Over the years we have all had the best regular season team in the league only to lose in the playoffs to a team much weaker that got hot at the right time. A lot of people just shrug this off as the luck factor of fantasy sports. In some regards this is true as fantasy football is very luck dependent. However, there are a lot of ways that we can increase our chances of taking down a fantasy championship and the cash and laurels that goes along with it simply by playing the right odds.
At times my style might sound extremely aggressive, rash, or just plain crazy. In some ways it might be. But that is only because of the way fantasy football payouts are typically structured which is that they are very heavily weighted to only a few weeks, the fantasy playoffs. For example, a lot of the leagues that I have participated in payout the winners of the playoffs 50%, 75%, or even 100% of the prize pool. Not to mention all of the leagues that just play for bragging rights where the only thing that matters is if you win the playoffs. I have even seen this go to the extremes where 50-75% of the league makes the playoffs so you don’t even have to try that hard to qualify for the playoffs.
Given all this emphasis on the playoffs it’s important to realize that not all points are created equally in winning your league. If we are really trying to maximize how much we will win we should weight each week by how important it is to winning the prize. How much weight should be placed on each week?
In order to answer this question I gathered data from several leagues and built a Monte Carlo simulation that would show exactly how important each week was in determining what your final payout would be. This works by essentially creating a thousand different fantasy football leagues and then simulating the results of every week thousands of times. Sometimes the best teams beat up on the bad teams and win as you would expect. Other times the best team’s stud RB and QB gets hurt in the first quarter of week one and the same high flying team finishes in the basement. By playing this simulation over and over we can see which weeks have the most impact on our final payout.
So how much is each week worth? Unfortunately there isn’t a simple single answer to this question because it is heavily dependent on the size of your league, scoring methodology, how your payouts are structured and how many teams make the playoffs. But for simplicity sakes a fairly standard 12 team, non-PPR league where 6 teams make the playoffs and pays 70% of the pot to the super bowl winner and 30% to the runner up it turns out that week 14, 15, and 16 are worth about 1.8, 4.0, and 1.6 times more than every other week in the regular season for an average team.
That said we aren’t average players because we have all the great tools from Footballguys at our disposal. If instead we assume we are actually in the top 25 percentile of all owners, the playoffs are worth even more because we are more likely to make it to the playoffs in the first place. In this case they increase to 1.8, 4.8, and 2.3 times more important than every other week in the regular season. In general, the better your team is and the higher the payout to playoff winners the more you should weight the playoffs.
We should actually continue to keep track of these multipliers as we move along in the season to see if we should start trading for players that have great playoff matchups or picking up high upside backup runnings backs. For example, if our top 25 percentile team started the season 6-2 they should be valuing the playoff weeks at 2.7, 7.4, and 3.2 times more than the rest of the regular season points which is a big deal and means you should be making aggressive moves to plan for the playoffs.
The big take away from this is it is very important to work on weekly projections and not just season long projections. I started laying out this approach in my pass ratio articles this week where I used Vegas spreads, defense tendencies, and potential weather issues to see if there were specific teams that are likely to pass or run more in any given week. I also adjust yardage totals by potential weather concerns and defensive matchups. It might sound risky to make some of these assumptions before the season even starts but these are the types of risks we need to take to put ourselves in the best position to win. If someone in my rankings looks a lot different from their ADP this weekly strategy might be why.
My goal is to help you best identify players that are going to score the most “payout points” and not necessarily the most points. Since our goal isn’t to just make the playoffs but to win the championship and take home the prize.
Let’s look at a couple of other specific ways that the ADP market is typically wrong and how we can capitalize on these types of market inefficiencies so we can maximize payout points.
Suspended players
The first subset of players I want to talk about is players suspended at the start of the season. The reason I like to start with this group is because it is the best place to exploit market inefficiencies. This inefficiency is so pronounced that last year I ended up with Josh Gordon, Justin Blackmon or both in nearly all of my drafts. The simple reason is a lot of people do a poor job of valuing these players when you are looking to maximize payout points.
What most owners do for these players is look at the player’s raw projected stats to see where they would rank compared to all other players. Then some compound this problem by bumping the player down even more because they aren’t getting as many playable weeks from the player. By valuing players in this way the owner is forgetting that the total points that they are projecting from that player isn’t an accurate way to rank the player because they can bench that player and get replacement level points in their starting position the weeks they are out.
But there is yet another advantage that we are interested in. When projecting the raw total points for the player the owner used a lower amount of games in their projection. This means that in the games that the player is actually going to be playing he will have a higher VBD than a player ranked near him. This is important because the weeks he is playing will include the fantasy playoffs that we saw are worth more.
Here is an example of these two factors in play. Ray Rice is currently the most interesting suspended player. Rice has a current ADP of RB27. In fact, if I run my season long projections for Rice he ranks as number 26 which is right in line with ADP. This part is important because we don’t want to just target players who fit a profile if we don’t believe in them. Based on a standard scoring system Rice should be projected to score 188 points this season over his 14 games. Ranked right above Rice in ADP is Rashad Jennings who is projected to score 190 points over 16 games.
Let’s walk through exactly why this ranking doesn’t make sense. We start by looking at their per game averages. Rice would be projected at 13.4 points per game and Jennings at 11.9 points per game. For simplicity's sakes, if we value points from weeks 14, 15, and 16 at 3 payout points we get a value of 255 payout points for Rice compared to 250 for Jennings. Rice should really be ranked higher than Jennings right off the bat!
