Lies, Damned Lies, and Fantasy Statistics - Defense vs Position

October 9, 2020

Pierre Camus dives into misleading fantasy football statistics based on small sample, schedule quirks, outlier performances, or other factors to help fantasy GMs find streamers to target or avoid and make smart lineup decisions for Week 5.

Believe it or not, there are plenty of ways to enjoy NFL football without referring to or even understanding statistics. Just ask the 76% of Americans who watch the Super Bowl each year without having watched a single NFL game the entire season prior to that day. Nearly 35% of those viewers can't name a single player on either team.

I made up those percentages, by the way. They sound believable, though. For all I know, they could even be true. That's not the point, however. Statistics can be used, misused, or downright fabricated for a variety of purposes which we won't touch upon in this space.

A one-game sample in which the running game doesn't work, the quarterback has accuracy issues, or a key injury to a wideout affects the play-calling can skew things dramatically. But what if, a quarter of the way into the season, things are still skewed? Not all advanced metrics are meaningful and some can be downright deceptive. My aim here is to point out those potential outliers that could steer you wrong when making key lineup decisions.

Now We Can't Even Trust Numbers???

Those of us who not only watch the NFL religiously but play fantasy football in highly competitive leagues will look at the statistical component of the game as essential. Some even find watching the games themselves largely useless.

Don’t watch any tape and evaluate the player using metrics only. Remove all subjectivity from your player analysis. https://t.co/9ySLcM49I0

— Addison Hayes (@amazehayes_) February 9, 2019

Addison, a former RotoBaller and very smart man, would also go on to state, "But statistics do tell the truth if you can accurately interpret the data. I think that’s one thing a lot of people hate about analytics is that I can frame three different stories from one number."

We know this to be true, yet it continues to happen because personal biases, whether intentional or subconscious, will always exist. Let's remove those biases regarding "good" or "bad" teams, and dive deeper below the surface for some real statistical analysis.

Defense vs Position Stats

One of the starting points (emphasis on starting, not end-all-be-all) for my weekly positional rankings is to analyze Defensive Ranking versus Position. This information is widely available and can show specifically where a defensive unit is vulnerable or impenetrable.

It goes without saying that Seattle is bleeding points to quarterbacks and wide receivers each week. You may not realize they've allowed 1,345 yards to WRs in four games which is 500 yards more than the next closest team (Cleveland). That means just the wide receivers on an opposing offense are averaging 336 yards per game. The worst defense in this regard for 2019 was Tampa Bay, which allowed 198.5 yards per game to WRs. So if you were streaming receivers against Tampa last year, should you be falling over yourself to grab every receiver who plays Seattle this year, including this week? More than likely, you should realize that Adam Thielen and Justin Jefferson are the only Vikings WRs with more than four catches this year. Sorry, but those Bisi Johnson and Chad Beebe shares in your DFS tourney will be wasted flex spots.

Conversely, don't assume that any defense outside of the most elite can't give up big games to quality players.

Here's a fun example: After Week 1, Jerick McKinnon was involved in the offense, scored a touchdown, and was one of the hottest waiver wire pickups I recommended. I also had the audacity to suggest he would be a smart flex play in a plus matchup against the Jets in Week 2 since, well, it's the Jets. Believe it or not, that didn't go over so well with some users on Reddit.

I received a direct line of inquiry based on this suggestion, the first comment of which read, "The narrative that the Jets are a good RB matchup is just false and undermines the analysis in this article." That's right, everything in the whole article was trash because of the failure to recognize how dominant the Jets run defense was! Better yet was the notion that nobody can run on the Jets because "thats literally the only thing that jets are good at... lol."

