I'm back with a bunch of visualizations -- let's go!
Time to dive back into more starter-worthy analysis. If you missed my first post from a few weeks ago, I introduced what I call starter-worthy scores for each individual position. Check out my first post to get all caught up. In short, I found point thresholds for each offensive position that best separates starter-level production from the rest.
Let's again dive into this by first looking at how often players are starter-worthy based off of Average Draft Position, and how that distribution changes by position. Ah, the good ol' ADP. Play around with the bar graph below to get a better sense of where starter-worthiness comes from in season-long drafts. The bar graph shows, across the horizontal axis, the proportion (or percentage) of games for each season shown (i.e. 16 games for 2015 to 2017, 9 games for 2018) a player is starter-worthy. The count on the vertical axis shows the number of players with that proportion. Using the proportion allows for easier comparison of 2018 to earlier seasons. The data shown is only considering players with least one starter-worthy week in a season. Enjoy!
A couple of things I come away with here:
Not only do we see more RB green and red on the high proportion side, we see it increasing over time. Check out the line graph below to get a better sense of this change since the 2015 season.
Another point of clarification -- the starter-worthy values I am using are derived from 2015 to 2018 data, which explains the QB dip in 2017 and spike in 2018. Some years there are just less starter-worthy players than others. This definitely fits the narrative this season that you can find an option at QB almost anywhere. Other than Buffalo. Anyway, top-of-the-draft RBs are either getting more consistent (see ADP 1 to 12) or staying level. Your pass-catchers are not quite there, although the 1 to 12 WRs are making a nice jump this season. Tight ends are a bit all over the place and we really see the importance of being one of the first to grab a TE in your league.
You are probably yelling at your computer, tablet, or phone "I WANT TO SEE WHICH PLAYERS ARE MOST STARTER-WORTHY!" Well here you go, this last chart I have for you is a fun one (and part of the reason this post took me a while). It shows individual starter-worthiness proportions by position and ADP, but wait there's more. You can:
First thing's first, that top left of the plot is all RBs, echoing with what we were seeing before. Switching to the "Ranked Position ADP" we see the fast decline of TEs and that QBs are more level throughout their ADP run. However, in this view comparing QB/TE "slopes" to WR/RB is not a great idea given that there are significantly less QBs and TEs drafted. The main idea to take away here is consistency. To win a season-long league you can't pick a dud with your first couple of selections and I think we dug into that a bit here.
As always, I'm very interested in what you see in these visualizations. Maybe I missed something! We can chat it up by email, or on twitter using the buttons below. You can also subscribe at the top of the post or on the right sidebar. I will be making posts more frequently over the next few weeks, so don't miss out. If the interactive graphs aren't working well for your device I posted links to the self-contained applications are at the below, too.
Tweet to @MathWithJerome Follow @MathWithJerome
https://mathwithjerome.shinyapps.io/2_sq_dists/
https://mathwithjerome.shinyapps.io/2_yearly_line/
https://mathwithjerome.shinyapps.io/2_adp_rank_scatter/
Time to dive back into more starter-worthy analysis. If you missed my first post from a few weeks ago, I introduced what I call starter-worthy scores for each individual position. Check out my first post to get all caught up. In short, I found point thresholds for each offensive position that best separates starter-level production from the rest.
Let's again dive into this by first looking at how often players are starter-worthy based off of Average Draft Position, and how that distribution changes by position. Ah, the good ol' ADP. Play around with the bar graph below to get a better sense of where starter-worthiness comes from in season-long drafts. The bar graph shows, across the horizontal axis, the proportion (or percentage) of games for each season shown (i.e. 16 games for 2015 to 2017, 9 games for 2018) a player is starter-worthy. The count on the vertical axis shows the number of players with that proportion. Using the proportion allows for easier comparison of 2018 to earlier seasons. The data shown is only considering players with least one starter-worthy week in a season. Enjoy!
A couple of things I come away with here:
- Quarterbacks are not that reliable, even those in the top 6.
- Running backs are much more consistent at hitting their starter-worthy score than receivers.
Not only do we see more RB green and red on the high proportion side, we see it increasing over time. Check out the line graph below to get a better sense of this change since the 2015 season.
Another point of clarification -- the starter-worthy values I am using are derived from 2015 to 2018 data, which explains the QB dip in 2017 and spike in 2018. Some years there are just less starter-worthy players than others. This definitely fits the narrative this season that you can find an option at QB almost anywhere. Other than Buffalo. Anyway, top-of-the-draft RBs are either getting more consistent (see ADP 1 to 12) or staying level. Your pass-catchers are not quite there, although the 1 to 12 WRs are making a nice jump this season. Tight ends are a bit all over the place and we really see the importance of being one of the first to grab a TE in your league.
You are probably yelling at your computer, tablet, or phone "I WANT TO SEE WHICH PLAYERS ARE MOST STARTER-WORTHY!" Well here you go, this last chart I have for you is a fun one (and part of the reason this post took me a while). It shows individual starter-worthiness proportions by position and ADP, but wait there's more. You can:
- Switch between a player's overall ADP and their ranking within their position.
- Add a linear or curved trend.
- Click on the scatter plot points to see the associated players and get a little more information in the table below the plot.
- Zoom (sorry, not on phones/tablets) by dragging a box on the scatter plot then double-clicking the box (helpful to click on points in a crowded area). Double-click again to reset.
First thing's first, that top left of the plot is all RBs, echoing with what we were seeing before. Switching to the "Ranked Position ADP" we see the fast decline of TEs and that QBs are more level throughout their ADP run. However, in this view comparing QB/TE "slopes" to WR/RB is not a great idea given that there are significantly less QBs and TEs drafted. The main idea to take away here is consistency. To win a season-long league you can't pick a dud with your first couple of selections and I think we dug into that a bit here.
As always, I'm very interested in what you see in these visualizations. Maybe I missed something! We can chat it up by email, or on twitter using the buttons below. You can also subscribe at the top of the post or on the right sidebar. I will be making posts more frequently over the next few weeks, so don't miss out. If the interactive graphs aren't working well for your device I posted links to the self-contained applications are at the below, too.
Tweet to @MathWithJerome Follow @MathWithJerome
https://mathwithjerome.shinyapps.io/2_sq_dists/
https://mathwithjerome.shinyapps.io/2_yearly_line/
https://mathwithjerome.shinyapps.io/2_adp_rank_scatter/