I am a creature of habit. I like predictability. I enjoy a good routine. I like waking up knowing that the schedule I’ve kept for the last two years is the same schedule I will keep today. I will wake up at 4:30 AM, have my pre-workout drink, thumb through Twitter, work out to some sort of fantasy sports or fitness related podcast, get myself ready for work, take my son to school, go to work for nine hours, come home and make dinner, get ready for bed, and fall asleep by 8:30 PM ready to do it all again tomorrow. Perhaps it’s the predictability of my life that draws me to NFL DFS.
However, I also like simplicity as much as I enjoy predictability. If my routine takes more work than deemed feasible or necessary, it probably means I’m not enjoying my schedule nearly as much. The simpler I can keep my routine, the easier it is to keep and the more time I have to enjoy doing the other things that comprise my life.
Outside of the pre-season, NFL DFS is the one true sport where every factor that determined a great lineup in week one will be true ten weeks later and even in the playoffs. Because of the short 17 week schedule you don’t have to worry about resting players destroying a locked lineup unlike the NBA or any of the various soccer leagues. Just because a player transitions from one stadium to the next doesn’t downgrade his fantasy potential unlike MLB where pitchers/ hitters can lose upside simply because of outfield dimensions or in golf where a golfer’s playing style isn’t suited for the layout of a course. The field that a football player caught a touchdown on last week will be the exact same measurements as the field he’s playing at the following week. Outside of the occasional windy game, and there’s even debate about how much steady 20+ MPH winds really do affect passing attacks, the factors that go into selecting each positional player will remain constant throughout the season.
The purpose of this article is to explain in context what these factors are and why they matter so much, yet also to keep the details to a minimum so as this process is clear and easy to replicate. Many daily fantasy sites pride themselves in player models that factor in twenty plus trends, stats, and rankings so as to give you the consumer a wide view of what makes a player a good roster decision. However, I have found beauty in simplicity and have realized after eight years of playing fantasy sports that their is a fine line between having enough information to make an informed decision and so much information that your judgment gets clouded. Practically all of my player positions breakdown to seven key factors and no more. Were I to start including more information my player pool would start expanding as outliers get enough of a bump to get considered, and now my upside is capped by including these low upside plays. By narrowing my pool of players down to seven key components I’m ensuring that only the best of the best are getting nods towards being in my lineups come the weekend.
These are the same factors that I use to build my weekly spreadsheet that outline, what I believe are, the best plays at each position. Once I have the top statistical plays the only thing left is to make these pieces fit, salary cap wise, into an ever changing puzzle that evolves based on player’s salaries rising or dropping.
However, before I begin talking about these ever important statistics I want you to know where and how I source this information.The purpose behind using these statistics will be explained later in the positional breakdowns, but for now I think it’s paramount you know where I access the numbers that I use. To save yourself time each week I’d be creating a folder in your bookmarks tab and saving each of these sites…
Vegas Odds: I get my Vegas metrics from two sites; Sharp Football and Vegas Insider. Instead of viewing what the current betting action is at a single casino or off shore book, both sites allow you to see multiple casinos/ books and gauge their action against one another. I prefer to use both sites because sometimes the books that Sharp references won’t have odds out as early as what Vegas Insider uses. That being said, if you were to only use one I would prefer Sharp. Sharp has a great feature that uses a color graph to show how long it has been since the money line or odds went down or up. This is especially helpful if you like to follow the big money that shows up on Sunday mornings, three to four hours before kick off, so you can gauge which direction the sharps are thinking.
Team/ Individual Statistics: The majority of my statistical research begins with Team Rankings. As soon as Monday Night Football is over, it’s a safe bet that Team Rankings has updated their database giving you pertinent information that you can compare for the season, last three games, or even to last year if so needed. Nearly 90% of my research is done there, all for free, for every position whether I’m looking to cross compare to another player or versus certain teams. Whatever information Team Rankings doesn’t have I check in with the team at Draftshot. We all bounce ideas and stats off of one another to see what sticks.
Rankings: Once again, if it’s just simple, quantifiable stats like where a defense ranks versus the run then Team Rankings is the place to find this information. However, sometimes you need rankings that a simple stat can’t explain. That’s where Fantasy Outsiders steps in and offers their team defense DVOA metric rankings that go above and beyond what a statistic like passing or rushing yards allowed per game could tell us. FO’s DVOA breaks down by position (QB, RB, WR, TE) and where on the field certain defenses excel and where they crumble. These DVOA metrics have been fairly spot on and help you see exploitable matchups that the public, who is just clicking on player cards, won’t see. On the flip side, Football Outsiders also offers offense DVOA rankings to help you see trends for the players or defenses you were planning on using.
