NextGen Stats - Running Back Breakdowns and Takeaways

With five weeks of play in the books and things starting to settle a bit, it's time to get back to studying some of the rushers around the league and see how they have performed through (almost) the first third of the season. Here's to hope that we don't have any more setbacks with regards to COVID and we keep getting games worth number-crunching going forward!

To gain the biggest edge in your fantasy football league, it's necessary to understand how to apply the advanced statistics being used in sports nowadays. Back in the day, it was all about wins and losses, passing yards, and touchdowns scored. It's not that those stats are now worthless, they just don't offer enough information to savvy analysts. While football is still in its infancy compared to baseball in terms of analytics, the evolution the sport has seen lately in those terms is notable.

Each week, I'll be tackling NFL's Next Gen Stats, bringing you data from the previous week's games with notable takeaways you should consider when assessing fantasy players for the upcoming week. In case you're new to the series, or Next Gen Stats altogether, I recommend you read our NGS-primer. Now, let's get to the data!

The Running Game Is Alive!

One of the most important concepts to consider when analyzing players are Air Yards. The metrics around it are key to know who is really over-performing or under-performing among receivers and passers, but it doesn't have much to do with rushers. For this last group, which mostly features on the ground, we can look at time, speed, and efficiency metrics.

Today, I'll present each of the stats from the NFL's advanced metrics site, its correlation with rushers-fantasy points, and a list of leaders and trailers in each category along with some notes and takeaways on both the players' and the metrics' impact on fantasy football as a whole.

For the 2020 season, the NFL has introduced the concepts of:

• Expected Rushing Yards (xRY; How many rushing yards is a ball-carrier expected to gain on a given carry based on the relative location, speed and direction of blockers and defenders?)
• Rushing Yards Over Expectation (RYOE; The difference between actual rushing yards and expected rushing yards on an individual play or series of plays)
• Rush Pct Over Expected (ROE%; The percentage of runs where a ball-carrier gained more yards than expected)

I will only focus on fantasy production as pure rushers, eliminating the pass-catching element from their game. This will concentrate entirely on their total rushing yardage and rushing touchdowns in terms of the fantasy points per game numbers shown (labeled ruFP/G). I will also include an extra column, "ruFP/15Att", which is accounts for the fantasy points a rusher is getting per 15 rushing attempts, which would be considered an RB1 workload on average and allows us to know how different players in different roles would be doing if given the same opportunities.

So let's dive in. Note: The cutoff is set at 25 rushing attempts.

Efficiency

Correlation with Rushing Fantasy Points: negative-5%

Leaders and Trailers:

Leaderboard Notes:

• Remember, the most "efficient" a rusher is, the least amount of yards he "wastes" going headfirst toward the opposition goal line in a straight route.
• With that in mind, and through five weeks of play, we have gone from seven players with an EFF below 3.00 to just two of them among the 49 qualified RBs: Sony Michel and Gus Edwards.
• Although the correlation between EFF and ruFP is negative (yet not very significant) both of them are finding relative success so far with averages of 13.4 and 8.5 ruFP on the ground per 15 attempts... That being said, their actual usage has both of them below 8 ruFP per game...
• Only three players (Nick Chubb, Aaron Jones, and Miles Sanders) are above 10 ruFP per game while approaching the running game in the straightest, simplest of ways. Of the 12 RBs with marks below 3.6 EFF, only them are averaging those ruFP.

• And at the other end of the leaderboard sits the man making that negative correlation look too small for what he's doing. Derrick Henry is third in ruFP per game, and dead-last in EFF. Makes sense given what we know so far...
• ...or not. Actually, four of the next five RBs in the bottom part of the leaderboard to rank above Henry are averaging a paltry 3.9 ruFP per game and right in the middle of them, Josh Jacobs is at a great 13.5 average. As you can see, the correlation is nonexistent and EFF is more descriptive than predictive, to say the least.

• Just a final note to disprove this stat as a fantasy-relevant one. The five-best RBs in ruFP per game (all above 15 ruFP/G) have the following EFF marks: 4.13, 3.69, 4.83, 3.31, 3.20. As random as it gets.

Percentage of Stacked Boxes Faced

Correlation with Rushing Fantasy Points: negative-14%

Leaders and Trailers:

Leaderboard Notes:

• The correlation between stacked boxes and fantasy points is going stronger than it usually does. It normally doesn't affect fantasy performances a lot, but this year through five weeks of play it sits at a reasonably-high negative-14%. Nothing incredible, but worth noting.
• Don't throw away stacked boxes completely given that there normally is an almost non-existent relationship, but keep in mind that it is heavily related to the role a player has on his offense and to what teams expect from him.

