Seth’s Week 4 Daily Fantasy Analysis and FanDuel Lineup

October 3, 2015 - by Seth Trachtman

My Week 3 FanDuel lineup scored some winnings in flatter payout structures, but my overall results ended up down overall because I entered it into more tournament-style contests than 50/50s and double ups. I scaled back to only five contests with my lineup this week, but three of those contests were tournaments paying out to the top 25th percentile.

Judging by my results, my lineup finished somewhere around the top 40th percentile of entries.

The Week 3 results:

5 entries
$45 in total entry fees
$16 net loss ($29 in contest prizes)
1 Championship ticket won (finishing ahead of 5 expert entrants to win a ticket to the RotoWire Football Championship finals)

After three weeks I’ve still made a solid profit on FanDuel:

Week	# Contests Entered	Total Entry Fees	Total Prize Winnings	Net Gain/Loss for Week	Finals Tickets Won	Cumulative Profit	Cumulative Finals Tickets Won
1	5	$32	$295	$295	4	$295	4
2	9	$82	$0	($82)	0	$213	4
3	5	$45	$29	($16)	1	$197	5

Overall Strategy

Given that my lineup is primarily built for the TeamRankings Football Championship, up until now I’ve been hedging my bets a bit.

In short, I’ve been trying to build a team with a relatively “high floor” in scoring to double or triple my money based on the weekly contest format, while also having a fighting chance to land a huge score, finish in the top 3, and win more tickets to the $25,000 Week 11 Final. (I scored one of those in Week 1.)

Going forward, I’m going to more specifically target my lineup to profit in double ups and 50/50s. That will hurt my chances to finish in the top 3 in the TRFC on any given week, but my ultimate goal in FanDuel is to be consistently profitable with a relatively modest number of weekly entries.

Week 4: The Beginning of Bye Weeks

There will be at least six teams unavailable for Sunday’s 1 pm ET contests, and possibly eight if the Eagles-Redskins game is postponed due to Hurricane Joaquin. They means fewer choices for FanDuel lineups, and specific contest lineups that are even more similar/popular than usual.

I briefly discussed pick popularity and ownership percentages over the last two weeks, but wanted to dig deeper this week. We’ve gotten our hands on all entry ownership percentages from FanDuel contests last season, which creates some great possibilities for studying the data.

This week I wanted to study the wisdom of the crowd, and look at how ownership percentages correlate with player performance. In other words, how good are FanDuel users at picking the highest scoring players?

The correlation for all of last season between player entry ownership percentage and points earned is 0.438. That medium correlation implies that the FanDuel crowd does have some ability picking the top scorers, but the correlation isn’t particularly strong.

Breaking this correlation down by position does provide some interesting observations.

2014 Fantasy Points Scored vs. Entry Ownership Percentage Correlation

QB	RB	WR	TE	K	DEF
0.390	0.544	0.492	0.436	0.181	0.278

It shouldn’t be a shock to see the weaker correlation for kickers and defenses. These positions have a small range of salaries, and especially with kickers, whose scored tend to be inconsistent from week-to-week.

However, looking at 2014 weeks with byes (Weeks 4-12) shows a higher correlation across the board, with the exception of slightly weaker correlation for running backs.

2014 Fantasy Points Scored vs. Entry Ownership Percentage Correlation (Weeks 4-12)

QB	RB	WR	TE	K	DEF
0.448	0.525	0.558	0.445	0.203	0.351

With fewer players to choose from, the correlation increases.

The takeaway here is that while FanDuel players shouldn’t by default shy away from picking the most popular players (especially in 50/50s and double up contests), going against the crowd and taking unique players with lower ownership percentages — particularly at the positions with weaker correlations — has value.

At the very least, it’s clear that the most popular players are far from a lock to deliver the highest scores.

Next week we’ll build on this ownership percentage analysis by looking at where the top players at each position ranked in ownership percentage each week.

My Week 4 Strategy

My 3-step strategy didn’t change much going into Week 4, though I think it’s wise to use more 2015 data for matchups after three weeks of games.

