Fantasy football season is fast approaching, and it's just about time to start plotting your draft strategy. In most leagues, you need to take elite running backs early, but how do you choose which one? Last year, there were four RBs who went at the top of most drafts: Arian Foster, Chris Johnson, Adrian Peterson, and Ray Rice. Foster and Peterson missed some time with injuries, but were great when healthy. Rice was a sensation, especially in PPR leagues. And Johnson was a massive disappointment, especially in the first half of the season. Was there a way to predict these things? In particular, could we have foreseen Johnson's lackluster year?
Most specifically, I've been curious about how running backs perform after their best seasons. We all know that players tend to regress after a great year. It's rare for everything to go right two seasons in a row, and defenses tend to key around a player like Rice or CJ2K. But how much effect does this have on a player's stats? To find out, I looked at the top 30 fantasy seasons by an RB in the last 20 years: 1992-2011. Then I checked their performances in the year before (n-1), year after (n+1), two years after (n+2), and three years after (n+3).
For all seasons, I used the following scoring system:
1 point per 10 rush yards
1 point per 10 receiving yards
6 points per TD (any kind)
0.5 points per reception
That's a pretty simple fantasy scoring metric. Notably, it's a ½-PPR system, with no fumble penalties, and I have completely omitted two-point conversions and passing statistics. Those are fluke plays, not really what we're looking at here. I'm listing total points per season, but if you prefer per-game averages, 240 is 15 points per game, 320 is 20 points per week, and 400 is 25 per week.
The "average" line does not include years when a player was in college (Edgerrin James in 1998) or retired (including Ricky Williams' temporary absence in 2004), nor years in the future (Ray Rice and Arian Foster). What the average shows, though — and pretty clearly — is that great running backs seldom continue to deliver at a high level if they're more than one year removed from a great season. The n-1 and n+1 years look pretty good, the exceptions being related to playing time the year before or injuries the year after. But there's a significant drop-off when we get to n+2.
The top n+2 seasons:
1. Marshall Faulk, 1998-2000 (415)
2. Marshall Faulk, 1999-2001 (382)
3. Emmitt Smith, 1992-1994 (340)
4. LaDainian Tomlinson, 2005-2007 (333)
5. LaDainian Tomlinson, 2003-2005 (329)
6. LaDainian Tomlinson, 2002-2004 (312)
7. Emmitt Smith, 1995-1997 (259)
8. LaDainian Tomlinson, 2006-2008 (252)
9. Marshall Faulk, 2000-2002 (249)
10. Steven Jackson, 2006-2008 (210)
Those are the only n+2 years that beat the 207 average. So unless you're drafting Marshall Faulk or LaDainian Tomlinson, you probably shouldn't choose a player more than one year after his breakout season. And once those guys start to fade (Faulk in '02, Smith in '96, L.T. in '08), they're almost always finished as players who merit a first-round pick in fantasy.
Out of the 30 seasons included, only eight featured improvement from n+1 to n+2, and some of those eight are jokes: Terrell Davis scoring 42 fantasy points instead of 37 when he was injured in '99 and '00, or Jamal Anderson scoring 198 (which is not worth a high draft pick) instead of 10 because he didn't get hurt in Week 1. The only legit increases were Faulk, Smith, and Tomlinson, who never really regressed in the first place, plus a couple guys who returned from injury but didn't have great years. Effectively, no one on this list went from being great to getting hurt or just being above average, and then ever returned to being great. It basically doesn't happen.
From 1992-2011, there were 44 running back seasons which scored at least 300 fantasy points in this system, about two per season. Among players who had ever hit 300, none of them dropped below 250 and returned to 300 in the future. Ever.
There are some exceptions if we use a slightly lower standard. Tiki Barber scored 299 in 2002, dropped to 220, then had three more great seasons before retiring (326, 332, 272). Edgerrin James was over 300 each of his first two seasons, injured in 2001, and returned to 283 and 290 in '04 and '05, respectively. There is also a distinction if we separate the averages by age. Of the 30 seasons above, 15 were players 25 or younger, and the other 15 by RBs 26 or older.
For players 25 and younger, we're probably seeing something close to a normal regression to the mean. No one produces 350 fantasy points every year, and 230-260 is still very good. The real takeaway is for players who have already hit their mid-20s. If they didn't have a big season the year before, they're probably not going anywhere good. The older players' n+3 seasons included three who were retired, four who missed most of the season with injuries, four part-time players who scored under 200 points and another at 202, plus the last gasps from Faulk, Smith, and Tomlinson, all between 249-254. Don't draft old RBs who didn't light it up the year before, or at least not in the first two rounds.
This actually is not the conclusion I was expecting. It seemed to me that elite runners often go through a down year following their breakout campaign, because of overwork or defensive attention or strength of schedule (or whatever else), and then rebound in the n+2 season. My research does not bear that out. It may apply to young runners who still had good n+1 years, but almost never to older players who have already peaked. The old guys may rebound following a return from injury or a trade to a new team, but never again to elite status, not to deserving a high draft pick in fantasy.
Of course, where you draw the cutoff line for a first-round RB depends on the size of your league (and differences in scoring system), but 250 is probably a pretty reasonable standard in the system I'm using here. Over the past 20 years, 121 RBs have met that figure, about six per season. Number of RBs with 250+ fantasy points:
It's probably safe to say that if you're investing a first-round draft pick on an RB, and your league is less than 12 people, you're hoping for 250+ fantasy points as a return. So how do you use previous production to decide who's worthy? Specifically, how do you do that in 2012?
