When trying to decide who to play as your flex, there are two basic factors to consider: the candidate player's ability and his matchup. Late last season, I did a series of articles on who to start at RB, WR, and TE that looked at quantifying those factors. You might want to look over those articles first, but I'll try to make this one stand on its own.

I'm going to use PPR scoring and only compare RBs to WRs as flex choices. The data is based on the past ten seasons, 2006-2015 (updated from the articles linked above) and the Top 100 RBs and Top 100 WRs in those years. So that's a total of 1000 players at each position. “RB1s” were the top 100 overall, “RB2s” were ranked RB101-200, etc. On a seasonal basis, the would be the Top 10 RBs, RB11-20, etc. To further differentiate the “Studs” from the rest of the RB1s, I broke the top 10% in half, so “Stud RBs” are the top 5% or top 5 backs in a year, with the extra nuance that if a back in the Top 50 from the last 10 seasons was not in the Top 5 of his year group then he counted as an “Other RB1” and not a “Stud.” I did the same thing with WRs.

I then ranked all the defenses over the last ten seasons by fantasy points allowed (RB or WR as appropriate). I divided them by quarters – the top 80 in that time span, the next 80, etc. I separated out the top 5% as "stud" defenses from the top quarter (this would be the top one or two defenses in a given season). I looked at dividing the defenses up into smaller groups, but I felt this was putting too fine a point on my ability to estimate a defense’s category – and it really didn’t make any difference in the conclusions.

Then I calculated the average FP for each set of matchups, for example, Stud RBs against Stud Defenses.

Here’s the results, RBs first:

Average PPR Points of RBs vs. Defenses, 2006-2015

RB Category

Defense Rank

Top 5%

Other Top Qtr

2nd Qtr

3rd Qtr

Bottom Qtr

Stud RBs

16.2

18.2

19.6

21.0

23.8

Other RB1s

9.8

13.6

15.0

16.7

19.1

RB2

9.8

11.8

13.5

14.8

15.9

RB3

8.7

9.4

10.3

11.4

12.7

RB4

6.0

8.2

8.9

9.4

11.8

RB5

5.2

6.8

6.9

8.0

9.4

RB6

4.1

5.1

5.7

6.4

7.0

RB7

4.1

4.3

4.8

5.3

5.8

RB8

2.8

3.2

3.5

4.3

5.1

RB9

1.8

2.0

2.7

3.0

3.3

RB10

1.4

1.6

2.0

1.9

2.6

For example, Stud RBs averaged 16.2 against Stud Defenses, but 18.2 FP against other top quarter defenses. As you’d expect, they progressively averaged more FP as the defenses they played got worse.

The same is true for other RBs, almost without exception. I'll come back to the highlighted cell.

Here's the WR table:

Average PPR Points of WRs vs. Defenses, 2006-2015

WR Category

Defense Rank

Top 5%

Other Top Qtr

2nd Qtr

3rd Qtr

Bottom Qtr

Stud WRs

15.6

17.1

17.5

20.7

23.6

Other WR1s

13.8

15.6

16.8

17.6

18.8

WR2

10.1

13.3

14.4

14.8

15.8

WR3

9.5

10.6

11.7

12.4

13.6

WR4

6.7

9.3

10.6

10.5

12.4

WR5

6.9

8.2

8.9

9.4

10.1

WR6

6.1

7.0

7.4

8.8

8.6

WR7

4.9

6.2

6.8

7.1

7.6

WR8

5.0

4.9

5.9

5.6

6.8

WR9

3.6

4.0

4.5

5.0

5.7

WR10

2.3

3.0

3.3

4.1

4.7

Now you can look at each table individually and make start-sit decisions at each position (look back at the old articles for help on that although I think it's probably fairly self-explanatory). But what I want to do is use the two tables to help with flex decisions.

Let's say you're in a league that starts 2 RBs, 2 WRs, and one RB/WR flex. You plug in your starters at RB and WR. Now you're left with a couple of RBs and a couple of WRs on your bench.

