Daily Fantasy Playbook: Quarterback Strategy

Aug 01, 2014
Daily Fantasy Playbook: Quarterback Strategy

Winning in the NFL boils down to which team can execute their weekly gameplan better. The DNA for these gameplans is often established months- even years- in advance, and is all contained inside perhaps the most sacred item in the NFL: the playbook. Since knowing the playbook is a prerequisite to being able to successfully execute your team's weekly game plan, studying it during the offseason is crucial, as there will be no time during the weekly grind of the season.

Daily fantasy sports (DFS) is very similar. Each DFS site (such as Fanduel, Draftkings, Draftday, and Starstreet) has a salary cap, and your goal is to assemble the highest scoring group of players for that week only that fit within the cap. Therefore, each week you need a game plan for constructing the best lineup given the matchups and player salaries. Most of your time during the season is best spent researching players, analyzing salaries, keeping up with new developments, and ultimately crafting lineups.

But you'd better know your playbook too. Think of your DFS playbook as a collection of strategies, philosophies, and overarching concepts that you draw from throughout the season to create the best possible lineups each week and gain an edge on the field.

Throughout the preseason, I will present you with strategy guides for each position that will collectively form your DFS playbook. Then, every week throughout the season, I will provide weekly game plans for both Fanduel and Draftday.

DFS Positional Strategy Framework

When developing positional strategy, the following questions about the position must be considered:

  1. How volatile is it, and how does this dictate how much salary we should allocate towards it?
  2. Which statistic(s) are the most relevant?
  3. How do the different types of game situations influence the likelihood of scoring fantasy points?
  4. How can I extract value?

Let's kick things off by taking a look at quarterbacks.

Quarterback Salary Allocation

“Which position should I pay up for?” is a question that's been raised many times. The answer is "it depends". Your main priority should be to maximize your lineup's floor in cash games (head-to-heads and 50/50 leagues) and it's ceiling in GPPs (guaranteed prize pool tournaments). The amount of value available at each of the positions will vary on a weekly basis.

Perhaps a more significant question is “Which position should I pay up for when there is similar value at multiple positions?”

To arrive at an answer, we need to know the volatility of each position, which can be measured by taking the standard deviation in fantasy points from game to game for the players at each position.

My sample included all players who played at least six games in 2013 and met the following fantasy point per game (FP/G) cut-offs using Fanduel scoring: 10.0 for QB, 8.0 for RB/WR, and 6.0 for TE. Since positional average FP/G varies, I divided each standard deviation by it's mean to get the coefficient of variation (CV), so that we can compare volatility across positions.

Positional Volatility, 2013

MEAN FP/G

STDEV

CV (STDEV/MEAN)

QB

16.36

7.32

45%

RB

12.46

6.84

55%

WR

11.56

7.48

65%

TE

9.02

6.02

67%

Let’s assume you have a QB, RB, and WR all priced at $10,000 and projected to score 20 FP. You can calculate an approximation of each player’s floor by multiplying the CV by the projection, and then subtracting the result from the projection:

  • QB: 20 - (0.45*20) = 11
  • RB: 20 - (0.55*20) = 9
  • WR: 20 - (0.65*20) = 7

In cash games, paying up for QB is generally the optimal play.

We can calculate an approximate ceiling in the same manner, except this time we add instead of subtract:

  • QB: 20 + (0.45*20) = 29
  • RB: 20 + (0.55*20) = 31
  • WR: 20 + (0.65*20) = 33

In GPPs, saving on QB is generally the optimal play.

Remember, this strategy only works when there is similar value at the positions you are choosing between. For instance, it wouldn’t make sense to pay up for a QB instead of a WR if there were an obvious cheaper value play at QB, but no cheaper WR values. Positional scarcity and overall value available at each position must always be taken into account.

What Are The Most Relevant Statistics When Analyzing Quarterbacks?

The table below shows how various statistical categories correlated to FP/G using two different scoring systems: 4 pts per passing TD, -1 per interception, and 6 pts per passing TD, -2 per interception.

A quick refresher on correlation: The scale is -1 to 1. Positive values indicate variables move in the same direction, while negative values indicate variables move in opposite directions. An absolute value of 1 means perfect correlation. Absolute values above 0.7 are considered strong, those between 0.3 and 0.7 are moderate, and those under 0.3 are weak.

Statistical Correlations to FP/G, 2013

STAT

CORREL TO FP/G (4 TD, -1 INT)

CORREL TO FP/G (6 TD, -2 INT)

Pass TD/G

0.86

0.91

TD %

0.80

0.85

QB Rating

0.79

0.85

Pass Yds/G

0.70

0.72

Yds/Att

0.62

0.62

Comp %

0.56

0.60

Comp/G

0.52

0.54

Pass Att/G

0.35

0.37

Yds/Comp

0.23

0.21

Rush TD/G

0.14

0.06

Rush Yds/G

0.08

-0.01

aDOT

-0.09

-0.13

Rush Yds/Att

-0.17

-0.24

Sack/G

-0.25

-0.27

Int %

-0.54

-0.60

Not surprisingly, passing TDs per game were the most highly correlated statistic to FP/G. It should also be no surprise that passing TDs matter slightly more in scoring systems where they are worth six points. Touchdowns should be your main priority when selecting a QB.