To review my strategy:

Start with Vegas over/under lines
Evaluate each team’s recent fantasy points allowed vs. each position
Evaluate salaries and contest format

The highest Vegas-implied point totals include:

Team	Vegas Implied Points
Indianapolis (vs. JAC)	28.8
Green Bay (@ SF)	28.2
Atlanta (vs. HOU)	26.8
Seattle (vs. DET)	26.5
San Diego (vs. CLE)	26.0
Buffalo (vs. NYG)	25.5

The lowest Vegas-implied point totals include:

Team	Vegas Implied Points
Detroit (@ DEN)	16.5
St. Louis (@ ARI)	17.5
Minnesota (@ DEN)	18.0
Tampa Bay (vs. CAR)	18.2
Cleveland (@ SD)	18.5
Jacksonville (@ IND)	19.2

The Colts have the highest Vegas-implied point total, though it’s important to note that Andrew Luck is questionable to play. If his replacement, Matt Hasselbeck, is called into action, the spread and total will likely fall sharply.

The Result: My Lineup for Week 4

Here’s my lineup for Week 4:

Some notes on the lineup, which includes both objective and subjective analysis:

By nearly any measure, Tyrod Taylor has been a great fantasy quarterback through three games, but he’s still reasonably priced at $7,600. Taylor sits as the fourth highest points earner. While the Bills will be without LeSean McCoy and Sammy Watkins on Sunday, McCoy has averaged only 3.4 yards per carry and Watkins sits tied for fourth on the team in receptions. That’s not to say Taylor won’t miss those weapons, but the Giants defense is enticing. They’re 10th in fantasy points allowed to quarterbacks through three games, and first in passing yards allowed (345 yards per game). The defensive line seems to be reeling without Jason Pierre-Paul, as the team only has three sacks thus far.
Frank Gore is admittedly a risk given Andrew Luck’s questionable status, particularly for touchdown upside if the Colts quarterback is unable to go. Still, he’s averaged 4.7 yards per carry, and Jacksonville’s defense has allowed the seventh most points to running backs, getting especially exposed by New England last week. Regardless of Luck’s status, I think Gore will likely figure heavily into Indianapolis game plan.
Look for LeSean McCoy’s backup, Karlos Williams, to be the most popular player in FanDuel lineups this week. He was owned by 31% of entries in my Thursday tournament contest, and it makes sense considering that he’s averaged 7.8 yards in his first 24 career carries and faces a Giants defense that is sixth in points allowed to running backs. His reasonable price puts him over the top.
The Green Bay wideout tandem of Randall Cobb and James Jones are also very popular this week after killing the Chiefs secondary on Monday night. With Davante Adams unlikely to play, Jones will be Aaron Rodgers’ No. 2 wide receiver against a 49ers team that has been the fifth worst in the NFL vs. wide receivers.
The Giants expected to have Victor Cruz back this week, but a setback puts his near future in doubt. Odell Beckham Jr. is starting to pick up where he left off last season as the ninth highest point scorer at wide receiver, and has back-to-back strong performances. Buffalo has allowed the fourth most yards in the league to wide receivers this season, and I expect the Giants to be playing catch up late in the game on the road.
Another week, another spectacular performance for a mediocre tight end against the Oakland defense. This time it was veteran Cleveland Brown Gary Barnidge, who had a career game with six receptions for 105 yards and one score. Barnidge had 48 career receptions in 94 games before Sunday’s performance. (Remember, Crockett Gillmore had a career day vs. Oakland in Week 2 with 88 yards and two scores, and Tyler Eifert had a career day in Week 1 vs. Oakland with 104 yards and two scores.) To put Oakland’s ineptitude vs. tight ends another way, tight ends that have played against Oakland this year have averaged 23.2 points in their three games vs. the Raiders, and 5.0 points in the six games they didn’t face Oakland. It’s a small sample size, but despite Jimmy Clausen playing quarterback for the Bears, I’m comfortable taking Martellus Bennett. David also selected Bennett this week in his Raiders stack based partly on this analysis.
Taking Arizona’s defense was a simple case of selecting one of the lowest Vegas-implied points teams (Rams). I had a bit of spare change left over at kicker, and Brandon McManus has 55 non-fantasy points scored in nine career home games in the kicker-friendly Denver air, and scored 18 fantasy points in Denver’s first home game this season.

That’s my lineup for Week 4. Don’t forget to compete against me in this week’s TRFC contest!