David’s Week 7 FanDuel Lineups: Targeting Targets

October 23, 2015 - by David Hess

I noticed something interesting in Week 6.

I intentionally devote roughly half my bankroll (or more) to head-to-head matches. I do this because it should reduce my variance. If my lineup is in the 75th percentile for a given week, it should win about 75% of its head-to-head matches, and if it’s in the 25th percentile it should win about 25%.

Compare that to 50/50’s, where a 25th percentile lineup should lose every 50/50 it enters, and a 75th percentile lineup should win every 50/50. Entering head-to-heads essentially smooths out the binary win/loss results.

However, in Week 6, a single user fulfilled almost all of the roughly 100 head-to-head contests I created. So instead of playing against 100 different opponents, I played mainly against one (though he did use a couple different lineups against me). My lineups were slightly below average this week, and that user ended up beating me, so I lost nearly every head-to-head contest.

This week, to prevent that from happening again, I plan on fulfilling other peoples’ open head-to-head contests, rather than creating my own.

New Lineup Research Sources

As is becoming a weekly habit, I spent some time Thursday looking for new information sources or strategy breakdowns online.

Here’s this week’s haul:

NFL most pass targets per game over the last 4 weeks. This is actually data for our site that I found handy when creating my lineups this week. When making a 50/50 lineup, one thing I want to avoid is a big zero from one of my receivers. It seemed to me that focusing on players who have been targeted a lot recently could be one way to try to avoid a super low score. So, in this week’s lineups I called a few audibles. In some cases I diverged from simply following the projections I found on other sites, and tried to fit some low-priced, high-target receivers into my lineups.
Why We Shouldn’t Think About Ownership by Single Players. This is a nice conceptual article from Fantasy Labs about player ownership rates. Specifically, it talks about the idea that having specific unique combinations of players in a GPP is important. This is kind of obvious, but a lot of times people focus on either low-owned single players or on unique lineups, and don’t think so much about low-owned player combos.
What We Know About Games In London. Another Fantasy Labs article, this one is a data dump rather than a conceptual discussion. Basically, in the games at Wembley so far (an admittedly small sample), players on the favored team have outpaced their projections, while players on the underdog have fallen short. Note that Sunday’s TRFC 1:00 pm ET deadline doesn’t include the morning game in London.
FanDuel Player Ownership database. This web app lets you view FanDuel player ownership rates for past contests.
Who Has The Best Fantasy Football Projections? This article tries to answer the question in the title. Somewhat unsurprisingly, a wisdom-of-the-crowd approach was deemed the winner.

Hopefully some of you find those links useful.

Week 6 Results Recap

Initial 2015 NFL Bankroll: $1,000.00
Bankroll Entering Week 5: $973.20 (-2.7%)

Head to Head & 50/50 Entry Fees: $130 (13.3% of bankroll)
Head to Head & 50/50 Net Winnings: -$83 (-64% ROI)
Tournament Entry Fees: $71 (7.3% of bankroll)
Tournament Net Winnings: -$71 (-100% ROI)

Total Entry Fees: $201 (20.7% of bankroll)
Total Net Winnings: -$154 (-77% ROI / -15.8% of bankroll)

Bankroll Entering Week 6: $819.20 (-18.1% for the season)

Week 6 Lineups

Head-to-Head & 50/50s

As a quick reminder, the original strategy here was to be willing to pay for volume points at QB & RB, even at the expense of a little bit of value.

In addition, it would be nice to pick RB’s on opposing teams if possible, but that’s not a requirement. And I’m trying to avoid choosing too many players whose projection errors are likely correlated, based on the positional scoring correlation data I found last week. Aside from the recent changes, you can read more background about my current strategy if you’d like.

Here’s my primary lineup for Week 7:

Some notes on this lineup:

Philip Rivers was the projected best value at QB according to both RotoWire and numberFire.
Devonta Freeman is probably going to be owned by everybody this week, so I want to use him in one of my two lineups. This is basically like picking the #1 seed to make the Final Four in a March Madness bracket. I’m following the crowd here, to prevent losing ground to the whole pool (or most opponents) if he does well. And I’ll differentiate myself elsewhere.
DeAndre Hopkins is another “chalk” play, and a target monster.
Stefon Diggs has averaged 9.5 targets in his two games, and Jamison Crowder has averaged 9.7 targets over his past 4 games, so they both seem to have fairly high floors for such cheap players. Diggs could lose snaps this week with Charles Johnson set to return, however.
The Atlanta defense gets to face Zach Mettenberger this week instead of Marcus Mariota.