• The top part of the leaderboard is super-damaging for those ranking high in stacked boxes faced so far this season. Of the top-8 players in 8+D% (all above 27.5%), all of them are averaging fewer than 8 ruFP per game, and only three are above 5 ruFP/G.
• Usage is not a factor in that equation. Even pro-rating the ruFP average to a per-15-attempts basis, only one of those eight players (Sony Michel; 13.4 ruFP/15ruAtt) would break the 10-ruFP/15ruAtt barrier. Other workhorses like Myles Gaskin and Mike Davis would be at just 7.2 and 9.1 ruFP/15ruAtt...
• As is always the case with lowly-related stats, the bottom of the leaderboard is pretty much showing the opposite trend. Of the bottom-11 rushers, all having faced stacked boxes less than 10% of the time, only two (Alvin Kamara and Miles Sanders) are averaging more than 10 ruFP/G... Again, on a per-15ruAtt basis, just those two and Brian Hill would break the 10 ruFP/G barrier.

• Flipping the chart order to the ruFP/G column, the relationship (or lack of it) shows: Dalvin Cook is averaging the second-most ruFP/G at 18.2 while facing 8+D on 18.5% of his attempts; Miles Sanders is averaging 12.4 ruFP/G (9th-most) yet he's faced 8+D on 1.6% of his carries.

Average Time Behind The Line Of Scrimmage

Correlation with Rushing Fantasy Points: 22%

Leaders and Trailers:

Leaderboard Notes:

• In the same vein as what is going on with the relationship between stacked boxes and fantasy points, the TLOS stat is showing early strong signs of being tied to fantasy production. Surprisingly knowing how, once more, it wasn't so high in 2019. It is still early with just five weeks on the books, so we'll see how this develops as we get deeper into the season.
• The difference in ruFP/G between the top-15 players in TLOS (lower values) and the bottom-15 (higher TLOS) amounts to 2.9 fantasy points. Not incredibly high, but notable enough, thus the 22% correlation.
• That difference if considering ruFP/15ruAtt on a pro-rated usage for all players is reduced to just 2.1 rushing fantasy points.

• When we last tackle running backs, six of them were averaging more than 3.0 seconds behind the LOS. It's down to two after Week 5, and Melvin Gordon III was this close to missing the cut at 3.02 seconds. Jerick McKinnon, though, is on a league of his own.
• McKinnon has seen a very reduced use, with just 35 carries over five games (7 per game). Pro-rated to 15 attempts per game he'd be leading the group proving the 22%-correlation as he'd be averaging 16 ruFP per game on that 15-carry usage.

• All but one player averaging 10+ ruFP/G are taking more than 2.74 seconds to cross the LOS. The outlier is Ezekiel Elliott, whose 2.58 TLOS is the lowest in that group yet he's still at a 13.3 ruFP/G average on the season.
• Frank Gore seems to be running against the clock... to no success at all. He's getting to the LOS in just 2.6 seconds but he's averaging a paltry 4.1 ruFP/G even on one of the heavier loads of all league RBs.
• David Johnson is taking virtually the same time (2.57 TLOS) but he's more than doubling Gore's outcomes at 8.4 ruFP/G. Talk about a non-predictive stat...

ATT & YDS & Y/A & TD

Correlation with Rushing Fantasy Points: 61% / 74% / 28% / 85%

Leaders and Trailers:

Leaderboard Notes:

• With five weeks gone, there are still two players averaging 6+ YPC, and as many as 12 with 5+ YPC. That will go down with time (and five of those 12 players have 35 or fewer rushing attempts...), but with more than a quarter of the season played out, it's not bad to have so many high-averagers in the league.
• Raheem Mostert and Sony Michel, both with a low number of attempts so far, would both jump to over 13 ruFP/G if given 15 carries per game... and that is having just one touchdown on the season! Give those men the rock and let them run wild with it!

• Take a look at the bottom of the leaderboard, and if you have any of those bottom-five rushers in your roster get an instant depression for free. They are the only RBs to meet the 25-attempt minimum while posting fewer than 3.5 Y/A... and all of them are averaging fewer than 5.5 ruFP/G. Yikes. Not even pro-rating their usage to 15 attempts per game do we get better results, with Hines leading the pack at just 8.1 ruFP/15ruAtt...
• Obviously, the most concerning cases among those five are the ones of Joshua Kelley and Frank Gore. Both have 60+ rushing attempts so far yet they're pretty much horrific in fantasy terms with putrid 5.3 and 4.1 ruFP/G averages respectively.

• Tame your expectations when it comes to playing both Derrick Henry and Josh Jacobs. Both of them are averaging a paltry 3.7 Y/A and only the touchdowns are keeping them at the top of the fantasy leaderboards. Even on super-heavy workloads (both over 100 rushing attempts through W5) their low Y/A averages aren't helping matters, but the 4 and 5 TDs added sweet fantasy points to their season-overall tallies. If the TD-supply is cut out, though, watch out for heavy regression.
• That's precisely the opposite of what is going on with Mostert and Michel, as they've been good on low-usage while not relying almost at all on scoring plays.