In regards to gathering this data, this easiest method is also the most time-consuming. That involves you visiting these pages and directly copying the information into a spreadsheet or notes. If you’re into time management, while being handy with the basics of MS Excel, I encourage you to start using Google Sheets and create spreadsheets where you’re data mining this information directly from the source. Draftshot also has pages and pages of stats and projections you can sort through to start your research.
One final thing of note; no where do I take in consideration what a player is producing themselves as a stand alone stat to consider. Every week is a new matchup that demands we investigate where the information is pointing. What a quarterback or tight end did or has done is noise. What I want to know is what they’re capable of based on this week’s matchup. Momentum is something I refuse to try and formulate. When I do look at a player’s production, it’s measured in opportunities to touch the ball afforded by his team. End product numbers can be deceiving, I want to know how he’s reaching those numbers dependent on how many times he touches the ball.
Now that we’ve addressed where to access this information let’s go position by position and see what exact information you need for building the most data congruent roster possible in NFL DFS.
Quarterback is where the majority of DFS players begin their lineup building process, and for good reason as it’s the default first option at every daily fantasy site. Now, while everyone may begin building teams at the quarterback spot, it doesn’t mean all are looking at signal callers through the same lenses. The temptation with quarterbacks is to over analyze the position simply because they will have the ball on nearly every offensive possession (quarterbacks will account for ~75% of their team’s total touchdowns). Thus, how do you know which stats are giving you clarity apart from the ones that are simply adding noise? For me, it’s finding which metrics correlate strongest to the way a quarterback can accrue fantasy points. Obviously, fantasy points are occurring more often as more real life points are being scored. Thus, my search for a quarterback begins with Las Vegas.
A quarterback should be a home favorite. Despite what you may assume, a QB who is a dog is not where you want to begin looking. When a QB is an assumed dog, come the fourth quarter, it means his play + his offensive personnel’s + the opposing defense have combined to create an effort where his production hasn’t been good enough to be winning. You obviously want the QB on the other end of that equation who has created numbers that have put his team in a winning position. Furthermore, we’re looking for a home team because they’re more likely to score more points than away from home. On average, an NFL home team scored 24.14 points while when on the road only scoring 21.43 points. In fact, only 6 of 32 NFL teams averaged more touchdowns on the road than at home.
In 2015, average fantasy points for quarterbacks (Draftkings scoring) broke down this way per $1,000 of salary…
|Vegas scenario||Home favorite||Home underdog||Road favorite||Road underdog|
|Points per $1K||3.35||3.06||3.01||3.00|
Interestingly enough, the numbers point to quarterbacks having more potential value as a home underdog than being a road favorite, but I believe that makes the point even clearer that being at home is a definitive value for quarterbacks as opposed to being on the road.
A quarterback’s team should have an implied team total of 24 points or more. On the surface this seems simple enough because if a team is scoring 24 points then it means that roughly three touchdowns were scored. However, when you compare the production of quarterbacks compared to their implied team total (Draftkings fantasy points) it becomes even clearer…
|Vegas scenario||18 and under||18-21||21-24||24-27||27 or more|
|Points per $1K||2.65||3.09||3.16||3.28||3.11|
The more points a quarterback’s team was pegged to score directly correlated to his own fantasy output up until a team reached 27 or more points. The slight drop after 27 points could be explained by a team, with that large of a score, leaning more on the running game as they’ve build a presumptive lead. The other option is that daily fantasy sites do bake Vegas figures into their salaries, and a quarterback with a team implied to score that much may end up with a higher salary than 95% of the rest of the quarterback player pool. Thus, he may be hitting the same fantasy points marks as a cheaper quarterback, but because of his higher salary he’s not nearly the same value. Regardless, the benchmark is 24 points and this will also spill over into other positions as the sweet spot for implied team totals. As far as Vegas is concerned, I also want to target quarterbacks playing in a game with an over/ under that is 46 points or higher. I’m not overly concerned about the actual spread because my interest is what Vegas believes each quarterback’s team to score and not what their opponents do. However, I don’t want to worry about the opposing team being enough of a liability that they may not keep the game competitive. While it’s not dogmatic to be that my list of quarterbacks all come from games with higher over/ under, it does add an extra layer of security about the type of game script the quarterback should see.
If a Quarterback can hit those two previous factors than I have a great place to start, especially in my cash games where I want safety. Obviously, there will be Sundays where Quarterbacks are in great positions but are on the road as a road favorite. In those cases, they will still probably rank high in my model, but for the sake of consistency, I will always side with the home favorite because of their projected output. Outside of what information is pouring out of Vegas, I have four simple stats that further point to fantasy production.
