Subtraction Of Mendenhall Statistically A Good Thing For Steelers Pass Protection

Matthew Marczi

Earlier this week, we have devoted two articles to the pass blocking prowess of the Pittsburgh Steelers’ running backs last season, and how it will be a chore for Le’Veon Bell to get up to speed with the rest of the group.

However, it is fair to note that the 2012 season was a recent high water mark for the Steelers backfield with respect to pass protection. Even Isaac Redman, who was the best in pass protection in the league a year ago, had his troubles in 2011, when he saw 77 snaps as a pass blocker.

Last season, Redman gave up just one hurry all season—in the last game of the year, in fact. In 2011, however, he gave up two sacks, one hit, and three hurries, which equates to a Pass Blocking Efficiency of 93.5. Going by that metric, Redman’s PBE would be nearly half a percentage below the mean from a year ago, when the league-wide PBE was 93.94.

One player who did not do the Steelers many favors in this area over his short tenure with the team was Rashard Mendenhall. Drafted in the first round in 2008, he missed most of his rookie season before winning the starting job in 2009, and then went on to miss most of the 2012 season.

As previously noted, Mendenhall played just four snaps in pass protection last season, and did so without allowing any pressure. But that was not indicative of his overall game in this area. The numbers between 2009 and 2011 will be more telling.

Conveniently, Pro Football Focus has already done some of the legwork in this area, as they have compiled a piece that looks at the pass blocking efficiency of backs league-wide over the three-year period between 2009 and 2011. The results? Mendenhall had the ninth-worst Pass Blocking Efficiency over that period of all backs with at least 100 snaps in pass protection.

While he can take solace in the fact that names such as Adrian Peterson and Chris Johnson ranked eighth and sixth, respectively, it is certainly a dubious distinction to be included in a bottom ten list of nearly any kind.

Between 2009 and 2011, Mendenhall played 148 snaps in pass protection. In those 148 snaps, he surrendered 14 total pressures, including four sacks and three hits. That is a success rate of just 90.54, which, by last year’s standards, is easily below league average.

In terms of Pass Blocking Efficiency, which, again, is a Pro Football Focus metric that takes into account the greater severity of sacks by counting hits and hurries as just three-fourths of a pressure, Mendenhall scored 92.23. That, too, again, is well below the average numbers that were put up league-wide a season ago, when the mean PBE came in at 93.94.

Interestingly, however, Mendenhall actually had a very representative season in 2011, when he allowed just one hit and one hurry on 42 snaps in pass protection. That earned him a PBE of 96.4, which did place him in the top 20 of backs with at least 40 snaps in pass protection that year. 2011 was a fine season; it was the prior two seasons in which he really struggled.

Taking a look back at 2009, Mendenhall was asked to pass protect more than he should have, playing over a quarter of his passing snaps in pass protection. In his 266 passing snaps, 68 were as a blocker.

In those 68 snaps, Mendenhall surrendered seven pressures, which was a success rate of 89.7. In terms of Pass Blocking Efficiency, he rated at 91.2, as three of his seven pressures were sacks.

2010 was an even worse showing, as he surrendered five pressures in 30 less snaps in pass protection. That was a success rate of 86.8. With one of his five pressures being sacks, his Pass Blocking Efficiency came in at 89.5.

Neither of those figures for either season cracked the top 50 at his position in those respective years. As relayed previously, the Arizona Cardinals had one of the least successful backfields a season ago in pass protection, so it will be interesting to see which Rashard Mendenhall they will be getting in 2013, if he will be asked to pass protect much at all.

Of course, Mendenhall is not part of the Steelers in 2013, and he played only a marginal role in 2012, so his absence does not do much of anything to reshape Pittsburgh’s narrative. Both Redman and Dwyer showed their prowess in pass protection last season, and both are still on the roster. Baron Batch and Will Johnson also held their own in their first seasons as active contributors. How many of them will make it onto the 53-man roster, and will they be able to live up to their own performances from a year ago?

About the Author

Matthew Marczi

Passionate Steelers fan with a bit of writing ability. Connoisseur of loud music. Follow me on Twitter @mmarczi.

