Given the same fantasy production projection and salary, having the lower owned player is the optimal GPP play. OK, now in English… if Aaron Rodgers and Ben Roethlisberger have the same salary, and are both projected to score 30 points, then picking the player that has lower projected ownership is going to be a better play for you in a tournament. I understand there are a lot of assumptions here, but I’m going to help you break this down using simple math.
Why is lower owned better in GPPs? A Beautiful Mind
GAME THEORY! First, let’s try to understand why having a lower-owned player is optimal given the same projection and salary. You are with 19 of your friends at a bar. There are only two girls in the bar. You can’t see them and you know nothing about them. The only knowledge that all 20 of you have about these two girls is that they are 9’s (in the traditional 0-10 scale) and have the same intellect. The bartender offers you a chance at getting a date with one of them. All 20 of you simply get to pick either girl number 1 or girl number 2, and go on a group date with all of the people that picked the same girl (I watch too much of The Bachelor with my fiancée). Now, you are savvy, and in knowing your friends, you know that 15 of them are going to pick girl number 1, even though they know nothing about her. If the goal is to impress this girl with your skills on the group date – in order to get a real one-on-one date with this girl – which girl do we go for? Ding ding ding….the low-ownership girl! Our goal is to limit our competition. Why would you want to compete with 15 of your friends with a girl of the same hotness and intellect, when you can compete with four of your friends with a girl of the same hotness and intellect? And yes, I kind of stole this from the movie “A Beautiful Mind.”
If we jump back to Aaron Rodgers and Ben Roethlisberger, it should be easier to see that given the same projections and salary, that having the player with lower ownership is optimal in a tournament. Would you rather be competing against 4,000 people for a million dollars or 15,000 people for a million dollars?
How Do I Know the Projected Ownership Numbers?
In this day in age, if you don’t know how to Google this, then we need to have a talk. This is the conversation that I have every day with my fiancée:
”John, how do I hard boil an egg?”
”Umm, Google ‘how to hard boil an egg’”
“Why/How do you not know how to hard boil an egg?”
Once a day I have conversations of this nature.
Primary Sources – remember that from research papers?
I’m first going to tell you here that there’s really no way to tell exactly what ownership levels will be on Sunday at 1PM. The good thing is that we do not need the exact numbers, we just need a ballpark. First and foremost, as crappy as these Thursday games are, it affords us the opportunity to learn about upcoming ownership. You can play in these Thursday contests on FanDuel and get the ownership numbers of all of the players you select (obviously don’t pick anyone in the Thursday game if you are only looking for Sunday ownership). An even more accurate estimator would be to play the 9:30AM EST contests on Sunday when there are awful London games.
Secondary Sources – the lazy way – the only way
DFSreport reports the percentage owned from the Thursday slate. The only reason I know that is because I literally just Googled it. I’m guessing that there are a bunch of other sites out there as well.
Now for the Simple Math that I Promised You – the Q-score!
What is a Q-Score?
A Q-Score helps us identify optimal GPP plays in DFS. It combines the projected points, salary, and projected ownership of a player and can be used to determine whether a player should be used in GPPs. The Q-score is a relative score, meaning that it is only used effectively when comparing scores to scores of other players.
A couple things before we calculate Q-score:
- I call it a Q-score because my last name is Quaile. It is such simple math that someone else probably already calculates this, and probably in a way better way than I do (I hope I don’t get sued). This is not a finite score. In fantasy, I like to create little calculations that highlight good situations for me. However, I never use any of my calculations as a final decision rule.
- Salary matters, and is really the only true number I use in the calculation.
- I realize that the ownership numbers only come from FanDuel. That is an inherent problem with the Q-score calculation. Ownership numbers differ across sites because of salary. Although ownership numbers differ, they are still in the same ballpark usually – that’s all we are looking for.
- Projections are projections. This is going to be wildly subjective. In tournaments, I am looking for maximum upside, so I use the ceiling projections here on Fantasy Labs. You can input whatever site you use, or use your own projections in this formula.
- The Q-score is not a tool that projects one player to score more points than another player. It is merely used to decide who is a better tournament play. There is a big difference between the two.
Ready for the Q-Score?!
Q-Score = Projected Points/Salary/Ownership Percentage X 1000
So, it’s basically points-per-dollar divided by ownership. Most of the formulas I use are multiplied by 1,000 because the decimals are so small. I like seeing numbers I can understand.
Let’s use our Big Ben and Aaron Rodgers example, but play around with the numbers.
Example 1: Same Salary, Same Projection, DIFFERENT Ownership
Example 2: Same Salary, DIFFERENT Projections, Same Ownership
Example 3: DIFFERENT Salary, Same Projection, Same Ownership
These examples may seem pretty obvious. Let’s look at a real example from a couple weeks ago — Demaryius Thomas vs. Larry Fitzgerald from Week 8.
This is something that we all are probably confronted with every week. We are deciding on two players for one spot in a tournament with similar salaries and projections. On one hand we have Demaryius projected at 21.5 points (according to Fantasy Labs) at $8,100, versus Fitz at 17.7 points at a $7,700 salary. It already looks like a tough math problem. The points-per-salary are pretty close, with the slight edge to Demaryius. We throw in projected ownership and here’s how it looks:
Conclusion
Be very careful when using the Q-score! There are some chalk plays that are unavoidable in fantasy, and may not yield a great Q-score. For example, in week 7, Todd Gurley was 60% owned in DK. That would have crushed his Q-score. A bad Q-score does not always mean that the player is a bad GPP play. You can put yourself at an extreme disadvantage if you do not have players like that. We know what happened with Gurley that week.
As I have stated before, this is a score that can highlight good situations. It is not meant to be a final decision tool. In other words, there are going to be a lot of high ownership plays in fantasy that are unavoidable. I’ll leave that to you – the DFS aficionado – to make those calls! By the way, I’m 33 years old and that’s the first time I’ve ever written or typed the word aficionado. I had to look up the spelling.
As a fantasy player, I’m always looking for things that can put me in a better spot to succeed. The Q-score is definitely an important tool for tournament play in any sport.
