Photo: Imagn Images

Most people will use last year’s statistics (or a three-year weighted average) as the starting point for their 2013 projections. From there, fantasy players modify those numbers up or down based on factors such as talent, key off-season changes, player development, risk of injury, etc. But in this article, I’m advocating that you use something besides last year’s numbers as your starting point.

There is a way to improve on last year’s numbers without introducing any subjective reasoning. When you base a player’s fantasy projections off of his fantasy stats from last year, you are implying that all fantasy points are created equally. But that’s not true: a player with 1100 yards and 5 touchdowns is different than a runner with 800 yards and 10 touchdowns.

Fantasy points come from rushing yards, rushing touchdowns, receptions, receiving yards, and receiving touchdowns. Since some of those variables are more consistent year to year than others, your starting fantasy projections should reflect that fact.

The Fine Print: How to Calculate Future Projections

There is a method that allows you to take certain metrics (such as rush attempts and yards per carry) to predict a separate variable (like future rushing yards). It’s called multivariate linear regression. If you’re a regression pro, great. If not, don’t sweat it — I won’t bore you with any details. Here’s the short version: I looked at the 600 running backs to finish in the top 40 in each season from 1997 to 2011. I then eliminated all players who did not play for the same team in the following season. I chose to use per-game statistics (pro-rated to 16 games) instead of year-end results to avoid having injuries complicated the data set (but I have removed from the sample every player who played in fewer than 10 games).

So what did the regression tell us about the five statistics that yield fantasy points? A regression informs you about both the “stickiness” of the projection — i.e., how easy it is to predict the future variable using the statistics we fed into the formula — and the best formula to make those projections. Loosely speaking, the R^2 number below tells us how easy that metric is to predict, and a higher number means that statistic is easier to predict. Without further ado, in ascending order of randomness, from least to most random, here is how to predict 2013 performance for each running back based on his 2012 statistics:

1) Rushing Yards (R^2 = 0.47). The best-fit formula to predict rushing yards is:

                   -731 + 3.73 * Rush Attempts + 180 * Yards/Rush

2) Receptions (R^2 = 0.42). The best-fit formula to predict receptions is:

                   11.1 + 0.39 * Receptions + 0.032 * Receiving Yards

3) Receiving Yards (R^2 = 0.38). The best-fit formula to predict receiving yards is:

                   83.7 + 1.65 * Receptions + 0.46 * Receiving Yards

4) Rushing Touchdowns (R^2 = 0.29). The best-fit formula to predict rushing touchdowns is:

                   0.1 + 0.0037 * Rushing Yards + 0.35 * Rushing Touchdowns

5) Receiving Touchdowns (R^2 = 0.23). The best-fit formula to predict receiving touchdowns is:

                   0.1 + 0.0022 * Receiving Yards + 0.25 * receiving Touchdowns

Using this formula, we can then come up with our starting projections.

2013 Starting Fantasy Projections

The table below shows the statistics for the top 40 fantasy running backs last year, with all players’ statistics pro-rated to 16 games. The final column, which is how the table is sorted, displays the number of projected fantasy points for each running back in a 0.5 PPR league. For an example of how little knowledge of the actual football world this projection system has, consider that it projects Knowshon Moreno to score 195 fantasy points this year. Obviously, you know better. The point is that this formula makes an excellent starting point for your projections. Then you can use things that you know about the non-statistical issues affecting particular players and teams to modify it. In Moreno’s case, you’d want to modify it way down, but in most instances, just a minor amount of tinkering is necessary (as of now).

Remember, the 2012 columns are all projected to 16 games, as is the projected 2013 points column. The system does not attempt to tell you who will get injured; it just makes a guess as to how many points the running back will score if he’s healthy all year.