  • woundedvet

    Subtraction of Mendy is statistically a good thing for the average of the roster’s intelligence, That went up.

  • Madi

    I’m not huge into stats, but it seems like what I’m reading is that overall his pass protection stats are poor, and therefore we are better off without him in that area.
    However, those same stats show he was only statistically poor his first two years, and in his third year he was good enough for 7th best in the league. Then last year he was perfect on a very small number of snaps. The last two years seem less like an anomaly to me, and more like an improvement.
    We didn’t dump 2009 Rashard Mendenhall, we dumped present day, good pass protector Rashard Mendenhall.
    I don’t think we’ll hurt too much in that area with him off the team, but only because he wouldn’t have seen the field anyway. And I trust Redman back there, although I’d rather have Mendenhall out there on passing downs for running and receiving reasons. Bell has the potential, but it’s very unlikely he’ll be as good at blocking as Mendenhall this year, if ever. It’s not something I’m sweating about, but it seems to me this was subtraction by subtraction, not addition by subtraction.

  • Rob H

    In my opinion, even though pass protection is important, unless they are a revolving door, the most talented “running back” is going to start, I think all of the talk by Wilson and others is just to keep all of them working hard to get better at it. If Bell isn’t officially named the starter before the third preseason game, I’ll be shocked, the other backs on the roster aren’t even in the same zip code when it comes to his overall ability.

  • Matthew Marczi

    You are misreading some things. The figures you see in that link are for the period between 2009 and 2011. He barely cracked the top 20, as I wrote, in 2011, and that was with finagling the numbers to include only backs with 40 or more snaps in pass protection. There were 18 players in 2011 with at least 10 snaps that did not give up a pressure. If you really want to include every back who took a snap in pass protection, Mendenhall actually ranked 75th. If you limit it to players with 10 snaps, he ranks 39th. Limit it to 20 snaps and he still ranks 30th. And having 4 snaps in pass protection in 2012 does not indicate anything at all.

  • Madi

    I’m not claiming his going 4 for 4 in 2012 means anything, so let’s not bump him down to 150th because 100 backs took one snap and didn’t allow a sack on it. I was just including 2012 to be thorough and not skip a year. It’s fair to limit it to backs with the number of snaps he took, and that got him to just inside 20th, it sounds like. I can’t see that stat so I’ll take your word for it.

    But that’s for that year. I don’t know if backs were especially strong that year or what, but if you project the 96.4 he earned in 2011 over three years using the three years PFF selected, then his 96.4 becomes good enough for 7th best in the league. You said it yourself: he struggled one year, struggled the next, and did well after that. But your conclusion mixes the three years together, as if his good performance was spread out over the first two years, and the overall impression is only slightly better than those not-so-good years. Instead, I look at it as a timeline in which he grew as a player. It’s not uncommon for young backs to struggle in pass protection and then get better. If you have a receiver with 20 drops his rookie year and no drops his second year, that’s 10 drops a year. But going into year three, do you really think of him as a 10 drop a year guy? After starting 16 games and finishing with no drops? I’m considering him a pretty sure-handed receiver at that point, not someone likely to drop 10 balls.

  • Matthew Marczi

    I think you are giving him too much credit to accept at face value that 2011 was a sign of clear growth rather than an outlier. In that sense, projecting a player’s best year over a three-year stretch is really kind of meaningless. By that same token, if you project his 2010 season over that same three-year stretch, it’s good for second worst in the league. Another perspective would be to look at Pro Football Focus’ actual grades. In 2011, the site gave Mendenhall a score of 1.4 for his blocking, which was 25th in the league. He graded out at -2.0 in 2010, and -3.0 in 2009. Those of course place him near the bottom of the league. (In case you were wondering, he gave up two hurries in four snaps in 2008).