It's true that in Week 1 the Jets limited Buffalo to 98 total rushing yards and 57 of those came from Josh Allen scrambles. The running backs, Devin Singletary and Zack Moss, combined for 41 yards on 18 carries. The Jets, for a brief shining moment, had a top-10 defense vs the RB and I failed to recognize it. To be fair, they actually ranked eighth in limited fantasy points to RBs in 2019, so it seemed logical. This is where the statistics couldn't be counted on. Jamal Adams isn't there anymore, C.J. Mosley and Patrick Onwuasor are on IR, and Adam Gase is, for whatever reason, still the coach. It wasn't going to last.

Aside from the sweet redemption of seeing the Jet torch the Jets for 77 yards and a touchdown in that game, it proved the larger point that you can't always trust the stats alone. Here are some early rankings for defenses that I don't quite trust yet.

Washington vs RBs

The Football Team from the nation's capital isn't in a great spot, sitting at 1-3 and ready to try out Kyle Allen as the starting QB. The offense, particularly the passing game, has been the main issue but it's not as if the defense has been much better. Washington has given up at least 30 points in three straight games since their opening-week upset over Philly.

However, a glance at the ranks for Defense vs. Position reveals that Washington is allowing the 11th-fewest PPG to RB, just behind Tampa Bay and impressive units like Buffalo, Baltimore, and New England. This is where things get tricky.

image taken from RotoWire.com

In Week 1, Miles Sanders didn't play so they only had to contend with Boston Scott and Corey Clement. Holding Kenyan Drake to 86 rushing yards in Week 2 looks far less impressive now considering that is his season-high. The Browns ran all over the place in Week 3 and then Lamar Jackson had his way with the Washington Football Team Defense last week. So, while they only allowed 88 yards to RBs, they allowed 144 total rushing yards and two touchdowns on the ground. For a team constantly facing negative gamescript, it would seem that starting running backs against this unit is a no-brainer going forward. Maybe this is the week Cam Akers arrives...

Dolphins vs QBs

Unpopular opinion - the Miami defense isn't terrible. I'm not saying it's good, just not one of the worst units in the league as many think. Statistically, this opinion is hard to justify as they have allowed 27.1 fantasy PPG to quarterbacks, the fourth-highest average, 24.6 fantasy PPG to running backs, which is eighth-most, and 27.7 fantasy PPG to wide receivers, good (bad) for sixth-most. This isn't a case to excuse their defense completely but we should think twice before lumping them in with the worst of the worst.

Against quarterbacks specifically, the reason Miami has done poorly is that they've faced some of the best talent so far - Cam Newton's renaissance game in the opener, the start of Josh Allen's breakout in Week 2, and then watching Russell Wilson cook in Week 4. They were torched twice but limited Gardner Minshew to his worst game with 275 yards, zero TD and one INT. Newton had a great fantasy day but did all his damage on the ground. He only threw for 155 and no TD.

image taken from RotoWire.com

The biggest factor has been the absence of Byron Jones at cornerback, who was a key free-agent acquisition. He missed the last two games with a groin issue but should be back in Week 5. Xavien Howard remains a top-20 CB with Pro Bowl upside, as we saw in 2018. This unit will be put to the test on days where Ryan Fitzpatrick has turnover issues but they aren't an automatic defense to exploit for mediocre offenses like the Broncos and Chargers in Week 6 and 7. By the time the Jets arrive in Week 10 and 12 with the bye in between, you may even consider streaming this DST.

PPR Allowed Above Average

This is a great chance to introduce the weekly #DFBeersReport by Mike Beers (@beerswater). Now that we have a four-game sample, at least from most teams, there is a starting point to compare how teams are performing defensively relative to the offensive norms at each position.

The title is mostly self-explanatory but here's the gist of it:

What is this chart saying?

Collectively ATL/NE/DAL/MIA WRs averaged 64.6 fantasy points per game against Seattle. But in all other games they averaged just 35.8.

Similarly DAL/PHI/BUF/NYG WRs averaged 31.1 FPG against the Rams, but 45.4 in all other games. https://t.co/4fKly3cMCw

— Scott Barrett (@ScottBarrettDFB) October 6, 2020

Now, let's look into the data to see what useful information we can cull for Week 5 and beyond.