A quarterback should have a match up against a defense that is ranked bottom twelve in fantasy points allowed. Instead of having separate metrics for defenses that are giving up the most passing yards, most rushing yards to quarterbacks, most passing touchdowns, and most total touchdowns I find it helpful to just simplify everything into one encompassing stat which is fantasy points allowed. While there might be a few outliers skewing numbers the end result (total fantasy points) is the end result. Ideally, you’ll want averages based on anywhere from four to eight games so as to have a more accurate picture of how defenses are playing whereas season long rankings may not tell the whole story.
A quarterback should have a matchup against a defense ranked bottom twelve in completion percentage allowed. The common myth for fantasy quarterbacks is that volume is what makes them great. Well, unlike our other skill positions where volume does matter, just because the quarterback is throwing the ball more than another quarterback doesn’t automatically make them better. In fact, if a quarterback is throwing more it may mean his team is losing and his stat line may be the reason behind that losing effort. Regardless, we prefer efficiency to volume because it means that passes are being completed which in turn means drives are being extended and yards are being accumulated. The more likely a drive is to be extended down the field, the higher likelihood the drive results in a touchdown.
A quarterback should have a matchup against a defense ranked bottom twelve in takeaways. Probably more so than incompletions, turnovers are a killer for the fantasy potential of quarterbacks. Once the opposing defense has recovered a fumble or picked off an errant pass, the quarterback waits at the mercy of the opposing offense until he can take the field again. Thus, we want to target scenarios where of a quarterback getting intercepted is lower. Besides the negative points that these turnovers accrue, it also means drives get killed. While I have no intention to quantify momentum for an offense, the mere fact that the quarterback no longer has the ball remains.
You could literally add tons of more metrics, and feel free to if you so desire, but I feel keeping the model limited to six metrics for a quarterback means I’m not adding too much noise. If you consider too many factors then borderline quarterbacks find their way into your player pool and your upside starts falling as the quality of player decreases. Ideally, the fantasy quarterback you roster will have a high implied Vegas total as a home favorite with a matchup he can exploit for maximum fantasy points output. Anything beyond that may be trying to out think your process.
Before you start comparing apples to oranges, let me stop you from assuming that the out of the box approach to quarterbacks, efficiency over volume, is how you should approach the rest of the positional players in daily fantasy. Probably more so than any other position, there exists a strong correlation between the volume a running back sees and his statistical output. While it should matter to an NFL talking head what the yards per carry average a starting running back is carrying, his volume, or lack thereof, means very little to us. For example, if a running back is averaging five plus yards per carry that look amazing, however, if that tail back is seeing only ten carries per game (think of a committee approach like the Philadelphia Eagles in 2016) that’s not the type of running back we want to target. Average that out and this running back is seeing only two to three rushing attempts per quarter. An average that says he’s not on the field nearly enough to be a regular contributor for his team. On the flip side, if a running back is sitting around the Mendoza line for ball carriers (three yards per carry), but he consistently gets twenty to twenty-five touches per game that is volume that we can trust from a player his team trusts to tote the rock. Beyond volume, which I will dig a little deeper into, I also tip my hat to Vegas once again looking for prospective game scripts that foreshadow which running backs will be in line for larger fantasy point producing days. These two criteria are the most predictive, in terms of fantasy production, for running backs and it’s essential any running back in my player pool meet these standards.
A running back should be a home favorite. If you’ll remember from the quarterback section, fantasy points were directly tied to their Vegas scenario with the highest outcome being a signal caller who was favored at home. In the same breath, in fantasy points (Draftkings) per $1000 of salary, running backs who were home favorites were scoring the most.
|Vegas scenario||Home favorite||Road Underdog||Home Underdog||Road Favorite|
|Points per $1K||2.91||2.60||2.46||2.39|
Running backs who were home favorites were averaging nearly three points per $1000 of cost on Draftkings, while the next closest scenario of a road underdog was more than .3 points behind. If you’re trying to contemplate why a road and home underdog were more valuable than a favorite on the road, don’t forget that Draftkings awards a full point per reception making it easier for a pass catching tail back to accumulate fantasy points (one point plus a tenth of a point for every yard he runs for after the catch) than a running back who is probably grinding out game clock getting only .1 point for every yard he runs for. Regardless, the game script scenario of a home favorite is the most valuable hands down as it ensures nearly everything we want to see our running backs have; touch opportunities, yards, and touchdowns. From 2013-2015 running backs who were home favorites saw the most rushing attempts per game (27.9), rushing yards (118.5), and rushing touchdowns (0.9) per game as well.