• Something similar will happen at the bottom of the leaderboard. Of last year's qualified rushers, only two of them finished below 3.5 YPC while there are currently five players below the 2.0 mark. Don't expect that to hold for long--the averages will go up, or the players get out of consideration with fewer carries than to qualify for the leaderboard.

• Dalvin Cook has played out of his mind. He's the only player with more than 80 rushing attempts averaging over 15 ruFP/G at a monster 18.2 with 7 touchdowns on the ground already (!). Of the three player with that amount of carries over the year, he's on his own world averaging 5.3 Y/A to the other three 3.6, 3.7, and 3.7. Also, if normalized to ruFP/15ruAtt, Cook would be averaging 14.8 fantasy points compared to the other three 7.3-to-9.6 range. Ridiculous.

YDS & RYOE & Y/A & RYOE/A & ROE%

Correlation with Rushing Fantasy Points: 74% / 27% / 28% / 17% / 15%

Leaders and Trailers:

Leaderboard Notes:

• Starting this year, the NFL has introduced the concept of Rushing Yards Over Expectation. Basically, it comes down to the extra yards a rusher gained (or lost) in a given play given the context of that play (down, distance, etc...).
• The RYOE metric indicates how many extra yards the rusher gained/lost on his carries over/under those he actually rushed for. Ex: YDS = 100, RYOE = 10, then the rusher was expected to rush for 90 yards, but overperformed that mark by 10 (RYOE) thus getting to 100.
• The RYOE/A metric indicates the average extra yards gained/lost per rushing attempt. Ex: Y/A = 0.5, RYOE/A = -1.5, then the rusher was expected to rush for 2.0 Y/A but underperformed that mark by -1.5 (RYOE) thus getting 0.5 Y/A.
• The ROE% metric indicates the percentage of rushing attempts the rusher exceeded the expected yards (gained more yards than expected). Ex: ATT = 10, ROE% = 50.0, then the rusher gained more yards than expected in 5 of 10 (50.0%) of his carries.

• Both of our earlier-discussed friends Raheem Mostert and Sony Michel have been balling. They are one full yard above no. 3 Nick Chubb in RYOE/A, exceeding the expectations of their carries by 2.9 and 2.8 yards per attempt respectively. The distance between them and Chubb is the same as that between Chubb and no. 12 (!) Brian Hill.
• Removing the yards over expectation from their actual Y/A, Mostert would rank 26th of 49 in Y/A, and Michel would be 32nd. Quite a bump down from the top-two spots, isn't it?

• Looking at what would be the expected Y/A leaderboard (not included in the chart), Miles Sanders would lead the NFL with 5.0 eY/A, which is actually just 0.1 yards below his current 5.1 Y/A through five weeks. Mark Ingram II (4.9 eY/A) and Matt Breida (4.9) trail Sanders, but they have underperformed with -0.3 and -1.2 RYOE/A this season.
• That -1.2 RYOE/A mark by Matt Breida is actually the second-worst among all 49 qualified running backs, only "bested" by Devonta Freeman's -1.5. Now, go wonder why the hell the Giants signed him to fill Barkley's role--are they under-the-radar tanking?
• Shout-out to Frank Gore, who once more appears in a bad place in our leaderboards and gets another mention in this week's column. Among rushers with 60+ carries through W5, Gore ranks second-worst (Kenyan Drake) in RYOE/A at a putrid -0.8. Also, drop Drake and if you still want Cardinals' backfield shares after that then feel free to pursue Chase Edmonds if anything.
• Another reminder: watch out for regression coming Derrick Henry's way. He's averaging 15.4 ruFP/G, sure, but he's needed 4 TDs to reach that mark and is rushing -0.6 RYOE/A in his four games and 101 carries...

• I have been saying that Kareem Hunt is a better player overall than Nick Chubb since the Browns added the former to the roster. See, Hunt is being less efficient than Chubb (1.2 RYOE/A to 1.8), but he's getting more yards than expected on 54.3% of his carries to Chubb's 40.4% mark. That difference amounts to two carries per game in which he exceeds the expected rushing yards over Chubb's, and he's doing so while already having logged a fat-enough 70 rushing attempts.
• Of the true workhorses through Week 5 (75+ rushing attempts), Ezekiel Elliott has posted the highest ROE% at 49.4% (virtually half of his carries go for more yards than expected). Kenyan Drake (85 ruAtt) and Clyde Edwards-Helaire (81) have been trusted a lot by their teams on the running game, but they have underwhelmed with just 28.4% and 32.1% of their carries going for more yards than expected.
• Rookie Cam Akers can't do more than you would expect from him with a league-trailing 23.1% of his carries exceeding yardage expectations. Same for Joshua Keely (27%) and Adrian Peterson (29.6). Oh, wait, perhaps AP doesn't belong in this group of youngsters...

That's it for today. Until we meet again next week, I hope you can crush your waiver wire, set up the best possible lineup, and get ready for another weekend full of fireworks!

Download Our Free News & Alerts Mobile App

Win Big With RotoBaller

More Fantasy Football Analysis