Positional PPR Points 2013-16 Part 3

Positional PPR Points

by John Bush

Positional PPR Points 2013-16

I have covered a lot of ground in

Part 1 (http://www.fakepigskin.com/?p=30813)

and Part 2 (http://www.fakepigskin.com/?p=30858).

I conclude with Part 3 looking a little deeper still into this question: Does each NFL Team’s Position PPR Points from the last 4 year give direction for our 2017 drafts? I answered yes, I have laid out a case for this data’s usage methodology. I conclude my case in this article.

Planning Fallacy

Planning fallacy is a human bias where one underestimates the time, skill and resources to succeed.  In fantasy football, since your draft is a plan you need to consider reference class of data for comparisons to mitigate this bias . Below is defines this process for use. I will add “FF Drafting” to apply it to our world!

Using distributional information from previous FF data sources  similar to the sources need for forecast of FF Drafting. This approach is called taking an “outside view”. Reference class forecasting is a method for taking an outside view on FF Drafting. Reference class forecasting uses the following three steps:

 

  • Identify a reference set of data for past and similar projects in FF Drafting.
  • Establish a probability distribution for the selected reference data sets for the FF Draft that is being forecast.
  • Compare the specific project with the reference data distribution, in order to establish the most likely outcome for the specific FF Draft.

See Wikipedia link from Reference class forecasting: Wikipedia

 

League Average of Last 4 Years 

In Figure 1, the overall 32 NFL Team breakdown of positional usage is shown. These percent of positional usage then should be the reference class benchmarks of our analysis. Go back into the Divisions data (Part 2 of Article)  now with these numbers.

Figure 1 League Average of Last 4  Years of Positional PPR Points

Positional PPR Points Slide59

The data from Figure 1 is clear that if for example the Team’s WRs are a group are used at 39% of the time, that is not unexpected. If the Team usages is below 29% or above 49% or above, that should get your attention. Understand what the significance of these benchmarks maybe for your 2017 draft. Skewed positions can point the way to drafting opportunities or to your do not draft list.

QB Usage by Team

The data in Figure 1 begins a series of team usage information by the whole league’s averages. The first data column presents each team’s 4-year % of Team Usage and the 2016 % of positional usage. Both data series are color-coded by highs (green) to lows (red). The entire charts were sorted by 4-year average of team usage.

Questions to consider

  1. What teams are in the highs and lows of each position’s usage?
  2. What is the chance in 2017 of a change in that usage by free agency, drafting or SOS?
  3. How does the 2016’s data compare to the 4-year average?
  4. Why was there a difference in 2016 vs. the overall averages?

Figure 2.   4-Year QB Usage by Team vs. 2016 Levels

Positional PPR Points Slide61

Observations.  The top teams in QB usage are CAR, SF, SEA and IND (Dark Green) while the bottom teams were MIN, BAL, CLE, OAK, and LAR. Note also the 2016 usages in the second column.

SEA QB usage had a down year and most likely will continue forward to a usage closer to 2016.  CLE, DAL, ATL, SD, WAS, and NE had nice 2016 moves up in QB usage. KC NO, and NYJ really dropped. NO should bounce back but with KC and NYJ they are not expected for an improvement in QB.

Consider this a caution list. This is not the complete story but a reference for your deeper analysis.

RB Usage by Team

Figure 3.  4-Year RB Usage by Team vs. 2016 Levels

Positional PPR Points Slide62

Observations.  KC, MIN, and DET have seen a big drop in RB usage. Can they bounce back? We would suggest caution. On the other hand BUF, CLE, OAK, NYJ, ATL, PHI, NO, NE, BAL, DAL, CIN and ARI can be expected to continue a nice usage of the RB position. Note improvements in PIT, TEN, SF and MIA in 2016. Expect bargains there on those teams.

TE Usage by Team

Figure 4.  4-Year TE Usage by Team vs. 2016 Levels

Positional PPR Points Slide63

Observations.  The top 4-year Teams in TE usage continued their ways in 2016.  Note LAR, NO, and DAL had have better TE usages in the past vs. the 2016 season.  Can those teams move back up? The bottom TE usage teams seemed to also continue their ways in 2016 as well. I note some improvement in HOU, SEA, and TB. Watch those for sleepers.

WR Usage by Team

Figure 5. 4-Year WR Usage by Team vs. 2016 Levels

Positional PPR Points Slide64

Observations.  The top teams usages or WR continued their levels in 2016. (Dark Green) We do see a drop in PIT and OAK a little but the HOU team dropped 41.7 to 32.6. That is a very poor 2016. Be cautious on Texan WRs.  As we dig into the bottom Teams, NO has moved up from 35.7 to 41 (Nice).  However, IND dropped from 39.2 to 33.1, SF (36.6 to 30.5), PHI (34.9 to 28.9), and BUF (34.5 to 30.5)  (Beware those teams for 2017).

2016 vs 4-Year Usage Team Usage Differences

I next turn to the 2016 differences within Teams at the positions. This metric emphasizes the extremes in the positional usages in 2016. Use these data to focus attention on the obvious changes. I would use this metric to target certain Team players. 