As a side point, the larger the favorite the better. Four points seem to be a happy medium because four implies late in the game that the lead is large enough that the player’s team will have the ball and being able to utilize him in the running game to help melt away the clock. More carries equal more fantasy point opportunities. Also, the larger the implied total, the quicker that team is using the running back to salt that game away because, more often than not, they’ve reached that lead by the third quarter instead of with seconds remaining in the fourth. In 2015, when a team was tied with their opponents they were rushing the ball 43.3% of the time. However, when a team had a lead it was rushing the ball 50.2% while that number dropped to only 33% when trailing. The more positive the game script is in the running back’s favor the more likely he is to see carries.
A running back should expect to touch the ball a minimum of eighteen times per game. Eighteen may seem a bit arbitrary, especially as even the Lions who ranked dead last in 2016 in attempts per game averaged 21.5, but eighteen ensures that the running back is seeing the bulk of his teams carries. Also, this reduces risk so as to not roster chunk yardage/ gadget-play tailback whose fantasy points ceiling is dependent on them breaking a long carry. Furthermore, it means tail backs who are part of a committee don’t start creeping into our player pools. If a running back exists in a committee it probably means a goal line specialist is waiting to vulture his touches once the team reaches the red zone. In 2015, running backs who touched the ball eighteen or more times scored ten or more Draftkings points at an 85% clip. If you’re looking for further safety to be built into those eighteen touches look for a back who averages at least five or more targets per game. A running back who is utilized in the passing game ensures they’re on the field whether the game script is in their favor or not. The ideal is a back like David Johnson or Leveon Bell who both can ring up double digit receptions while also toting the ball thirty times, but not every running back is as gifted athletically like those two. In any case, the more touches a back receives the more valuable they become.
Furthermore, if a running back isn’t getting red zone carries than we’re banking on his value coming from a plethora of rushing yards and or receptions. Truth be told, more often than not, a running back will end up with zero touchdowns instead of one. However, pinpointing the weeks that backs do find the end zone, maybe even multiple times will be our most profitable. Thus, besides seeing which teams are giving up more or less touchdowns via the carry than through the air, it’s important to note which running backs are seeing more or less of their respective team’s red zone work. In data compiled from 2006-2015, less than one percent of handoffs outside the defense’s own twenty-yard line (red zone) resulted in a touchdown. However, if you move the ball to the eight-yard line than 10% of those carries ended in a touchdown while a rushing attempt from the one-yard line resulted in a 54% chance of a touchdown. If a tailback is being removed once the team gets within the opposing team’s twenty-yard line, the information shows they have a one percent chance or less of turning one of their carries into a touchdown. The closer the ball is to the end zone the more likely the running back is going to turn the play into six points and considering that a touchdown is worth as much as sixty rushing yards, you need to target these situations for your teams to have as much upside as possible.
After filtering out who should see the most opportunities and positive game script it’s time to move on towards the other predictive matrix for fantasy running backs.
First, a running back should have a match up against a defense ranked bottom twelve in rushing yards allowed per game. The easiest path to fantasy points for a running back is going to be by accumulating rushing yards. However, earning only .1 point per rushing yard gained is also going to be the lowest method by which fantasy points are earned. Regardless, we want to target those teams more likely than not to give up rushing yards so as to have a firm base of fantasy point expectations to work off of. Another way to solidify this list is to cross reference it against defenses that rank bottom twelve in rushing attempts allowed per game. If a team is allowing an exorbitant number of rushing yards and rushing attempts you can rest assured that this is a team worth attacking. The team that should immediately come to mind is the San Francisco 49ers who allowed the most rushing attempts per game (34.2) while also allowing the most yards per game (165.9) and touchdowns (1.6) as well in 2016. The 49ers inability to stop an opposing team’s rushing attack made them one of the most exploitable matchups in fantasy last season and more than likely another dream scenario this upcoming season until we see that their new free agent signings and draft picks have changed things. On the flip side, the Dallas Cowboys were dead last in running back carries allowed (21) per game which manifested with a league leading 81.7 rushing yards per game allowed.
In previous years I would have made a point to target matchups against teams that rank bottom twelve in fantasy points allowed or Football Outsiders DVOA rankings. Yet, now I see so much more value in gauging opportunity as opposed to who faces what team. That being said, I wouldn’t fault someone for keeping those metrics in their running back evaluation, especially with DVOA as you can tell which running backs get a bump receiving wise. On the other hand, simply checking yes or no, based on fantasy points allowed, is what the majority of uninformed daily fantasy players are doing as they scroll through player cards. In order to maximize our edge allow yourself to look beyond simple Player A versus Team X scenarios, and seize situations where positive game script and opportunity blend to form the perfect fantasy running back conditions.
