Let’s Stream Defenses: Week 9 (Fantasy Football)
Theoretically, the goal of fantasy scoring is to align real-world performance with fantasy points, so managing your fantasy team is something like evaluating talent (and opportunity). Unfortunately, the way defenses are scored does a pretty garbage job of this. 17 points allowed is an excellent real-world performance, and would give any competent offense a chance to win. If you do that with no turnovers and 3 sacks (the league average is about 2.5/game), that’s only 4 fantasy points. On the other hand, to allow 30 points with a pick-six and 2 sacks would earn 9 fantasy points. Because fantasy scoring is so focused on sacks, turnovers and touchdowns, the following facts are important when choosing a defense:
- Sacks can’t occur on hand-offs.
- About 35% of fumbles occur on sacks.
- About 64% of turnovers are interceptions. Combined with sack fumbles, this means that at least 80% of turnovers occur on passing plays. (This number doesn’t include fumbles after the catch by receivers.)
- For both sacks and turnovers, the tendencies of the offense are much more predictive than that of the defense.
This all adds up to the fact that you really want your defense to be facing a team that passes a lot. Simply targeting the teams that pass a lot would be a bad idea with no other factors – Green Bay passes at the 2nd-highest rate in the league. We want to pick teams that are going to be forced to throw because they’ll be way behind, not because they’re good at it. Predicting who’s going to be behind is kind of dangerous – it’s easy to get burnt playing a marginal running back on a team you think is going to win easily, but ends up playing from behind. Still, you’re never going to have the same confidence in your defense that you would in a good running back. No defense is a guaranteed 7 points – we have to pick the teams that have a chance of the game really going their way, and a low likelihood of melting down.
In real football terms, the Kansas City Chiefs are a below average defense, according to Football Outsiders. They’re close to average against the pass, but worse against the run. They do, however, have the best offense in the league. Their fantasy defense has feasted in 3 of the last 4 weeks against offenses that are prone to make mistakes when trying to catch up. (The other week was Tom Brady.)
I’ll take an average defense attached to a powerhouse offense (like Kansas City) against an error-prone opponent (like Cleveland) over a good defense in a bad matchup (like Baltimore vs Pittsburgh) any day.
Show Me Your Ranks
Fortunately, there are a few low-owned teams in the top two tiers. Unfortunately, the list of teams I would be comfortable starting this week is very short. Vegas is projecting this to be the highest-scoring week in an already historic season. Only two teams have implied totals under 20 points, while we can usually fill out a good portion of our top 10 with such teams.
|The Start Them With Confidence Tier|
|The Still a Fine Choice Tier|
|The Maybe You Can’t Find Something Better Tier|
|The Surely You Can Find Something Better Tier|
Tell Me About Your Top Picks
- CHI @ BUF: Some of the information I use to make my projections isn’t available until after the Monday night game. Not only do the stats from that game play a part, but sportsbooks will sometimes wait until Tuesday to set lines for the following week’s games involving teams that played on Monday night. In this situation, I take a guess at what the line will be, so that I can generate preliminary projections on Monday. For Chicago at Buffalo, I guessed that Chicago would be favored by 6 points, with an over-under of 43.5. That was an implied total of 18.75 for Buffalo, which was already enough to rank Chicago as the top defense this week. That turned out to be the most wrong I’ve been about a line this season – the total is much lower at 37.5, and Chicago is even more heavily favored. If you’ve been holding on to Chicago, expect to be handsomely rewarded this week.
- KC @ CLE: This is a classic example of where a team can benefit from the disconnect between fantasy scoring and real football performance. It’s certainly possible that Cleveland’s defense rises to the occasion and keeps Patrick Mahomes in check, but I’m happy to bet on Cleveland playing from behind and passing a lot, opening the door for sacks and interceptions.
- DAL vs TEN: The Cowboys’ defense has quietly had an excellent year, allowing an average of under 17 points per game. In their two outliers – the games where they allowed over 20 points – they’ve only given up 24. The Cowboys’ schedule has been relatively soft, but that’s not changing this week against a Titans offense that has only exceeded 20 points once.
- DEN vs HOU: The model loves this game for Denver at home. Deshaun Watson has made some mistakes this year, but Denver is knocked out of the top tier by the fact that Watson was lights-out last week, and the risk that he does it again.
- MIA vs NYJ: The Jets might be the best rookie-led offense, but they’re just not scary. Miami’s defense has been up an down this year, but should be in a position to do well at home this week, and Vegas agrees, favoring them by a field goal with a relatively low point total.
- CAR vs TB: It’s probably not a good thing for Carolina that Ryan Fitzpatrick was named the starter again, but there’s a reason he’s plan B. The wheels will fall off sooner or later, and the Panthers at home are good enough to make it sooner.
- SF vs OAK: There’s a strong chance that this is the worst Thursday Night Football game of the season. These are two of the worst teams in football, but only Oakland is terrible on purpose, and they happen to be playing on the road.
- BAL vs PIT: This is a Vegas line that surprised me, giving the Steelers just 22 points. Still, Baltimore’s defense has been excellent, and Ben Roethlisberger’s home-road splits are well documented, even if I have doubts about how predictive we should expect them to be. If you want to hold on to Baltimore this week, I’m uneasy about it, but you can do a whole lot worse.
- NE vs GB: This is sort of a last-resort starter. New England was a popular streamer last week, so if you don’t already have them, there’s a good chance they’ll become available when waivers clear this week. The Packers are on the road and have allowed an above average 3 sacks/game, despite how much time Rodgers seems to have when he’s not getting sacked.
- MIN vs DET: This is another one that I’m not super comfortable starting, but if you have the Vikings, you’re probably not all that interested in dropping them for a streamer. It should be fine this week, especially if the Lions trade one of their three stud receivers.
How Did We Do Last Week?
Our top streamer, New England, was excellent. Houston almost was – they had a fumble sack inexplicably overturned. The officials seemed to forget that getting the ball knocked out of your hand backward isn’t a forward pass. The rest of our recommended starters were fine, though the model picked the wrong side of the Giants-Washington game. Unfortunately, two teams that scored defensive touchdowns weren’t ranked as starters – Cincinnati was just outside, and New Orleans was ranked 23rd. Seattle also did well with 3 turnovers and only 14 points allowed, at a rank of 19. Both tiers of starters averaged 8 fantasy points, while teams not recommended as starters averaged 4.
|Rank||Team||Opponent||O/U||Spread||Proj. FPTS||Act. FPTS|
|The Start Them With Confidence Tier|
|The Still a Fine Choice Tier|
Jacob Herlin is a Senior Data Analyst for FantasyPros. For more from Jacob, follow him @jacoblawherlin.