Figure 6.  Differences of Team 2016 vs 4-Year Usage   QB/TE

Positional PPR Points Slide66

QB Observations. The green colorized teams had the best 2016 positional usages seen in Figure 6 left. The red colorized teams have the worst 2016-year. Questions, what teams can change their usages? Why? Answer those questions for sleeper picks. Interesting QB differences data here can be seen in CLE. This team has an an interesting 3 QBs and given this data, I expect a continued move forward for one of these QBs ( Kessler, Cody Osweiler, Brock,  and Kizer, DeShone). The other high DIFF teams are the usual suspects.

TE Observations. The top teams are fairly predicable as shown in Figure 6 right. I do see that HOU is the top Team with a +7.8 difference.  That change is a strong buy signal for Fiedorowicz, CJ and Griffin, Ryan especially being very late in ADP.  CJ is a strong sleeper as evidenced by this usage change last year. A great second TE for sure. Draft and collect type of player.

Figure 7.  Differences of Team 2016 vs 4-Year Usage RB/WR

Positional PPR Points Slide67

RB Observations. The top teams in RB difference usages are not surprising. I not that SF was the clear top and that strengthens the hand of C Hyde. I am drafting his far and wide. Also the bottom feeders of GB, DET and KC give me concerns about their RBs. I would demand a nicer ADP prices for those Team’s RBs.

WR Observations. The top teams of NO, KC and LAR are predictable and the ADPs of those Team’s WRs are strong. The opposite situation and a warning is for HOU at -9% difference. I am drafting HOU WRs lightly and wanted to get a ADP price bonus for the risk taken. The IND team was due to Luck’s injury and poor year. I remain cautious. I am strong on Garcon as SF went out to change their WR crew. The issue is that is he the only WR that has the skills. M Goodwin is a dark horse their but its wait and see for me.

Figures 7 to 10. Colorized Scaled Differences in QB, RB, TE and WR

These Area Graphs give us visual to the DIFF numbers presented in the previous tables. Multiple views are essential for deeper analysis. I would just scan around and note the interesting positions and Teams. Add those notes into your cheat sheets.

Positional PPR Points Slide69

Positional PPR Points Slide70Positional PPR Points Slide71Positional PPR Points Slide72

Figure 11. Landscape View  of all 32 Team View 2016 vs. 4 years Positional Extreme Differences.   (Green Circles Highlight Extreme Changes)

The data shown in Figure 11 highlight some extremes in differences in each position across the league. The green circle hammer home the standout metrics of change.

Positional PPR Points Slide74

Team analysis over 2013 to 2016 positional usage ratio calculations.

The next figures 12 to 19,  present the ratios of positions vs. each other by teams and year.  Color-coding highlights the highs (green/blue) and lows (red) over 2013 to 2016 by team.

RB/WR (R/W),  RB/TE (R/T),  and WR/TE (W/T). Ratios below 1 point to lower usage by the top position vs the bottom position. Ratios above 1 point to higher team usage by the top vs bottom position by year

Questions to consider:

  1. What are the yearly changes in the three ratios?
  2. Did the 2016 ratios fit with the team’s past usage?
  3. Use a entire view to note what are the extremes by year?
  4. If 2016 was different, what was the source of that change?
  5. How does the 2017 season seem to be predicted by the team’s past?

Positional PPR Points Slide76

 

Positional PPR Points Slide77

 

Positional PPR Points Slide78

 

Positional PPR Points Slide79

 

Positional PPR Points Slide80

 

Positional PPR Points Slide81

 

Positional PPR Points Slide82

 

Positional PPR Points Slide83

Figure 20 Positional Extreme Ratios Over the 4 Years Top to Bottom

The last figure, gives a overview of the ratio extremes and highlights the 2016 extremes. Use to frame the research you might begin using this table as a starting point.

Positional PPR Points Slide84

The data shown in figure 20  presents the top ratios with the team and year. The 2016 performance was highlighted by deep green colorization.  For example, the BUF team in 2016 had 1.13 R/W, which supports the higher team usage in the RB vs. the WR.  In a STD league, the BUF RBs are a good target for drafting. In the PPR the WRs from BUF should be placed in the caution zone.  Note PHI, SF, NE, CLE and TEN also are highlighted by the data. Focus on the RBs in those teams.

The W/T ratio gives a insight into the TE position. High ratios numbers suggest low TE usage. Note the NYJ team owns the top W/T ratios meaning they have not used the TE position at all. Avoid the NYJ TEs.  Note GB, ARI, OAK, NO, DEN, DET, NYG, and MIA all under used their TEs. Watch free agency etc. to see if that team is going to make a move to use their TEs move. Downgrade those team’s TEs if nothing changes in the summer of 2017.

Article 1 to 3 Series Conclusions

  1. Do not be a slave to the past. Use the massive data-set to frame today’s questions for you.
  2. My comments are my current opinions and may change right up to the first 2017 season game.
  3. I have lived in this data pool since last February and I have these tables and graphs in my thinking going into a draft. I would summarize the data extremes and populate your cheat sheets with those data points.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>