Note: this article is part 1 of a series on player performance and age. During this series, we will be looking at Running Backs, Wide Receivers, Quarterbacks, and Tight Ends. For this study, we will be using NFL Statistics back to 1990. For the purpose of fantasy points scored, Rushing and Receiving Yards are worth 1 point for every 10 yards. Passing Yards are worth 1 point for every 25 yards. Rushing and Receiving Touchdowns are worth 6 points. Passing Touchdowns are worth 4 points.

Age versus performance.
Experience versus ability.
Less Filling versus Tastes Great.

All tradeoffs that both NFL teams and fantasy owners debate every year.

Never pick a running back over 30.
A good wide receiver always thrives in his third year.

Fantasy magazines are filled with rules for drafting players at various stages of their careers. Often times, these suggestions are taken at face value without additional statistical analysis to support these findings.

Would it be possible for pcDrafter to modify the player value based on a dynasty league draft, rather than a yearly draft?

There it was. The question was asked that gave me an excuse to research age and performance in the NFL. It seemed intuitive that there should be a correlation and that is should be relatively simple to define the math for those calculations.

I wanted to start this study with an open mind. So the first question I asked had to be: Does age affect performance? I didn’t want to spend a lot of time working on a formula to show player projections if there wasn’t even a real correlation, so I first tackled this question.


Does age affect performance?

Data Set
The first thing I did was collect the data for all NFL players since 1990. I could have gone further back, but this was a statistically valid sample size. Since this is primarily for use in Fantasy Football, I immediately calculated the fantasy points scored by each RB since 1990. I used a simple performance scoring of 1 point for every 10 rushing/receiving yards and 6 points for all rushing/receiving touchdowns. I also wanted to eliminate the plethora of players who didn’t receive any playing time, so I limited my study to those players who had at least 10 rushing attempts in any given year.


General Findings
Since 1990, there have been an average of 104 running backs each season that have had at least 10 rushing attempts. The range for this average is very tight, with the fewest players being 96 in 1992, and the highest number of players at 111 in multiple years. So this average seemed stable. These numbers are displayed in the table below.



I then queried the database to return the fantasy points scored by players in their rookie year, 2nd year, 3rd year, … through their 11th year. (The summary information for this data is located in the table below.)

As you can see, there is a definite correlation between age and the number of running backs that perform in the NFL each year. As players get closer to their 11th year, they make up a much smaller percentage of the active running backs in the NFL. (Yes, I realize that didn’t require a lot of research to find that correlation, but it does help validate the data.)

More importantly for our study, there also appears to be a pretty strong correlation between the number of years in the league and the number of points scored. It appears that from a player’s rookie year to their 6th year in the league, their performance will increase, with the sharpest increases occurring between the second and third years. Once a running back reaches their 7th year in the league (age 30 for most players), even the running backs whoa re still playing begin to decline.



Note in the table above, I did remove 15 outlying data points. Our goal is to show the relationship between age and performance. Since we are trying to come up with the most accurate formula, we need to remove the noise to ensure we are as accurate as possible. Remember that there will always be individual performances that go against history, but these should not be included as part of our general formula. This modified our findings somewhat.

Who would ever draft a running back that would only have 10 carries is a season?

Good question. With this in mind, I also looked at a smaller subset of these running backs. This time, I only considered running backs that carried the ball an average of 5 times per game (80 carries per year). These are the running backs that are at least getting a reasonable opportunity to play, and at least could be considered part of a RBBC. While there is a little more variation in the numbers, the graph is very similar, as seen below.



Much like the findings in our previous study, there is a pretty significant boost in performance from the rookie year to the 2nd year in the league. What is interesting is that the next boost in performance for these running backs does not occur until year 5. However, after year 6, we do see the same decline as when we considered all running backs, with a pretty steep decline in year 7 (again, when most running backs are turning 30). What is interesting is the very consistent drop in the number of these running backs after year 4. This ranges from 15-20 players less every year.

Also, we can see that the average running back who follows this pattern will typically see a very nice increase in average scoring from their rookie year to their 6th year. For both 80+ carry running backs and 10+ carry running backs, there is a 50% increase (112-163 and 61-97) in points scored.

Sure, that sounds great in theory, but does it hold up when applied to real players?

Let’s take a look at Shaun Alexander as an example. Here is a list of Shaun’s fantasy points by year since his rookie year.



Compared to our baseline, Shaun had a much larger jump between his rookie and 2nd year. This can be primarily attributed to backing up Ricky Watters during his first year. Outside of that jump, the rest of the statistics hold up to the comparison. There is a relatively flat line in year 3 and 4. There is a double-digit jump in year 5 (granted this number is larger than our baseline, but still comparable). Then there is another reasonable jump in points going into year 6.

It will be disappointing for many Alexander owners if Shaun does experience a drop back to the 300 point range, as expected by our running back baseline numbers. Keep this in mind if you do have one of the first few picks in a new keeper draft.

Conclusions
So what does this data tell us?

I think it does very clearly show that there is a drop in performance of running backs once they hit the magical 30. While there are definitely exceptions to this rule (Emmitt Smith, Marcus Allen, Terry Allen, and Earnest Byner all had exceptions seasons after turning 30.), this data is pretty clear.

More importantly, I think this information very clearly gives us a formula that can be used when determining the value of a dynasty player. The factors that need to be taken into account when valuing a player are:

1. How many years has the player been in the league already?
2. How many years will the player remain on your team?
3. What is the projected value for their fist year statistics? (Rookies only.)
4. What was the player’s previous year statistics? (Veterans only.)

Of course, there are other variables such as free agency, offensive coordinators, and quarterback changes that can also have an effect on your running backs. However, once they established themselves, running backs tend to be fairly consistent.

Since 1990, 65% of all running backs who rushed for 1000 yards in a season repeated that feat at least once. In fact, since 1990 once a running back reaches 1000 yards, they will reach that plateau an average of 4 times in their career. (That number could also be much higher if you can spot and avoid the one-hit-wonders such as Napoloeon Kaufman, Rashaan Salaam, and Anthony Thomas.)

By combining that consistency with the numbers above with the experience to spot talented players accurately, you can make efficient and knowledgeable roster moves to help your team in even the most competitive dynasty league.

Using our example above, Shaun Alexander’s value may never be higher. Considering the return you could probably get in a trade, it might be worth exploring those possibilities.

This is not to say that you should simply give Alexander away, as even a 20% decrease in his numbers would be a good year for most other players. But his move is not as laughable as many other owners may think. It is this additional knowledge that keeps your team one step ahead.

Stay tuned for the next article in this series, where I will be looking at Wide receivers and their development over time. Let’s see if the Year 3 rule holds up as well as the 30-year-old rule for running backs.

Also, be sure to keep an eye out for the latest 4for4.com Keeper league cheat sheets. They’ll be published soon as well. As you might expect, they fully leverage a suite of fact-based analysis including age-performance measures.