When it comes to wide receivers, their upside starts and begins with their targets. If a wide receiver isn’t getting the ball consistently they bring a legitimate floor of zero points. Targets trump every other metric in daily fantasy for wide receivers and it’s not even close. As long as a wide receiver is receiving a consistent share of passes he can overcome a low average depth of target, low catch rate, low touchdown rate, or low yards per reception. However, if a receiver fails to receive a healthy dose of targets he becomes a one trick pony wherein his value is tied to his ability to excel in a particular part of the receiving game (yards per reception, red zone usage, catch rate, etc…).
A wide receiver should see eight or more targets per game. If targets matter so much their has to be a baseline for proper expectations so as to distinguish the good from the bad. While aiming for double digit targets is ideal, if all ten are caught you have a floor of ten points, not all receivers command that high of a market share from their quarterbacks. Because targets directly influence the total yards a receiver will accrue (around 80% of the receiver’s fantasy production will come from receptions and yards), we need an attainable goal that doesn’t leave our wide receiver pool shallow. Eight seems to be the magic number as 85% of wide receivers who scored at least twenty fantasy points did so with eight targets or more while 61% had ten or more targets. In 2015, 507 games occurred wherein a wide receiver drew at least eight targets, 86% of those wide receivers scored at least ten Draftkings points. Ideally, these eight or more targets are going to result in the wide receiver having a market share of 24% or more of his team targets. Owning at least a quarter of his team’s respective passes ensures his volume and role will be consistent providing further safety for our cash teams.
While yards and passes account for the majority of a wide receiver’s value, a portion will come from touchdowns and it’s finding these touchdown weeks, or even multiple scores, that determine the ceiling of a fantasy team. Much like running backs, the easiest path to touchdowns is to find who is utilized the most in the red zone. Only 32% of touchdown passes occur outside of the opponent’s 20-yard line meaning a pass inside the red zone is twice as likely to end up going for a score. When you look at targets alone, the touchdown rate outside of the red zone falls to a measly 2%. The target your receiver gets near midfield and beyond is much more likely to just up as another catch instead of field breaking score. Thus, find out who dominates red zone targets and don’t be afraid to utilize guys who haven’t broken the plane of late knowing that regression is real and can reverse at any time.
It goes without saying that not every target is equal just like not every wide receiver on the end of those passes is equal. Targets can especially differ based on the individual matchups that receivers have with cornerbacks. To gauge matchups, I prefer going deeper than just looking at fantasy points allowed because one bad corner can unfairly skew the collective numbers of a group that is probably tougher than the numbers hint at. Perhaps the best way to analyze this is looking at Football Outsider’s DVOA rankings where you can view how teams defend the #1, #2, and other receivers from the opposing team. Scrolling down further on that same page you can also view where teams are better or worse at defending the field. In 2016, the Broncos were ranked #1 by Football Outsiders defending opposing #1 and #2 receivers. If you dig deeper you’ll see the Broncos (Aqib Talib, Chris Harris) were also ranked #1 against deep passes, short passes passes to the right, and passes to the middle. The Broncos only “weakness” was an 8th place ranking against passes to the left. Needless to say, if you were targeting wide receivers against the Broncos there’s a good chance your wide out under performed that day.
After all this discussion about targets, you may be wondering what role Vegas plays in determining wide receiver plays. Quite frankly, it’s mostly noise as far as Vegas and odds are concerned, but if anything it’s the inverse of what makes a great running back play. If being a home favorite is beneficial to a team’s running back then it can’t be beneficial for the team’s passing game as well. Wide receivers can benefit from being both a favorite or underdog, it’s just a matter of how the game script plays out for them. Regardless, daily fantasy sites will tend to incorporate Vegas odds into their player salaries meaning you can roster a road underdog at a discount. Interestingly enough, in 2015 this value played itself out as road underdogs registered more points per $1000 of salary on both Draftkings (2.76) and FanDuel (1.77) than the perceived top Vegas scenario of a home favorite. Even though road underdogs are going to produce the fewest receptions and yards of all four scenarios, their under priced value is still making them a worthwhile play.
Furthermore, wide receivers on teams with low totals may also be unfairly getting underpriced. Wideouts on teams with an implied team total at or under fifteen points produced 3.53 points per $1000 of salary, the highest of any implied point classification. Further showing implied points total being baked into salaries, the lowest point per dollar return was from receivers whose team was projected to score 24-27 points.
This may sound like a broken record but just like wide receivers, fantasy value for tights ends is tied directly to their opportunities for targets. Even more so when you consider that tight ends are often the most underutilized receiving position on most NFL offenses. Most generally, on a passing play, the running backs (once they leave the tackle box) and wide receivers are all legitimate options for the quarterback’s pass. On the other hand, the tight end could be on all forty-eight of the quarterback’s attempts and never once see a target much less be an actual option.
