Welcome back to the Data Viz DFS NFL Preview utilizing data science & data visualizations – today our focus is tight ends. We will be utilizing % of team air yards, targets per game, yards per game, & total TDs to understand what TEs have the best value for week 8.
If you recall from the WR post – the reason I chose these 4 statistics because they all have an r squared score > .5 when compared to fantasy points – which means they have a high positive correlation. Well an interesting thing happened when I looked at the TE scatter plots – TDs had an r square score < .5. But, before we jump get into the analysis lets take a huuuuuuge step back.
So now you may be saying to yourself “That’s all great and everything – but I have no idea what the hell r squared score or high positive correlation mean & frankly don’t know why it matters.” Well luckily for you – I am going to break down correlation, scatter plots, & basic data science principles to give you a more informed choice when picking week 8 TEs. Basically I’m going to use nerdy math shit to win you money – so if you like money, and only if you like money, keep reading for a full breakdown & analysis.
Today’s first viz is 4 different scatter plots – with each being compared to total fantasy points (y axis). The four statistics on the x-axis are: receiving yards per game, receiving TDs, % of team air yards, & receiving targets per game. Each circle represents a different TE, the color is based on total fantasy points (blue is good & orange is bad), & all are positively correlated.
Before we jump head first into the analysis, let’s define a few key terms.
Key Terms:
Scatter Plot: A graph where the values of two variables are plotted along two axes (x & y), the pattern of the resulting points reveal if any correlation is present. The purpose of a scatter plot is to show possible associations or relationships between two variables.
Correlation: The process of establishing a relationship or connection between two or more variables.
Positive Correlation: The values of one variable increase as the values of the other increase. Basically if you improve one variable – you will see improvement in the other.
R Squared Score / Correlation Score: The main purpose is to predict future outcomes on the basis of other related information. In general the higher the R-squared score, the more confident you can be when predicting future outcomes.
% Of Team Air Yards: The sum of the receivers total intended air yards (all attempts) divided by the sum of his team’s total intended air yards. Represented as a percentage, this statistic represents how much of a team’s deep yards does the player account for.
Intended Air Yards: The vertical yards on a pass attempt at the moment the ball is caught in relation to the line of scrimmage. Air Yards is recorded as a negative value when the pass is behind the Line of Scrimmage. Additionally Air Yards is calculated into the back of the end zone to better evaluate the true depth of the pass.
Key Takeaways:
Takeaway #1: The standard equation for fantasy points combines yards & TDs. Based on this I would assume both yards & TDs would have a high correlation score when compared to fantasy points – but that is not the case for TEs (correlation score = .21). This tells me that tight ends get the majority of their points from yards rather than TDs – basically tight ends do not score enough TDs overall to have a high correlation to fantasy points. Which makes receiving yards more important that TDs, but also puts a premium on the TEs that are able to score TDs (shout out Austin Hooper). As you can see from the scatter plot, the high TD totals are NOT all blue circles – which means total TDs DOES NOT directly lead to fantasy points for TEs.
Takeaway #2: Percent of team air yards & receiving targets per game have high positive correlation (r squared score = .53), but are not directly involved in the equation to produce fantasy points. This is where it starts to get interesting – we can utilize these 2 stats with relatively high confidence to predict future outcome due to the r squared score being > .5.
Takeaway #3: Percent of team air yards has a correlation score of .52 when compared to fantasy points. The more likely the TE is involved in explosive plays downfield – the better chance the TE puts up fantasy points.
Takeaway #4: Receiving targets per game has an outrageous correlation score of .71. This is great news for us because targets per game leads directly to fantasy points. If we can predict correctly what TEs will have high targets in week 8 – we will be able to predict what TEs will have a good fantasy week with high confidence.
Takeaway #5: We are going to take a deep dive into all 4 statistics using 1 easy to read visual. This way we can utilize all factors to predict week 8 TE performance. I will present players I like & don’t like for week 8 below, but remember this is just my opinion based on the data. There is no correct answer when it comes to Fantasy Football, so do your own research & always make your own picks based on how comfortable you feel with the player. Hopefully this gives you a great starting point & a general direction to follow when researching for the week.
This viz is a vertical bar graph where the length of the bar shows % share of team air yards, the width of the bar represents targets per game, the color of the bar represents yards per game (blue is good & orange is bad), & the grey circle indicates total TDs. The TEs are ordered by DK fantasy points per game. All four statistics that have a positive correlation are included in this viz – continue reading for the TEs I like & don’t like for week 8.
Top 15 Active TEs
TEs I Like For Week 8
Hunter Henry vs Bears| Austin Hooper vs Seahawks| Gerald Everett vs Bengals
Evan Engram vs Lions | George Kittle vs Panthers| Zach Ertz vs Bills| Greg Olsen vs 49ers
Ask me any questions about Data, Visuals, DFS, etc. on Twitter (@StephenDataViz) or Instagram (@StephenDataViz)