I examined 20 RBs who might merit a high draft selection, not including any rookies (or other first-year starters, including DeMarco Murray). We're using past performance to predict future production, so we need players with a track record. We'll consider Ray Rice, LeSean McCoy, Maurice Jones-Drew, Arian Foster, Darren Sproles, Marshawn Lynch, Michael Turner, Ryan Mathews, Steven Jackson, Reggie Bush, Chris Johnson, Matt Forte, Adrian Peterson, Fred Jackson, Frank Gore, Shonn Greene, Willis McGahee, Ahmad Bradshaw, Darren McFadden, and Jamaal Charles.
For each running back, I calculated fantasy points using the same system explained earlier, and looked at: best year (BY), how many seasons ago that was (-BY), last 250-point season (L250), last 300-point season (L300), and total points for each of the last two years (n-1 and n-2).
A few of these are older players who are probably starting to decline, and some of them young players on the way up. We're using a player's best season as the frame of reference, but for several of the names above, that best season probably hasn't happened yet. So how do we sort these 20 RBs?
Apart from injuries and suspensions, the ones who scare me are those whose 2011 were below 230 and whose best year came more than two seasons ago: Turner, Steven Jackson, Johnson, Forte, Peterson, Gore, and McGahee. I'll add Fred Jackson to that list, because he's a 31-year-old RB who missed six games in 2011 and has never topped 237 carries. If that's your RB1, I hope you're starting Aaron Rodgers and Calvin Johnson.
Everyone is down on Michael Turner. He's a value pick at this point, someone you add if he drops too far. I don't think anyone expects him to replicate 2008 (1,700 yds, 17 TDs), but he's always produced as a starter. Steven Jackson is a capable running back, and valuable in PPR, but you want your top RB to be a game-breaker. Jackson's a solid 12-points-a-week guy, but he gets hurt a lot, and if he's the best player on your fantasy team, you're not going to win your league. Jackson is a great RB2, but not a quality RB1. If you pick him as your top RB, you're just patching a hole, when you could be taking an elite QB or WR, maybe grabbing the first game-breaking TE off the board.
Chris Johnson still looks like a first-rounder in most leagues. A shooting star in 2009, he was good in 2010 and a huge disappointment last season. His total fantasy points by year: 230, 372, 255, 199. This year will be Johnson's n+3 season. He's got great speed and he's an asset in the passing game. No RB has ever had a season as good as Johnson's 2009 and not had another superb season, a 20 points per game kind of season. He did okay at the end of 2011, and his poor performance early may have been linked to an extended holdout. But at this point, I wouldn't bet on CJ. If he falls to the second or third round, yes, of course, grab him and do a victory dance. Even in the late first round, you get someone who's probably worth about 225 (14/gm) and has huge upside. But in the first five or six picks, I'd say you're reaching for a player whose magic only lasted one year. I'd rather roll the dice on McFadden or Mathews staying healthy.
Matt Forte will go in the first round of most drafts. In four seasons, he's had two very good years (about 250 pts), one pretty poor year (164), and one very good year shortened by injury (199). Considering the Bears' acquisition of Michael Bush, it seems very unlikely Forte will top 250 this season. The track record for guys whose best season was four years earlier is really bad, unless you're Marshall Faulk, Emmitt Smith, or LaDainian Tomlinson. Forte's a nice player, but he's not nearly of that caliber. Based on previous production, I see Forte as a second-round value. Someone else will draft him before it's worth it for you to take the gamble. If you don't play PPR, don't touch the guy.
Adrian Peterson, a reliable fantasy stud coming off a serious injury, is a lottery ticket. He's probably not worth an early-round pick at this point. I'd be surprised if he earns 200 fantasy points this year, and shocked if he tops 250. I know some people are drafting him with the assumption he'll be available for the fantasy playoffs, but I don't like that kind of gamble on a guy trying to come back quickly from knee surgery. That usually takes more than one year, and you may not make it to the playoffs in the first place if you're waiting for your best player to get healthy. AP is getting drafted based on name recognition and false hope.
Frank Gore, a high-quality RB who can't stay on the field, could also be considered a lottery ticket. The odds of getting 15 or 16 healthy, high-productivity games out of Gore are probably about the same as the odds of actually winning the lottery. This will be his n+6 campaign, and I think I'm done with The Inconvenient Truth as an RB1.
I'm not sure why I included Willis McGahee in this exercise. He's obviously not an elite RB, but he can still play and he gets the ball. Grab him as an RB2 or flex option if he slips to the middle rounds.
The one other player I want to comment on is the ultimate lottery ticket, Jamaal Charles. As a part-time back in 2009 and 2010, he scored 210 and 264 fantasy points, respectively. He averaged about six yards a carry, and was a first-round pick in virtually every fantasy league last season. He got hurt in Week 2 and missed the rest of the season. Charles is a small (200 lbs) RB who's never been a full-time starter, coming off a serious injury and sharing time with Peyton Hillis. If he cracks 250 points, I will buy a hat and eat it. Charles has nice upside, but you can't afford to throw away a first-round pick on a guy whose realistic ceiling is probably close to 200 fantasy points. If you draft Charles, you also need to handcuff with Hillis, blowing another mid-level draft pick. I wouldn't seriously look at him until the bottom of the second round, maybe even late third in smaller leagues.
Leave a Comment