Are running QBs less reliant on passing TDs? Using only QBs who rushed for over 400 yards last season, I calculated the same correlations. It turns out passing TDs have a strong 0.96 correlation with 4 points/passing TD and 0.98 with 6 points/passing TD. Rushing TDs only had a moderate correlations at 0.45 and 0.40. Passing TDs are still essential for running QBs because QB rushing TDs were low frequency events- last year, the league leader had only six. The same goes for QB rushing yards- Cam Newton's league-leading 580 were barely one tenth of Peyton Manning's league leading 5,477 passing yards.

There is only a weak correlation of passing attempts per game to FP/G. This suggests that for QBs, we should prioritize efficiency over volume. Studs like Manning and Drew Brees finished in the top eight in passing attempts per game, but so did Joe Flacco and Ryan Tannehill, who failed to crack the top 20 in FP/G.

Average Depth of Target (aDOT, courtesy of Pro Football Focus) had basically no correlation to FP/G. While selecting a QB who consistently slings the pigskin deep down field may seem advantageous, the reality is that deeper throws are more difficult to consistently execute and this type of QB is best left for GPPs. Top FP scorers such as Brees, Manning, and Aaron Rodgers finished 29th, 32nd, and 33rd in aDOT, respectively. To maximize your floor in cash games, you want passers who make high percentage throws and can consistently drive their team into scoring position.

How Can We Maximize Passing Touchdowns In Our Lineups?

Passing Statistics by Field Position, 2013

Field Position

Passing TDs

% of All TDs

TD/Pass Att

Own 1-10

2

0%

0.003

Own 1-20

13

2%

0.004

Own 21-50

58

7%

0.007

Opp 49-20

218

25%

0.04

Red Zone

574

66%

0.25

Opp 1-10

398

46%

0.34

Since attempting passes from inside the red zone astronomically increases the likelihood of passing TDs, we need to target quarterbacks who teams are likely to put them in good field position.

Many DFS players tend to target a QB on the basis of passing matchup versus the opposing defense alone. This fails to account for the possibility that if a QB’s running game or defense is overmatched, it will likely result in poor field position for his team’s offense and fewer red zone trips.

Hidden value may exist in a QB who appears to have a neutral matchup with the opposing pass defense, provided his own defense or RB has a favorable matchup. Alex Smith last year was a perfect example: He scored over 20 FP seven times by capitalizing on the dominance of Jamaal Charles and the Chiefs’ initially strong defensive play.

A shortcut to pinpointing teams that have a great all-around matchup is to check the Vegas lines for heavy favorites. Also, be sure to check out our schedule adjusted defensive fantasy points allowed (aFPA, available to premium subscribers), which are much more reliable in matchup analysis than raw, unadjusted stats.

Extracting Value From Home/Road Splits

Home/Road Splits, 2013

FP/G (4 TD, -1 INT)

TD/G

QB Rating

30+ FP Games

Brees FP/G

Home

16.05

1.75

87.5

20

29.0

Away

15.17

1.62

82.1

11

17.1

Over the course of a season, there will inevitably be instances where you have multiple QBs projected to perform similarly. In these cases, a small edge may be had by favoring the home QBs. Home QBs also tend to have more upside, as they put up nearly double the amount of 30+ FP “monster” games than road QBs did.

If a QB playing at home is coming off poor road games in the last week or two, his salary may be artificially depressed and thus exploitable for value.

I’d be remiss if I did not mention the DFS landmine known as “Brees on the road.” Brees’s performance on the road tends to suffer because the Saints offense relies heavily on frequent substitution packages and pre-snap communication, which is more difficult to execute away from the home crowd. Paying for Brees on the road would have netted you only 74 percent of his mean production, which puts you in a negative expected value situation because his salary is generally based on his cumulative performance.

Finding Hidden Value in a QB's Supporting Cast

In 2013, the Browns trotted out three below-average QBs in Brandon Weeden, Brian Hoyer, and Jason Campbell, and they were priced accordingly. Yet due to the emergence of (#Free) Josh Gordon- and to a lesser extent, Jordan Cameron- those three combined for a respectable six 18+ FP weeks. Similarly, when Josh McCown stepped in at QB for the Bears, he put up four 20+ FP games in five starts by utilizing the supreme WR combination of Brandon Marshall and Alshon Jeffery.

When a QB does not have a previous track record of success, his price tends to lag behind his true value, even when he is surrounded by playmakers that could have a direct, positive impact on his statistics.

Looking ahead to this season, the most obvious candidate to emerge and drag otherwise pedestrian quarterbacks to respectability is Cordarrelle Patterson. If Patterson develops as a receiver and can start producing long touchdowns in the passing game the same way he did in the running game and on special teams last year, whomever the Vikings start at QB may have some surprisingly valuable weeks.

Another situation to monitor is Ryan Fitzpatrick with Andre Johnson. Johnson is the best receiver Fitzpatrick has ever had, and he along with a healthy Arian Foster could boost Fitzpatrick’s numbers.

These situations don’t come along often, so it is important to jump on them before the salaries catch up. I mentioned earlier that saving at QB is generally a better strategy in GPPs, so finding cheaper quarterbacks to stack with playmaking receivers who may enhance their value is a preferred tournament stacking strategy whenever possible.

Recap

  1. Pay up for QB in cash games, save on QB in GPP.
  2. Touchdowns are the most important stat for QBs, and efficiency trumps volume.
  3. Field position is crucial for a QB to score, so target QBs on teams with not only favorable passing matchups, but favorable running and defensive matchups as well.
  4. Target QBs at home if possible.
  5. Target cheaper QBs who have playmaking receivers, especially in GPPs.

Running backs are up next. Remember to study your playbook, so you can successfully execute your daily fantasy game plans during the season and get this money!

Latest Articles
Most Popular