Translating the Regression into English

This system loves Jamaal Charles, jumping him up to nearly the levels of Alfred Morris, Ray Rice, and C.J. Spiller. Why is that? Charles ran for only 5 touchdowns last year, an abnormally low number for a player with 1500 rushing yards. (Of the 84 running backs to rush for 1,500 yards in a season, only four rushed for five or fewer scores that season.) Meanwhile, Morris ran for 13 touchdowns with 1600 yards, so the regression expects fewer touchdowns for Morris and more for Charles. With respect to Ray Rice, he only scored more fantasy points than Charles because he had more receptions and more rushing touchdowns, but Charles’ edge in rushing yards is more sustainable. As for Spiller, the regression thinks his absurd yards per rush and yards per reception averages are less likely to hold up; historically speaking, players with more touches tend to hold their value best.

Meanwhile, Trent Richardson is a player that formula is bearish on. Much of his fantasy value was due to touchdowns and receptions, not rushing yards, making him a riskier pick. On the other hand, more is expected from Chris Johnson in 2013. He caught 36 passes, but they only resulted in 232 receiving yards and 0 touchdowns: if he catches 36 passes in 2013, he’d probably end up with closer to 300 yards and a score or two. In addition, history suggests a player with 1200 rushing yards is likely to rush for more than six touchdowns.

One player you might be surprised to see projected so optimistically is Darren McFadden. After all, he sure felt like a bust last year, and averaged a pitiful 3.3 yards per carry. But the regression tells us that rush attempts are extremely important, and last year McFadden had three games with over 25 carries and four more with 17 or more rushes (in only 12 games). He also caught 42 passes in 12 games, so this formula is impressed with a player so heavily involved with the offense. Throw in the fact that he ran for just 2 touchdowns, and he’s likely to be undervalued.

Adding the Human Element

These projections do not include any non-statistical information. At all. So you would want to use real-world knowledge to modify them. Fortunately, that’s where my Footballguys co-staffers are extremely valuable, and they’ll provide you with many articles this offseason explaining why players are under or overrated. For example, you might want to upgrade or downgrade these computer projections because of factors such as:

• The player’s previous history (prior to last season)
• His risk of injury
• His risk of being benched
• His offensive line and other surrounding personnel
• A change of teams or a change in coaching staff
• His attitude and intangibles
• Your intuition

With Charles, he has never been used as a goal line option, so he may continue to have a very low rushing-yards-to-rushing-touchdowns ratio. The computer might be down on Trent Richardson, but he was a rookie dealing with injuries last year; we should project improvement to counter some of the expected regression in his touchdown numbers. As for McFadden, the Raiders are moving away from the zone-blocking scheme that hampered him in 2012, providing even more reason for optimism there. On the other hand, with McFadden, assuming 16 games played is also a very optimistic projection. Allow me to run through the top 32 running backs according to our starting projections and tell you how I’d modify them (as of April 29). Your mileage may vary greatly.

Add Maurice Jones-Drew, David Wilson, Lamar Miller, Jonathan Stewart, and running backs for the Packers, Rams, Cardinals, Jets, and Steelers, and I come up with a top 32 for 0.5 PPR leagues that looks like this:

1. Adrian Peterson
2. Doug Martin
3. Marshawn Lynch
4. Arian Foster
5. Jamaal Charles
6. LeSean McCoy
7. C.J. Spiller
8. Ray Rice
9. Alfred Morris
10. Trent Richardson
11. Matt Forte
12. Stevan Ridley
13. Chris Johnson
14. Reggie Bush
15. DeMarco Murray
16. Maurice Jones-Drew
17. Darren McFadden
18. Steven Jackson
18. David Wilson
20. Frank Gore
21. Willis McGahee
22. Lamar Miller
23. Ryan Mathews
24. Darren Sproles
25. Vick Ballard
26. Jonathan Stewart
27. BenJarvus Green-Ellis
28. Eddie Lacy/Johnathan Franklin
29. Isaiah Pead/Zac Stacy
30. Rashard Mendenhall/Stepfan Taylor/Andre Ellington
31. Mike Goodson/Chris Ivory
32. Le'Veon Bell/Jonathan Dwyer