SOCIAL LINKS:

If you have any questions about the code here, feel free to reach out to me on Twitter or on Reddit.

Shameless Plug Section

If you like Fantasy Football and have an interest in learning how to code, check out our Ultimate Guide on Learning Python with Fantasy Football Online Course. Here is a link to purchase for 15% off. The course includes 16 chapters of material, 14 hours of video, hundreds of data sets, lifetime updates, and a Slack channel invite to join the Fantasy Football with Python community.

Plotting YAC distributions

In this part of the intermediate series, we're going to continue looking YAC yardage for some of the top receivers of the

2020 season. This time, we are going to be looking at all receivers instead of just the WR position.

What we'll be doing in this post is plotting the distribution of YAC yardage as a kernel density estimation

for the top 10 YAC receivers this season (it's week 16 as of the time I'm writing this). The point of plotting

YAC distributions is to find those players with the potential to break out huge runs after the catch. As we'll see

later in this post, 2 of the 10 players we'll be plotting the distributions for are RBs (Kamara and McKissic), and we don't typically of

RBs when thinking of YAC numbers.

Given that, I can encourage you to edit the code to not only include the top receivers in terms of YAC yardage,

but maybe the top 10 receivers who are most relevant to your team, current matchup, and league.

The first thing we'll do is, as always, install the nflfastpy package. If you are working from Jupyter and already

have this package installed on your local machine, you can skip this step. If you're working from Jupyter

and don't have the package installed, conda install nflfastpy or pip install nflfastpy from your terminal.

If you're using Google Colab, though, you'll have to install the package to your runtime using the command

below in a seperate cell.

Next, we'll import our standard libraries. requests and BytesIO will be used later in the code to

display player headshots through pyplot's imshow function.

Next, we'll grab and compile some data for use later in this post. The comments below

run you through what each line is doing.

gsis_id receiver_player_name posteam position yac
264 00-0033906 A.Kamara NO RB 716.0
74 00-0030506 T.Kelce KC TE 543.0
100 00-0031381 D.Adams GB WR 535.0
115 00-0031610 D.Waller LV TE 504.0
265 00-0033908 C.Kupp LA WR 498.0

Awesome, so now we have our df object ready for use in plotting with matplotlib.

Above, we can see the top 5 YAC receivers this year. We can see that Kamara is the leading YAC receiver this year by a healthy margin.

If you run head(10) off our df, you'll be able to see that JD McKissic is the #10 receiver this year in terms of YAC yardage.

Robert Woods is in the top 10 as well, which means the Rams have two receivers this year in the top 10.

Kelce and Waller both in the top 5, and the only TEs in the top 10 (unsuprising). Kelce has a narrow lead of 40 yards over Waller. Both have had fantastic seasons this year, and if you had either

on your team this year you probably made it to the playoffs, given how valuable a good TE was this year.

Moving on, what we'll do now is create a DataFrame that has the top 10 YAC receivers and use that to filter our original df with play by play data.

gsis_id receiver_player_name yac
261 00-0033906 A.Kamara 716.0
72 00-0030506 T.Kelce 543.0
98 00-0031381 D.Adams 535.0
113 00-0031610 D.Waller 504.0
262 00-0033908 C.Kupp 498.0
76 00-0030564 D.Hopkins 489.0
155 00-0032688 R.Anderson 477.0
65 00-0030431 R.Woods 475.0
435 00-0035659 T.McLaurin 465.0
154 00-0032602 J.McKissic 435.0

Now, using this top_10 object, all we have to do is filter our DataFrame using the isin method.

gsis_id receiver_player_name posteam yac position headshot_url team_color team_color2
501 00-0030564 D.Hopkins ARI 0.0 WR https://a.espncdn.com/combiner/i?img=/i/headsh... #97233f #000000
502 00-0030564 D.Hopkins ARI 5.0 WR https://a.espncdn.com/combiner/i?img=/i/headsh... #97233f #000000
503 00-0030564 D.Hopkins ARI 0.0 WR https://a.espncdn.com/combiner/i?img=/i/headsh... #97233f #000000
504 00-0030564 D.Hopkins ARI 0.0 WR https://a.espncdn.com/combiner/i?img=/i/headsh... #97233f #000000
505 00-0030564 D.Hopkins ARI 14.0 WR https://a.espncdn.com/combiner/i?img=/i/headsh... #97233f #000000

We have now have our df DataFrame fully ready for plotting with matplotlib. Below, I've wrote a comment

for each line detailing what each line is doing. Essentially, we're going to be creating 10 axis's (2 columns x 5 rows)

and plotting each player's KDE of their YAC yardage on a sepearate axis. Each axis will have a player image and KDE plot with the player's team color as the line color.

And that's it for this post. We can see most players have a similar YAC distribution. Interestingly, all players have their

maximum likelihood around 0, but it looks like RBs are less likely than WRs to have their plays stopped short at the yardline where the

reception was made. When you think about it, that makes sense. The goal of designed RB pass plays is to acrrue yardage through YAC, not air yards (in other words, RBs, typically, aren't running deep route patterns).

Thanks for reading. If you have any questions about the post, shoot me an email [email protected]