The first thing to do is categorize their ability as RB3, WR2, etc. based on their overall merit (not on their position on your roster, but as fantasy players in the league as a whole). Then you need to look at their matchup for the week – taking the opposing defense and categorizing it as a stud defense, top quarter D, 2nd quarter D, etc. Remember to evaluate the defense against the appropriate position: it might be really good against RBs and porous vs. WRs.

Let's say you have the following:

Player

Matchup

RB2

Stud

RB4

3rd Qtr

WR2

1st Qtr

WR3

2nd Qtr

You've got a pretty good team, your 3rd best RB grades as a RB2 overall; your 3rd best WR is a WR2 and your #4 WR is a WR3. Then you go into the tables to see what the average result is for each matchup - I'll call that expected fantasy points (FP). For example, your (flex available) RB2 can expect to score 9.8 FP in his matchup. And your WR2 would typically average 13.3 (1st Qtr matchup = "Other Top Qtr" in the table). See the highlighted cells in the tables above to find these numbers.

Doing that for all four players gives these numbers:

Player

Matchup

Expected FP

RB2

Stud

9.8

RB4

3rd Qtr

9.4

WR2

1st Qtr

13.3

WR3

2nd Qtr

11.7

It looks like you'd start that WR2 at your flex, despite his tough opponent.

Now let's take real players and defenses this year. Let's say my flex choices are:

Player

Wk 5 Opp

Category

Matchup

Expected FP

Duke Johnson

NE

     

Giovani Bernard

DAL

 

 

 

Randall Cobb

NYG

     

John Brown

SF

 

 

 

Now I need to assign the category and matchup grades. For the players, I can use the site's season projections, which have Johnson and Bernard as roughly RB3s and Cobb and Brown as WR3s. I can downgrade Brown for this week if I think he's worse than a WR3 with Drew Stanton instead of Carson Palmer at QB. I can upgrade Cobb if I think he should be a WR2. I can upgrade Johnson if I think his team will have to throw a lot against NE or move Bernard down if I think the Bengals will use Jeremy Hill to kill the clock and take touches from Bernhard.

On the matchup side, Dallas is currently 25th in FP allowed (19.18 FP/G) to RBs and NE is 13th (24.25 FP/G). Going strictly by that, DAL is a 2nd qtr defense (fewer points allowed means a better defense) and NE is a 3rd qtr. But both these RBs are pass-catching options for their teams. NE is allowing the 8th most points to RBs when looking just at the passing game, while Dallas is steady at 25th Maybe I should make NE a bottom qtr defense for this analysis. SF is middle of the pack in PA to WR while the Giants are 13th. I could upgrade SF thinking they will be better against a crummy QB – but I might be "double-counting" Palmer being out (you may think that is warranted).

So I made NE a bottom quarter defense against pass-catching RBs, downgraded Brown to WR4 and rounded SF up into the 2nd quarter of defenses vs. WRs, then looked into the tables for the Expected FPs:

Player

Wk 5 Opp

Category

Matchup

Expected FP

Duke Johnson

NE

RB3

Bot Qtr

12.7

Giovani Bernard

DAL

RB3

2nd Qtr

10.3

Randall Cobb

NYG

WR3

3rd Qtr

12.4

John Brown

SF

WR4

2nd Qtr

10.6

Based on this analysis, Johnson edges out Cobb for my flex choice. However, if I treat NE as a 3rd qtr defense that drops Johnson's expectation to 11.4 and moves Cobb into the starting flex role. Or if I think Cobb's offense coming out of a bye is going to use him better, and I bump him to a WR2, that would make his expectation 14.8 FP. Based on the combination of Duke's downside, and Cobb's upside, my gut tells me to start Cobb.

The point is not to make it all numbers or take away your subjective analysis. The point is to help you quantify how much difference it makes how you value a player or how tough you think his matchup is. I think you are better off with a structured way to make decisions rather than going purely by the seat of your pants.