pass gm.
-Evan Silva (@evansilva)

Torrey Smith is going to shatter every previous career mark. Was talk of diversifying his routes in past seasons. It has finally happened.
-Evan Silva (@evansilva)

Watching BAL-BUF: Ray Rice looks fine to me. I'll be putting offers out there for him in all of my leagues.
-Sigmund Bloom (@SigmundBloom) 

THE NUMBERS GAME:

Two QBs are on pace for 700+ pass attempts....Sam Bradford and Robert Griffin III.
-Chad Parsons (@ChadParsonsNFL)

Never would have guessed it, but Robert Woods is WR7 in Deep target rate. #Bills
-Chad Parsons (@ChadParsonsNFL)

Garrett Graham wasn't even targeted last week. Already to 4-68-1 this week.
-Mike Clay (@MikeClayNFL)

CJ Spiller's season-high for carries in a game was 17. He has 17 carries in the FIRST HALF today.
-Adam Levitan (@adamlevitan)

Jamaal Chrles on pace for 92 receptions this season. #Chiefs
Jamaal Charles targets by week: 6 - 10 - 8 - 9 . Don't care about Reid's past or Charles' ability. That's still enormous for a tailback.
-Mike Clay (@MikeClayNFL)

Doug Martin on pace for 400 carries, yet to break 15 FF pts in any game thru four weeks.
-Sigmund Bloom (@SigmundBloom)

Eagles are first team in NFL history to net 430 or more yards three straight games and lose all three.
-Reuben Frank (@RoobCSN)

Opposing No. 1 WR stats vs. #Patriots through 4 games: Julio Jones 4-39; Vincent Jackson, 3-34; Santonio Holmes 3-51; Stevie Johnson 3-39-1.
No. 1 WR stats vs #Browns: A.J. Green 7-51; Mike Wallace 1-15; Torrey Smith 7-85; Greg Jennings 3-43. No TDs.
WR stat lines vs #Bills: Torrey 5-166-1; Santonio 5-154-1; Stephen Hill 3-108-1; Ginn 3-62-1, Amendola 10-104, Edelman 7-79-2. Gordon next.
-Evan Silva (@evansilva)

Stevan Ridley (21.8 pts.-RB50) still trails Shane Vereen (22.9) in PPR scoring.
-Rich Hribar (@LordReebs)

Le'Veon Bell played 60 snaps, while Jones/Dwyer combined for 14. No quesiton he's healthy/the workhorse. RB2 worth targeting.
-Mike Clay (@MikeClayNFL)

#Bengals backfield last 2 games: Gio Bernard 80 snaps/30 touches. Green-Ellis 41 snaps/16 touches. Gio 4.59 YPC. Law Firm 2.73 YPC.
-Evan Silva (@evansilva)

AJ Green on pace for 200 targets (2nd only to Shorts), yet is only 10th in catches, 21st in yards. #daltoneffect
-Ryan McDowell (@RyanMc23)

Eli Manning throwing to Victor Cruz yesterday = 125.4 QB rating. Throwing to Hakeem Nicks = 5.6
-Pro Football Focus (@PFF)

No player with 25+ targets has a higher aDOT than Michael Floyd (16.9). Another thing to like.
-Mike Clay (@MikeClayNFL)

David Wilson has handled 73% of the Giants' carries over the last two weeks. Big number.
-Mike Clay (@MikeClayNFL)

Emmanuel Sanders' 2013 low for targets in a single game: Eight.
-Mike Clay (@MikeClayNFL)

Clear separation between #Cardinals WRs M. Floyd & A. Roberts since Housler's return. Roberts down to 56% of the snaps. Floyd is above 90%.
-Evan Silva (@evansilva) 

THE SAM BRADFORD ERA

Sam Bradford is going to be a great backup quarterback one day.
-Chase Stuart (@fbgchase)

Screw Alex Smith. The 4-year rule works. If your QB hasn't shown up by year 4, it's not happening (Bradford).
-Scott Kacsmar (@FO_ScottKacsmar)

This is 10th time Sam Bradford threw 35+ passes & averaged < than 5 yards per pass. That's more than twice as many as any other QB since '10
-Chase Stuart (@fbgchase)

I finally got around to watching Josh Freeman's performance last week. It wasn't nearly as bad as I expected, or Sam Bradford's.
-Football Robert (@FootballRobert) 

THE JOSH FREEMAN SAGA

Josh Freeman has lost the locker room…according to Greg Schiano’s Minister of Propaganda.
-Rumford Johnny (@RumfordJohnny)

Waiting for the Bucs to reveal that Josh Freeman has watched some "pretty freaky stuff" on his team-issued iPad.
-Bill Barnwell (@billbarnwell)

Freeman just took the metaphorical PR grenade that the Buccaneers threw at him and threw it back with a nuclear bomb attached to it.
Schiano is going to get absolutely pummeled for this. Rightfully too.
-Cian Fahey (@Cianaf)

It's funny the Bucs thought this would help their PR battle.
-Gregg Rosenthal (@greggrosenthal)

Filed to ESPN: Bucs have released QB Josh Freeman. Bucs tried to trade him but couldn't.
-Adam Schefter (@AdamSchefter) 

ALSO OF INTEREST:

Surprising stat: The only two teams with less sacks allowed than Peyton Manning and the #Broncos....#Lions #Bears.
-Chad Parsons (@ChadParsonsNFL)

I spend more time thinking about HOW I think rather than WHAT I think. One is a cause. The other is an effect. The cause is more important.
-Joe Bussell (@NFLosophy)
(If I had an award for "Tweet of the Week", this would be your winner. On an unrelated note, perhaps I need to consider adding an award for "Tweet of the Week".)