A tight end should see at least eight or more targets per game. In 2015, there were 49 games in which tight ends scored 20 plus Draftkings points. In those 49 games, 78% of those tight ends had eight or more targets while a little more than half (58%) had ten or more. Targets and yards will account for nearly ¾ of a tight end’s fantasy production, however, the peaks in production occur when tight ends find the end zone.
More so on Draftkings than FanDuel, you can absolutely scrape by on a minimum priced tight end ($2500) so long as the tight end has touchdown upside. Quite frankly, spending as little as possible on a position that is prone to underperforming is not a bad thought process, especially in cash. Not to start throwing up qualifiers everywhere but if you are going to go barrel scraping for tight end value so you can fit in that high priced running back, that tight end absolutely must be a red zone option for his team. Tight ends with week winning production are doing so off of touchdowns, and considering the easiest path to a touchdown is receiving a pass in the red zone it’s this variety of passes your tight end needs. Tight ends converted 32% of their red zone passes into touchdowns in 2015, while wide receivers did so only at 23% and running backs at 18%. Furthermore, over 90% of the touchdowns scored by tight ends from 2012-15 were scored within the red zone. You can easily break down into a pie chart how the quarterback spreads the ball around when his offense is near the goal line. If the tight end isn’t seeing at least 15% of those red zone passes you’re taking the chance of rostering a zero on your team. The assumption is that the tight end would at least get a couple points based on a few targets, but if his salary is that low his per game targets are small enough that one or zero passes is a conceivable outcome.
Besides hunting for red zone targets the simpler path to find touchdowns is going off predictive game script via implied game totals. The math is simple, the higher the team’s total the more likely that team is to make repeated trips into the red zone. The more red zone trips a team makes the more likely a tight end is to catch a pass for a touchdown. In 2015, tight end fantasy points (Draftkings) broke down the following way in points per $1000 of salary and their implied point total.
|Vegas Scenario||Under 18||18-21||21-24||24-27||27 and above|
|Points per $1K||2.20||2.5||2.79||3.05||3.07|
The clear point is the worse Vegas projects a tight end’s team the worse his production should be. Based on the information above, 24 or more implied points appears to be the sweet spot with their being little difference between a tight whose team is projected to score 25 and one whose team is implied to score 30. Hearkening back to previous data, we often see our highest projected teams as home favorites while our lowest teams are road underdogs. Thus, target tight end’s whose teams are home favorites with an implied team total over 24.
Besides analyzing targets and touchdown probability, looking at fantasy points allowed isn’t a bad metric to consider. Whether the tight end is matched up against slower linebackers or undersized safeties, tight ends can consistently win in situations where good matchups present themselves like Dallas last season or the Giants for what seems like the last decade. Most generally defenses will put cornerbacks on islands with wide receivers but rarely assign a shutdown defender to the tight end which should mean more favorable matchups once the tight end gets past the defensive line. However, some teams like the Chiefs, have no issue taking their best defender (Eric Berry) out of coverage to shadow a tight end basically nullifying their impact on the field.
Defense/ Special Teams
Pegging the top DST plays each week can feel like a bit of a crapshoot because often the top plays are determined by touchdowns. Trying to predict which special teams group will return a punt or kickoff to the end zone or which quarterback will toss a pick-six is a bit of a fool’s errand. If someone could properly predict when these type of touchdowns will occur they should be doing something else besides playing DFS. However, just because the peak of DST scoring occurs under random circumstances doesn’t mean there aren’t predictive measures, as at any other position, that give us baselines as to which team is better than another.
Just like with tight end, there are no concrete factors that a DST must meet, but I do have seven metrics that I want to see a team check off. Because the defense accounts for nearly all of the fantasy scoring that occurs, remember the only time the special teams accrues points is via a touchdown, all seven of my metrics are based on the respective offense that defense will encounter. Remember, I’m not looking for a defense to meet all seven criteria, but the more boxes a team can check off the better.
First, a defense should be a home favorite. The following graph breaks down the average stats that home favorites, road favorites, home underdogs, and road underdogs are putting up in regards to the three most predictive fantasy point metrics for the DST position; points allowed, sacks per game, and interceptions per game.
|Vegas scenario||Points allowed per game||Sacks per game||Interceptions per game|
As you can see, rostering a home favorite means access to the fewest points allowed, the most sacks, and the second most interceptions. Because road favorites are not that far off in those three categories and have even better interception numbers which can turn into touchdowns should the blocking be favorable, it might make sense to roster more road favorites in your tournament teams whereas home favorites in your cash teams. Home favorites are going to provide you with a layer of security, which is ideal for your cash games, but pivoting to a road favorite should give you a higher floor for interceptions thus giving you a little higher ceiling.