Atlanta vs WRs

By this point, it's obvious you want to start every receiver possible against Seattle. But what about Atlanta's porous and injury-riddled secondary? In trying to keep away from the debate about how Dan Quinn still inexplicably has a job, let's focus on the data.

This has a Groundhog Day kind of feel to it, as the Falcons always seem to have serious injuries in the secondary and give up a lot through the air. They allowed the sixth-most passing yards in 2018, 11th-most in 2019, and second-most so far this year at 341.5 yards per game.

We see them at the top of the list for QB and TE, so stream accordingly this year if that's how you roll at those positions. But how can they be firmly in the red at -13.3 for WR?? This is definitely misleading, seeing as how the Falcons have allowed more fantasy points to wide receivers than all but nine teams.

The key is the opponent's average combined with some unique circumstances. Atlanta has allowed five 90+ yard games to WRs already: DK Metcalf (4-95-1), Tyler Lockett (8-92-0), Amari Cooper (6-100-0), CeeDee Lamb (6-106-0), Allen Robinson II (10-123-1). Those are all fantasy studs who have put up better performances in other weeks, so compared to their averages it actually makes Atlanta's defense look less horrific.

If Davante Adams or Allen Lazard had played in Week 4, we would be staring at six or seven such games and probably four 100+ yard WR games in four weeks. Instead, Jamaal Williams collected 95 receiving yards as did a second-string TE.

It also just so happens that Atlanta has been so terrible against tight ends that it makes the wide receivers slightly less impactful in comparison. Both Dalton Schultz and Robert Tonyan performed like All-Pros when facing the Falcons. Jimmy Graham went for 60 yards and two touchdowns in Atlanta and that came with both Mitch Trubisky and Nick Foles at QB in the same game.

Both primary receivers should be ranked highly in a week where Atlanta is the opponent. Don't forget to play whoever is at tight end too.

Buffalo vs WRs

Small sample size alert: this stat is likely skewed by one specific game against a division rival. Buffalo now fancies itself an offensive juggernaut but we know the defense is the heart and soul of a Sean McDermott team and the Bills still have one of the best units in the AFC. Specifically, Tre'Davious White is one of the best in the biz and third-year man Levi Wallace grades out as a top-20 CB. Yet, they are 15th in overall team defense and allowing the eighth-most passing yards after a quarter of the season. What gives?

Part of this can be attributed to the high-octane offense that has made it necessary for opponents to pass more often to play catch up. Defensively, the Bills face the sixth-highest pace of play at 25.46 seconds between snaps. That puts pressure on the defense and forces them to face more situations where an opponent will eat up yardage. Whether that remains the case depends on how much you believe in Josh Allen's early-season performance.

The most important reason is the schedule, however. During their 4-0 start, Buffalo has had the pleasure of facing the Jets, Dolphins, Rams, and Raiders. The Rams are legitimately good, as are their receivers. The Dolphins are legitimately not, yet the combo of DeVante Parker, Preston Williams, Isaiah Ford, and Jakeem Grant combined for 157 receiving yards in Week 2. That's not overwhelming but it's more than that WR corps achieved in Week 1 (126) or Week 3 (109). They then faced Seattle in Week 4, so you know how that turned out...

Similarly, the 133 yards gained by Raiders receivers against the Bills isn't much but it's pretty good for that bunch considering they don't have a single WR who's reached 200 total yards and only two who've reached the 100-yard mark after four games (Hunter Renfrow and Nelson Agholor).

This Buffalo secondary is good, their linebackers are back healthy, and the game flow should slow down enough eventually against tougher opponents coming up on the schedule. You aren't playing Tennessee receivers this week for other reasons but don't feel too confident streaming the likes of Sammy Watkins, N'Keal Harry or whoever the Jets trot out there in the coming weeks.

More Fantasy Football Analysis