Rice’s value is calculated by taking his 10 weeks * 13.4 Points/Game *1 payout point + 3 weeks*13.4 Points/game * 3 payout points = 255 points
This would be a good starting point if we had to play Rice every week. But in reality we can bench him the first two weeks and use a player from our bench. If we add in a replacement level player of 7.5 points per game for the first two week of the season that Rice is out we end up with Rice at 270 payout points to Jennings 249.
These adjusted payout points for Rice would actually rank him closer to RB20 instead of the RB27 that he is currently going at. RB20 is currently going off the board at pick 49 instead of pick 59 where Rice is currently going. We have now gotten a player worth considerably more a round later without even changing our projections.
Unfortunately there aren’t as many exciting suspended players this year as there has been in years past. However, that could change if Josh Gordon gets his suspension cut from the full season to something much less.
Injured stars
Players that are expected to miss the first few games of the season due to injury are typically mispriced in the fantasy community in the same way that suspended players are. Fantasy owners don’t properly value their replacements and don’t factor in the fact that they will score you more points per week during the playoffs. As such injured players are also high on my target list.
For example, last year Rob Gronkowski was another one of my most heavily targeted players during drafts. It didn’t work out last year but I would continue to do it every time. Gronkowski is a little different this year because he isn’t getting the same missed game discount as last year. But I will be watching to see which player is projected to miss the first few games and how much his ADP changes so I can take advantage of it.
However, I don’t get quite as excited about injured players as I do suspended players because of the additional risk surrounding their injury. Most owners assume that injured players will come back at full strength and play out the rest of the year at their historical averages. Rob Gronkowski, Danny Amendola, Arian Foster, and Robert Griffin III III are good examples of why that doesn’t always happen. Additionally, with suspended players you know exactly how long they will be out whereas injured players you can only speculate. I am still going to be interested in injured players, especially difference maker stars that are mispriced, but I am just a bit more leery than I am about suspended players.
Backup running backs
The last group I want to mention is backup running backs. Here my goal isn’t to accumulate the most points at all, which is how rankings are generally compiled. My goal is to try to accumulate the most VBD late in the season.
For this example we will look at Knowshon Moreno and Knile Davis. During week one of the season you would expect Moreno to score 5-6 points because he is already in a time share in Miami. You might also expect Davis to score 0-2 points since he probably won’t play much unless Jamaal Charles gets hurt in the game. But really those points from Moreno shouldn’t hold any value to us over the points from Davis because you didn’t start Moreno anyway. You can keep applying this same rationale all season long until one player gets into a situation to make your starting lineup.
Now assume that both Lamar Miller and Jamaal Charles get hurt in week 13. Up to this point in the season the value for both Davis and Moreno has been zero since neither has started a game for us. This week we can finally plug them into the lineup. Who are we going to start? Here I see Davis as the clear favorite over Moreno for several reasons. Just to name a few,
1) Davis doesn’t have anyone to share his carries with whereas Moreno still has Daniel Thomas
2) The Chiefs tend to use a single back where as the Dolphins don’t mind a backfield by committee
3) Davis is a highly talented highly drafted player that could offer great upside
If Miller and Charles were both to stay out for the rest of the year there is a good probability that Davis has more value than Moreno.
To recap, neither player will hold any value until they take over the starting job, if they both do then Davis has more value. Moreno currently has an ADP of 97 while Davis has an ADP of 164. The only way this can make sense is if Moreno’s chance of taking the starting job from Miller and keep it all to himself is very high. I don’t think it is.
If total points don’t make sense for ranking backup running backs what does? What we really need to do is estimate how many points per game the backup running backs will score if they were to take over the job and what the probability is of them taking over the starting job.
These are some of the qualities I use to evaluate these potential points:
1) Injury history and age of the current starter - To help determine if there is a higher chance of the starter to get injured. If the current starter is already injured that is even better.
2) Skill level of the current starter minus skill level of the backup. - This will help identify if the backup has a better chance of overtaking the current starter’s job. However, if the current starter is a veteran coaches are often slow to go away from these starters even if the talent gap is small.
3) Skill level of the backup minus skill level of the rest of the backs. - The purpose of this value is to approximate the chances that the backup will have to share the workload with other players
4) Coaching tendencies and how likely the coach is willing to give a full workload to the backup and what their pass ratios would look like with a backup.
5) Offensive firepower. - If the running back has Peyton Manning or Aaron Rodgers on their team they are going to have a lot more chances of scoring TD’s
6) Pass catching ability – Will the back be on the field in passing situations? Is the back just an asset in the passing game and won’t really pick up that many additional carries if the starter goes down?
If we are looking to win our leagues and not just score the most points we want to value the playoffs weeks with their true value. This varies by league but is approximately three times more than the regular season. This all starts with creating weekly projections so you can identify which matchups are most favorable for each player. We can also target specific player profiles like suspended or injured players and high upside running backs that are mispriced by the fantasy community. By setting rankings this way you won’t win any most accurate award rankings. But accurately predicting total points isn’t the goal of fantasy football. The goal is to win the league and focusing on the player’s true value is an easier way to do that.