Second, a defense should be favored by four points or more. If you’ll remember from the running backs section, a four point or larger favorite typically means the favorite is running out the clock towards the latter part of the game. Thus, the less the opposing offense has the ball in their hands the fewer yards and points they can accrue. While it could be considered metric 1A, something else I like to note is a team that is a large favorite in a game with a lower (44 points or lower) over/ under. The fewer the implied points the opposing offense is predicted to score the better. Something else to consider is that people will naturally be attracted to a double digit favorite as opposed to the team favored by only four points. However, little evidence exists to show that double digit favorites actually score more fantasy points than smaller favorites, provided they’re four point or larger favorites. Keep this information in your back pocket as a way to find differentiation in large field tournaments when large home favorites hosting the Browns, Rams, or 49ers become the chalk.
Third, a defense should have a matchup against an offense that is ranked bottom ten in points per game. As mentioned above, points allowed per game is one of our three most predictive criteria when it comes to DST scoring. Thus, let’s single out the offensively inept teams who are scoring the fewest points per game. If you feared that the points per game stat won’t paint a clear enough picture, I would cross reference that against what that offense is doing snaps per game and time of possession wise. If an offense is a bottom ten in snaps per game and time of possession you have a clear explanation for why their average scoring is so low. An offense that isn’t converting third downs and extending drives is in a bad pattern (Chip Kelly, cough… cough… cough…) of going through the play clock too fast and handing the ball back to the opposing team. Thus, the less time and plays they have with the ball in their hands the fewer opportunities they have to score. This ties into my fourth metric, a matchup against an offense ranked bottom ten in total yards per game. Total yards created are not 100% indicative of a healthy, fully functioning offense thanks in part to garbage time and how a defense, sitting back in prevent can allow yards to accumulate as they protect a lead. However, if an offense (like the Rams in 2016) can’t manage to total at least 300 yards per game you know the passing game isn’t even a threat to take advantage of garbage time when it presents itself.
Fifth, a defense should have a matchup against an offense that is bottom ten in sacks allowed per game. Sacks are our second most predictive path to fantasy points, thus we want to focus on teams with protection issues or quarterbacks who have a tendency to hold the ball too long. We can easily see this when we look up sacks allowed per game, but another way to analyze this is seeing which teams are throwing more than others. The more a quarterback is asked to pass, the higher likelihood he is to get sacked. On the flip side, an offense that is more likely to hand the ball to their running back, even when losing like the Jeff Fisher era Rams, the less likely your defense is to accumulate sacks as game script turns on their quarterback. Furthermore, the more a quarterback passes the ball the greater the chance he is at throwing interceptions, which brings me to my next valuation.
Sixth, a defense should have a matchup against an offense that is bottom ten in giveaways per game. While defenses should always have more sacks than fumbles and interceptions, it’s important to consider which offenses are more likely to turn the ball over. In 2016 the top ten quarterbacks, who registered the most pass attempts (minimum 300 passes), averaged 14 interceptions. Quarterbacks ranked 11-20 averaged 11.1 interceptions on the season while those ranked 21-30 only threw 8.3 interceptions for the season. Furthermore, since defensive touchdowns occur at a more frequent rate than touchdowns converted from punts or kick offs, it should be our focus to exploit quarterbacks who drop back more. Touchdowns are what separate good DST plays from great plays, therefore, it’s only natural we focus on our easiest path to them, interceptions and fumble recoveries as this graph from 2016 shows.
|TD scenario||Kickoff||Punt||Fumble recovery||Interception|
Furthermore, besides the fantasy points that these giveaways accrue, and the possible touchdown opportunity they create, a turnover also means the opposing offense is leaving the field. If the opportunity is the name of the game, fantasy points wise, for offenses then decreased opportunity is the name for defenses.
Finally, a defense should have a matchup against an offense that ranks bottom ten in Football Outsider’s DVOA offensive efficiency rankings. This rating is going to give you a snapshot of an offense’s running attack and passing game efficiency that also compares success rates for every single play to league based averages and opponents. Furthermore, it also weights false starts and delays of game which impact the number of snaps, and their quality (3rd and 4 becoming 3rd and 9 because of a false start), that an offense executes.
Despite what you may think, there is more to selecting a kicker than looking for the cheapest guy who averages the most points per game. The kicker spot has metrics too that can’t be ignored if we’re looking to maximize the fantasy points potential of our lineups.
