How Well Did High-Ownership Players Perform for Daily Fantasy Football in 2016?

Overall Conclusions

If there were to be just one takeaway from all of this, it'd be pretty simple: the chalk be bustin', y'all.

Before we dive in too far, we should probably set out the parameters for how all of this was determined. For each week, the "chalky" players were determined to be the five most popular running backs, eight most popular wide receivers, and then the three most popular quarterbacks, tight ends, kickers, and defenses. This way, we're accounting for the different roster requirements on FanDuel without watering things down into under-owned players.

For each player or team, the data recorded was their price, salary, score, spread, over/under, implied team total, and whether they were at home or on the road. Additionally, the points they scored the previous week were recorded so as to get a gauge on how important points chasing was in the public's decision making.

Using a player's final score output and price, we can calculate the "value" they produced while being high ownership. Value is that player's total points scored divided by their salary and multiplied by 1,000. So, if David Johnson scored 18 points at a salary of $9,000, he would have hit 2 times value.

Because each position will have a different baseline for what a "desirable" value is, we had to calculate these individually. Baseline values were determined by finding the average value for all players at that position who were at least 1% owned on each slate. This way, we're weeding out players who were not used due to injury or who didn't have roles worthy of a roster slot.

These positional baselines were then compared to the output of each player when they were high ownership. If they met or exceeded the baseline value, then they were considered a success. Remember that this is a pretty low bar to set. We're merely asking that these guys be average, and that's fairly generous given that these are the most popular plays on a given week. Even still, they often had trouble reaching it. Players whose value was below the baseline when they were chalky were labeled "busts."

Just to give you an idea of the expectations here, this table shows the baseline value for each position.

Generally, you're going to want a lot more value than this from your tournament plays. Keep in mind throughout the piece that it didn't take much for a chalky player to hit value, and yet there were still abundant times when they struggled to do so.

Now that we've got that mumbo jumbo out of the way, let's take a peep at how these high-ownership players performed. Again, these were the most popular plays of the week at each position, and they weren't exactly doing so hot.

Mother of all things beautiful, tight ends. Why do you hate happiness?

The threshold for tight ends to hit average value was fairly low, but they hit that just 39.22% of the time, the lowest of any position. If you are using a popular tight end in a tournament, you are asking to get burned.

This is largely the same with defenses, as well. While average defensive output is fairly predictable with some help from Vegas lines, events such as defensive and return touchdowns are harder to predict. Here, we want to find a lower-owned defense that aligns with our process, even if it means we're avoiding what would be considered the "optimal" play.

All other positions are roughly even in their respective success rates, but that doesn't mean we should simply swallow chalk there. None of these positions hit or exceeded value 60% of the time, and remember, this is after we set our threshold for hitting value at a relatively low mark. This is not a glowing endorsement of playing the chalk.

A big part of these failures could be contributed to chasing previous performances. On a given week, our baseline assumption would be that a player would score more points than he did the previous week around 50% of the time. You'd hope this would be higher if the player were coming at high ownership. That was not the case.

The table below shows how often players at each position scored more points when they were popular than they did the week before, illustrating that our decision-making processes may have been too focused on past performance.

The only position where at least half of the chalky options out-performed their previous-week's outing was defense. Tight ends scored fewer points three-fourths of the time. If you're looking to what the player scored the previous week for guidance, you're snagging faulty insights.

For a different angle, here's the average output by position both the week they were popular and the week before. Every position performed below the previous outing.

Again, tight ends. Come on, fam.

What does all of this mean? It means we should feel fully comfortable fading the chalk in tournaments almost every time. If you have a reason for trepidation, don't hesitate to look elsewhere. We should go even further with tight ends and defenses to actively avoid players we believe will be popular. This isn't a comforting strategy, but it seems to be the necessary one based on 2016 data.

There will obviously be times where this is a different situation, though. There may be instances in which a chalky player is worth it. The best way to figure this out is by checking out each position individually to see what caused the chalk to bust, and we'll start things off with the quarterbacks.