Thursday, July 6, 2017

A Look at Last Year's CFF Scores in Predicting Value for This Year

Follow me on Facebook, YouTube, and Twitter!
Read about all of my Fishing Adventures!
Check out my Fish of 2014
Follow my Fish of 2015

While it isn't quite time to start breaking down some individual position rankings, I think it is the right time to discuss some trends from last season and how it should affect your preparation going forward. First things first: last year wasn't a banner year in terms of fantasy football for me. Now, I did make the playoffs of every league I played in, but I only cashed a check in one of those leagues. To be fair, of course, I was in only three money leagues and was second in the biggest of those leagues. 

Anyway, that's enough about me. 

Two weeks ago, we began our preparation talking about teammate double ups. You can check out some of the very best double ups and some great sleepers for this year by reading the post below.

Today I want to discuss a few things and I really hope to get your feedback. 

Because I don't believe in downloading people's rankings (or buying them), I began to build spreadsheets (just like last year) based upon the statistics of returning players. From there, I will begin to rank. Until then, let's take a look at some trends.

First, let's define our scoring system used last year: 
The positions were QB, RB1, RB2, WR1, WR2, WR3, TE, FLEX, K, DST. 
1 point for every 20 passing yards, 4 points for a passing touchdown, -3 for any turnover.
1 point for every 10 passing yards, 6 points for a rushing or receiving touchdown. Return yards and touchdowns were scored as rushing yards. 
3 points for a field-goal, 1 point for an extra point.
For the DST, you began with 30 points and lost 5 points per touchdown but gained 3 points for a turnover and 6 for a defensive touchdown.

Ok, so it was fairly standard. Here is what you really came to see. In our 20 team league, the average winning score was 152. Simple enough. Now, obviously, there are some blowouts. There are some freak low and high scoring weeks. Obviously, there are also some really good and bad teams that skew numbers. Right? 

Not so. In fact, if you took the top 8 teams (only those who made the playoffs), you find that the average winning total was still 152

How does this affect your draft preparation? Well, let's take a look. 

I crunched the numbers for all returning players from last season (I needed to do that anyway and I felt that using only this year's players would help more than it hurt). 

The PPG average for the top 50 QBs is 20 
The PPG average for the top 100 RBs (PPR) is 13
The PPG average for the top 300 WRs (PPR) is 12
The PPG average for the top 30 TEs (PPR) is 11
The PPG average for the top 20 Ks is 9
The PPG average for the top 20 DSTs is 25

Ok. so I know that last one has thrown you a bit. Again, consider that I am only using data for returning players, meaning that about a third of last year's elite players are gone at each position. But, I will make an argument (in another post) at why it doesn't matter. 

If you add all of that up, you come up with an average of 140 PPG. That's not woefully behind the 152 point average, but it's basically playing a position down each and every week. While a great player (a Lamar Jackson) can make up a lot of ground, not everyone is so fortunate. This comes down to point differential per position, something we discussed in depth last year. Click the link below and read about it.

Assessing Value to Position for Your 2016 Fantasy College Football Draft

So, you need to find 12 points more than your opponet each week. Most people's first thought is that a great QB is the easiest way to get over the hump. The issue is, there are typically five QBs who beat the standard deviation per year. In terms of returning QBs, there are are just two: Lamar Jackson and Quinton Flowers. Both of these beat the 5.8 Standard Deviation between the other top QBs. 

In other words, they are the only two who don't have a value-based ranking (where you can decide to draft or not draft based upon value of other positions). Is your luck that good? Well, unfortunately, if you are in an odd spot in the draft, you are just out of luck. 

The Standard Deviation at RB is a paltry 3.5. So, while there there isn't a great amount of point differential from one to the next, there is a plethora of running backs over the average. This means that there is inherent value in taking three above average RBs over a QB, especially the elites, provided you can get one. However, there around 3-5 of these guys.  But, used together, 3 (assuming the use of FLEX) such RBs can make up 10.5 points. 

The same can be said with receivers. While the position scores a modest 12 PPG (based on the top 300, of course), the top 30 score over 20 points per game. With a Standard Deviation of 4.6 points, there is a drastic step from one to the next, especially at the top. There are around ten WRs out there that will beat the Standard Deviation, making them a priority. Taking 3 of these players equates to almost 14 points per week.

Tight end has a deviation of just 2.8, meaning that there is virtually no way to make up a major point differential. There are only two TEs per year that typically beat the model, though it is worth mentioning that they, alone, can make up the point differential. There is obviously a ton of risk with using the TE to make up points. 

In summation, there are a couple of ways to ensure that you can draft a team capable of hitting 150+ points each week, even if you don't get a premier QB. Even with an average QB, understanding the standard deviation and average scores of each position can help you find that winning formula.