Here’s my secondary lineup:

Notes on this lineup:

Carson Palmer was the next best consensus value at QB in the projections I surveyed.
Le’Veon Bell figures to get a lot of touches since Landry Jones is likely to start for the Steelers.
Latavius Murray is facing a Charger defense that has allowed the second most FanDuel points per game to running backs.
John Brown is questionable for Arizona, and if he misses the game, Larry Fitzgerald seems like a good candidate for increased targets.
Steve Smith and Donte Moncrief were simply consensus good values.
Delanie Walker is 4th among all tight ends in targets per game over the past four weeks, with 8.7, and has the lowest salary from among those four. In addition, Atlanta has allowed 19 targets and 67 yards per game to tight ends (compared to an average of 12 targets and 50 yards).
The Cleveland defense is facing St. Louis, who have “allowed” the second highest average FanDuel points to opposing defenses.

Tournament Lineups

As was the case last week, my first step this week was to come up with some single-team stacks that seem to have the potential for a big day. Some of these stacks were based on fantasy-points-allowed to each position from this season (using RotoGuru as my source), and others were keyed on a particular player or two who were well-projected by consensus sources.

Here’s what I ended up with:

New York Jets (@ NE) — Ryan Fitzptrick (QB), Brandon Marshall (WR), Eric Decker (WR). Ignoring the Dallas game where they faced Brandon Weeden, the Patriots have faced eight #1 or #2 receivers, and have given up a touchdown to six of them. They’ve allowed the 7th-highest average FanDuel points to opposing receivers.
Indianapolis (vs. NO) — Andrew Luck (QB), T.Y. Hilton (WR), Donte Moncrief (WR). The Saints have the worst pass defense in the league on a per-play average, once you’ve adjusted for opponent strength, according to Football Outsiders. In addition, this game has the highest over/under of the week, and the Colts are favored. It seems like a great spot for Andrew Luck to go off.
Pittsburgh (@ KC) — Landry Jones (QB), Martavis Bryant (WR). This is what I’d call a “value stack.” The Chiefs have allowed the most wide receiver FanDuel points in the NFL, so it seems like a solid spot to take a gamble on a cheap backup QB getting his first start. As for why Martavis Bryant instead of Antonio Brown? Jones hooked up with Bryant for two touchdowns against Arizona after he replaced Michael Vick. It makes sense that Jones would have a better connection with Bryant than with Brown, since Jones and Bryant played together frequently in the preseason, while Brown was generally only in the game when Ben Roethlisberger was playing QB.
San Diego (vs. OAK) — Philip Rivers (QB), Malcom Floyd (WR), Ladarius Green (TE). Even after a 0-point game from Owen Daniels in Week 5, the Raiders have still allowed the highest average FanDuel points to tight ends. My original plan here was to use Antonio Gates, but his injury woes led me to substitute Ladarius Green, who filled in admirably for Gates over the first four weeks.
Arizona (vs. BAL) — Carson Palmer (QB), Larry Fitzgerald (WR), Michael Floyd (WR). The Ravens have allowed the second most FanDuel points to wide receivers. John Brown’s status for the game is uncertain, and if he’s out, Michael Floyd figures to be in line for an increased role. (He already scored last week, even with Brown playing.)

Filling in around the stacks with solid consensus values, here’s the result (click to enlarge):

As I did last week, I’ll be entering these lineups in a FanDuel Sunday Million tournament, as well as a smaller 100-person or 250-person tournament. The idea is that if they do astronomically well, I’ll be rewarded for that in the Sunday Million, but if they just score in the top 1% or so, I’ll get a better return in the 100-person tournament.

The most likely outcome for any specific tournament lineup is that I win nothing, but with 5 stacks, hopefully at least one will still do well enough to recoup my buy-ins, and let me tread water while I wait for a big payout. That strategy worked to perfection in Week 3.

Postscript: My Wife’s Lineup

At least I beat one opponent at FanDuel last week: my wife.

Those of you who read last week’s post may remember that I published a lineup created by my wife, which we entered in a GPP.

It was another down week for me, perhaps the worst week of the season. Given my results, I was expecting my wife’s lineup to have outscored mine. However, hers did even worse than any of mine, finishing in literally the bottom 1 percent of her tournament.

Crisis averted. I would have never heard the end of it.