Daily Fantasy Sports
Sharp App
Season Long Fantasy
Free Month
Wall of Champions
Discord

NFL DFS Week 3 Correlation Matrix Leverage Plays

Hey guys, this is going to be a very simple extrapolation using our Correlation Matrix grid as an illustrative point in spotlighting potential GPP winning plays for NFL Week 3.  The grid itself is fairly easy to follow, if you haven’t dabbled with it yet.  As the weeks go on, it will become more and more matchup-based and even more accurate in terms of spotting which players are most positively correlated with which players.

Next week, we might see CeeGee take over this column and be our regular Matrix correspondent.  For now, though, I get to put my little twist on it.  For those I haven’t met, I’m @Choppodong in Slack and on Twitter, and I’m one of our Senior Writers and current Content Editor.  I’ve been in the DFS Army for over three years and am one of the “Original 6” members of our community.  I specialize in Bankroll Management, but am more of a Jack-Of-All-Trades….Master-of-None.

Captain Obvious…

The most common correlation play is obviously the QB/WR1 stack.  Take a quick peek at the Top 15 plays on the week based purely by the numbers.

Many won’t obviously be using Eli/OBJ, but this graphic points out how strong this play can be this week.  Saquon Barkley plays a huge role in the passing game, too, and is actually 2nd on this team with 22 targets of his own.  Eli doesn’t look to his other receivers often, so it makes sense OBJ gets the lion’s share in a game the Giants are set to trail.  O’Dell is perhaps the best contrarian GPP wide receiver on the slate with all the focus going to New Orleans, Pittsburgh (on it’s slate), Atlanta, Kansas City, etc up in the higher priced receiving tiers.

Another interesting note to point out is the Steelers on this list twice.  You have to play the Sun-Mon or uber short Showdowns/Primetime slates to get them, but look at that correlation!  So strong for both Antonio Brown and Juju Smith-Schuster.  Let’s not overlook the connection/correlation shown between Ryan Fitzpatrick and Mike Evans in this same game, either.  You have three of the top 15 pairings in the same game…….we should see some fireworks at higher ownership in that Monday Night burner. 

Where to Go From Here…

On the first tab of our correlations matrix page, you can see all these top pairings.  But, let’s dig a little deeper to some harder to see matchups that correlate well.  We won’t likely focus on terrible games with super low team totals like the Jacksonville/Tennessee or Chicago/Arizona.  But, we can easily find some hidden gems using these matrix graphics…

SF49ers @ KC Chiefs

Everyone will want a piece of this game.  Mahomes/Hill and Mahomes/Kelce are pretty obvious and correlated very well.  However, did you notice the second best correlation on this team is NOT with Tyreke Hill (.28).  It’s with Sammy Watkins (.30).  Sammy has only 2 fewer targets at 12 (14 for The Freak) and makes a sneaky under the radar play in a chalky game this week.  Our ownership projections (found in Domination Station) have Sammy projected at under 10% in a very highly owned game on both sides.  If you want a piece of the game no one else has, Sammy might be your guy.

New Orleans @ Atlanta

Again, we know where the chalk is.  We even know Brees/Kamara (.24) shows a strong correlation (David Johnson is .30 with Bradford and CMC is .23 with Cam Newton, for reference).  But, after all the big dogs, we still have some strong correlations.  Take a look at Ryan/Calvin Ridley (.29) or even Ryan/Austin Hooper (.30).  Even Brees/Ginn (.31) rings in at a more than respectable number.  Of course, we need to understand the longer shot natures of these plays before just plugging and playing them.  Both offenses lean VERY heavily right now on their RBs and primary WR1s for workload.  It’s not like a New England Patriots system that is designed to spread the ball all over.  “Strong correlation” in this example should be used with the context of it still being very “gpp only” in nature due to lower target volume.  Ridley hasn’t seen 10 targets yet all year, and Hooper has only seen 9, as well.  These are highly correlated plays, but still very much touchdown dependent and not for the feint of heart.  They will, however, be lower owned in another chalky game…..which is the point.

Next Level Matrix Usage (Make Neo Proud)…

Now, you you have some of the basics down, let’s ramp this tool up a notch.  For this, we need to revisit the Transitive Property from our early math days.  (If a=b and b=c, then a=c….to jog your memory)

Again, everyone knows how strong the MIN defense play is this week due to the Vegas 17 point spread and paltry Buffalo implied team total of 12.  But, just how strong is it?  Let’s look at the matrix to see how the MIN defense correlates with Josh Allen, the BUF QB.

That -.57 points out that we are beyond stupid if we use a stack containing Josh Allen/MIN DST, right?  I mean like dumber than Jim Carey riding a frozen scooter with Jeff Daniels dumb….

 

However, if we use another great math principle, we can take the “reciprocal” of that -.57 and extrapolate just how strong it must be if we look at it from the MIN side instead of the Josh Allen side.  Here is where the Transitive Property steps in:  It makes incredible sense to see that if the ratio of QB to Opposing Defense correlation is horrible, then the Defense to Opposing QB correlation must be fantastic.

Many of you are only looking through the matrix looking for the green numbers.  However, complete use of the tool requires us to look at the red numbers, too, but think of how strongly correlated we are to the reverse situation.  This tool requires thinking through if you really want to maximize its capabilities for your game.

Finding Defenses…

We can now use the matrix to go searching for the strongest correlated plays on the slate in terms of reversing the QB/Opp D stack.  Like turning a weakness into a strength, if you will.   Here are your top 5 plays on the Main Slate for Week 3.

We have what we expect, largely, except for #1 (a neat little GPP twist, possibly).  Philly is always a gpp-relevant defense, but who knew they might be the top spot this week?  The matrix knew….that’s who.  And, #5 in Seattle is a bit sneaky, and might be indicative of the entire matrix using historical data so early in the season still.  As stated above, we expect these numbers to grow in accuracy as the year progresses.

PHI – .59
MIN – .57
JAX – .56
BAL – .56
SEA – .56

Other Notables We Thought Were Strong…

LAR – .43
HOU – .51

The Rams will appear like a decent play, and a lot of people plan to go “back to the well” with the Texans.  Buyers beware!

Invest the Time…

I will readily admit the Correlation Matrix (tutorial here) is one of the most complex, and potentially useful, tools we at DFS Army have ever provided our members.  It is complex, it can be confusing, and it definitely sucks up a lot of time.  However, don’t let that frustrate you to the point you start ignoring it.  Investing the time to get comfortable with the numbers and grids will make life easier when you can simply scan it over a few minutes, looking for both strong green and strong red, looking for standout matchups.  Like everything in our community, you will get back exactly what you put in when it comes to effort.  If this DFS world was easy, everyone would be a pro.  How bad do you want it?

As luck would have it, you also possess the one thing the rest of the industry does not have access to………….coaching.  We are here to answer questions.  We learn from you, too, and often have to add layers to our own processes to better explain things when asked.  You are a part of the greatest community DFS has ever created.  And, I’d like to see you all take full advantage of everything at your disposal from content to tools to coaching.

Let’s crush Week 3 and post another Wall of Wins historical gpp beatdown!!

Yours in DFS,
Chop