A kicker’s team should be a four point or larger favorite. Once again, if a team has a lead larger than a field goal not only are they consuming game clock, but the pressure to convert drives into touchdowns is less as well. If an offense has a three point lead and turns a drive into a successful field goal they can still lose should the opposing team turn their next drive into a touchdown and extra point. However, should a team have a four point or larger lead the need to score a touchdown isn’t nearly as urgent as just putting up three more points and extending the lead into seven points or possible two possession territory? On the flip side, if a kicker’s team is trailing late in the game we should expect them to utilize him less as they’re looking for touchdown opportunities, and even two point conversions should they score, instead of field goals. Just as with running backs, don’t underestimate the role that game script can play on the utilization of a kicker.
A kicker’s team should have an implied point total of 24 points or larger. Simply put, the more points a team is projected to score the more field goal/ extra point opportunities that kicker should see. The more opportunities a kicker has the more points he should score. In 2015 the impact of implied points and fantasy points per $1000 on FanDuel broke down thusly:
|Implied points||18 and under||18-21||21-24||24-27||27 and above|
|Points per $1K||1.30||1.43||1.52||1.65||1.62|
The only time the trend of rising implied points correlating to rising fantasy points getting bucked was when teams were projected to score 27 or more points. Regardless, a kicker whose team was implied to score 30 was still outscoring a kicker whose team was projected to score 23 points. That being said, it’s entirely possible that a kicker’s team scores 28 points leaving him with only four points off of four extra point tries, the same can be said of a kicker whose team has a low implied total. If his team were to score only seven points then that kicker would end the day with only one point off of his single extra point.
A kicker should have zero worries about weather impacting their output. In an attempt to mitigate weather concerns to a factor of zero I will prefer dome or closed roof kickers so that any worries about precipitation, wind, or field conditions don’t have to be considered. If by chance all the kickers are outside, or have had their retractable roofs removed, I need to figure out which kickers who are home favorites with an implied team total of 24 or greater have the best on field weather conditions. For example, has it been raining at his stadium or is it currently raining? I don’t want a sloppy field to impact how likely the head coach is to use the kicker. Also, what does the wind look like during game time? The windier it is on the field the tougher it becomes for kickers to nail their height and angle projections for where they kick the ball.
A kicker should be averaging top ten or twelve in the league in field goal attempts per game. What offenses are putting their kickers in a position to kick multiple field goals per game? In 2016 the league average for field goal attempts per game was two, a benchmark hit by only three teams; New England, Dallas, and Los Angeles (interestingly enough, Steven Gostkowski, Dan Bailey, and Josh Lambo were three of the more chalkier kicker picks week to week). Meanwhile, Washington led all teams with 2.6 attempts per game while San Francisco was dead last with only 1.31 attempts. If your prospective kicker isn’t seeing more than two attempts per game, you’re hoping for the rest of his value to be made up by extra points or that his possible lone attempt comes from 40 plus yards away. Remember, FanDuel awards all field goals 39 yards and under as a three point score. Field goals in the 40-49 yard range earn four points while anything beyond 50 yards earns 5 points. Side note, if you were wanting to maximize the attempts that your kicker makes, a simple search can show who is attempting the most kicks in the 40-49 yard, and 50+ yard range (In 2016, Justin Tucker made all fourteen of his attempts in the 40 yard range and all ten of his attempts from 50 plus yards).
Of course, all of this is for not if the kicker is not converting those field goal attempts. In 2015 and 2016 the league average for field goal attempts made hovered around 85.7%. Last season, nine of the fifteen teams that averaged under 85.4% of field goals converted were also at or under the league average for field goals attempted. The less likely a kicker is to convert a field goal opportunity, the less likely his team was to ask him to do so, especially in regards to any kick 50 yards or longer. Teams under the league field goal attempt threshold of 85.4% were a collective 34-78 (43%) on kicks longer than 50 yards, with nine of those fifteen teams attempting three or fewer of those attempts.
Kicker tends to be a throwaway position, especially for those who primarily play on Draftkings and are just looking to maximize their profitability by taking the team they built there and moving it to FanDuel. Because these people, or just lazy FanDuel NFL players in general, aren’t giving the research time to the kicker position that they are to other positions, they will often search out the cheapest option that makes lineups work. Use the edge that you have in selecting kickers while others are just ignoring what amounts to 11% of lineups.
Despite what you may assume, or what other daily fantasy sports sites have tried to convince of, having a statistically informed player model doesn’t have to be this extensive formula that takes a data scientist to decipher. If you will simply take each player position, investigate no more than seven metrics per position, you can reach the same conclusions as people who are divesting way more time and energy into NFL DFS. Allowing your decisions to be informed but not overly invested into too much information keeps your process simple and easy to replicate throughout the NFL season.