Since Jets hired Rex Ryan in 2009, he and Tom Coughlin have same number of regular-season wins, 36, same number of post-season wins, four.
-Adam Schefter (@AdamSchefter)

Jay Glazer on Rob Gronkowski: "This is Gronk's decision & only Gronk's decision." Targeting return in 2 weeks against Saints.
-Evan Silva (@evansilva)

Yay fantasy football......... RT @SethWickersham: Sometimes I think we'd understand football better if we didn't keep individual stats
-Chris Brown (@smartfootball)
(It wouldn't hurt so much if it wasn't true…)

Teh eJtss hjeve sow myna pentliaes thIm n0w drrnk.
-Chase Stuart (@fbgchase)
(Real friends don't let Jets fans play drinking games.)

If you Breaking Bad people really wanted dark TV, you'd be tuned into whatever Tom Brady does to Kenbrell Thompkins if he drops another one.
-Robert Mays (@realrobertmays)

Peyton Manning's last 4 TD drives Sunday didn't require a single third down conversion.
-Gregg Rosenthal (@greggrosenthal)

Second Thoughts

Scott Kacsmar mentioned the "4 year rule" for quarterbacks, which posits that if a quarterback isn't showing it by the end of year four, he never will. I'd actually go even further; if a quarterback has not become fantasy relevant by the end of his second full season as a starter, he is unlikely to do so. This will probably be the topic of its own "Big Takeaway" towards the end of the season, but for now, know that the list of exceptions is very short and pretty much just boils down to Drew Brees and Matt Ryan. Even Tom Brady, generally known as a "game manager" who didn't compile stats early in his career, managed to finish as the #10 fantasy QB in his second year as a starter.

Touching slightly more on this week's Takeaway, I know it seems like buying low on some underperforming players or selling high on some overperforming players is a terrible idea. And in some instances, it might be. The key to building a contender in dynasty, however, isn't to win every single trade. If you're winning a lot more often than you're losing, you can survive making some terrible trades in the process. It's also worth remembering that just because players are still likely to regress towards their ADP doesn't mean you have to pay prices commensurate to their ADP to land them. There's too big of a discount already baked into the system for this year's underperforming stars, which means in many cases, you won't have to stick your neck out too far.

I've devoted a lot of attention in this space to the Josh Freeman saga, and it's not because I'm a Freeman fan, or because I think the saga will send huge ripples throughout the fantasy landscape. I've devoted so much attention to it because I think this is a case where the easy narrative is the wrong narrative. Josh Freeman is not a great quarterback, but he's not anywhere near as bad as he's made out to be. He's better than a lot of guys who are keeping their jobs. His struggles trace pretty strongly back to a coaching staff that has never embraced him or set him up to succeed. It's interesting that Sam Bradford has a firm grip on the franchise QB job in St. Louis, while Freeman was just cut by Tampa, because Sam Bradford is no better than Josh Freeman. He's arguably even worse. He's certainly shown less to this point in their respective careers. Some team is going to take a chance on Josh Freeman. If that team gives him an honest chance and puts him in a position to succeed, then this is not the last we'll hear of Freeman. For a primer on what kind of difference a simple change of scenery can make, compare Jake Plummer's numbers in Arizona to his numbers in Denver some time.

Perhaps it's my imagination, but it seems like a greater percentage of NFL teams occupy the "middle class" this year than in most seasons. That tendency has already shown itself in a ridiculous rate of one-score games. As a fantasy fan, this is an exciting development, because we're seeing more of a team's "true" offense, unskewed by huge defecits or leads. I don't know if that will wind up making fantasy football more predictable this season- I suppose we'll know in week 5 of next year when I re-run the correlations- but in the meantime, it's certainly making for a much more entertaining slate of games. As a fan of the NFL, what more could you ask for?

While there are a lot more teams in the middle, this year has also produced some pretty notable "Haves" and "Have Nots". Las Vegas has just opened betting on next week's game between Denver and Jacksonville. Denver opened as 28-point favorites; if that line holds up, it will be the largest spread in NFL history, besting the 27 points the Steel Curtain Steelers got against the 0-14 expansion Buccaneers. This says a lot about the Denver Broncos, but nothing everyone isn't already familiar with- fantasy owners everywhere know to start their Broncos where they have them. The fantasy implications of Jacksonville's awfulness might be just as important, and for most owners, they're actually in a position to do something about them. In recent years, streaming defenses has become more popular. The idea is that it's the gap between defense and offense that matters, that a mediocre defense going against a terrible offense will score as much as an elite defense going against an average offense. As terrible as this year's Jaguars are, everyone knows to roster whatever defense is facing them. I would suggest, however, that Jacksonville is so awful that it might be worth pre-emptively rostering some of the teams that will face them a week or two in advance. A lot of people are vehemently opposed to rostering multiple defenses, preferring to devote that space to a prospect, instead. In this case, however, with as bad as Jacksonville has been and how quickly they're trading away their best offensive players, I think devoting that extra spot to a defense will result in a lot more points than rostering an 8th receiver or a 7th running back.

I'm pretty obviously an advocate for the usefulness of Twitter. The ability to get breaking news first has netted me players like Willis McGahee and Austin Collie before anyone else had a chance at them, and it's nice to hear what so many others are thinking about a particular player. With that said, it's not without its annoyances. I'm probably the only guy in America who's excited for Breaking Bad to finally be over so the tweets in my timeline can finally get back to focusing on football again.

Best of luck to everyone in their Week 4 games. I'll see you back here next week with plenty of NFL action to break down and plenty of dynasty implications to discuss!