    A better way to look at it is that even with an above average season in 2011, his career average was still well below the mean. Believe it or not, I actually like Rashard Mendenhall, and outside of financial concerns, I would have had no problem if the Steelers re-signed him. I’ve always liked him as a runner and as a receiver and felt that a lot of his critics were unjustified. I hope that he has success in Arizona and has the opportunity to prove that he can be a Pro Bowl-caliber running back. But the fact of the matter is that until I see him duplicate the success that he had in 2011, I have to look at it – and thus write about it – relative to the rest of his career, as somewhat of an aberration, and not as a glorious new chapter in pass blocking efficiency.

    I’m not saying it’s impossible, but pass protection numbers have the potential to vary significantly from year to year, especially from players who only take 50 or less snaps in pass protection, and that needs to be taken into consideration. And honestly, the fact that he is going to the Cardinals, with a weak offensive line, a stable of running backs already poor in pass protection, and a quarterback that has a proclivity to hold on to the ball too long, I don’t like his chances of improving. But even so, Mendenhall’s absolute best year was still worse than the team’s average from 2012, so the move was still addition by subtraction.

  • Madi

    Ugh, I had a much longer response typed and I lost it. So I’m only gonna put about 30 more seconds into this…

    Why would we make a projection from 2010 instead of 2011? That completely ignores the concept of time. I didn’t cherrypick, I used the most recent evidence. And to simply group years together because it’s easier to come up with stats that way is to completely lose any kind of improvement or decline in a guy’s play. By the same logic, we can all expect Keenan Lewis to go back to being barely worth a roster spot, because if you average out his entire career (complete garbage for two years, serviceable depth player for one year, good starter for one year) it probably works out to that of an okay #3 or #4, instead of the strong #2 and #1 capable guy he proved he was last year. What do you think his PFF numbers would be if he’d been put on the field back when the only reason he was even still on the team was his draft status? Bad enough to drag his average into the mud, I can tell you that. Of course, no one would do that because that’s crazy and his $25 million contract is considered a steal. The people that agree with me on that look at Lewis’ career in the order that it happened, not all scrambled up to a statistical average.

    I concede that Rashard’s best year is 0.3% lower than the 2012 Steelers’ backs. If you concede that Rashard’s same performance on the same chart puts him ahead of all 31 other teams. That’s not too shabby, right? Better than the whole rest of the NFL?

    Of course, I don’t really think he’s better than the whole rest of the NFL, but that’s what numbers can do for you. And what if you were to take into account the rest of Mendenhall’s game? As a runner and receiver he offers much more than Redman and Dwyer, even on passing plays. I’m sure PFF probably has a stat that proves me wrong, but you and I both know he’s better. He’s the one you want to dump the ball off to, given the choice. Don’t you think an entire season of him instead of them – getting a few extra yards here and there, a couple of big plays – would make up for the 0.3%? And what is 0.3%, anyway, like 1 extra hurry every 80 snaps? And isn’t that incredible 96.7% the team sported in 2012 an aberration? Or is that legit due to a genuine improvement? How do you know? What is it when you average it out over the last three years? (Don’t really look into that).

    Dammit, that was more than 30 seconds, and it was entirely different from my first post. Oh well.

  • Matthew Marczi

    Please consider this my final, all-purpose response. To be frank, some of the answers to your questions can be found in the previous two articles I’ve written about the Steelers’ backs in pass protection.

    I believe that I made it very clear, even in the title, that this article exclusively pertains to pass protection, so whether or not Mendenhall would be a better first option running or catching than other backs on the roster is irrelevant to my point. Nothing in this article is critical in any way of Mendenhall’s abilities as a runner or receiver. In fact, rushing and receiving have absolutely nothing whatsoever to do with this article.

    The truth is that a lot of starting running backs are not the most adept pass protectors (hence why Chris Johnson, Adrian Peterson, and others were on that list, and why Mewelde Moore saw more snaps in pass protection than Mendenhall; and after him, Redman). It is a very frequent occurrence that third down backs see more snaps in pass protection than starting running backs, and there is good reason for that.

    Of course we would not project a player’s average based on his worst season. I cited that as an example to point out the equal absurdity of citing a player’s best season as his average. 2011 may be his most recent representative body of work, but that does not dismiss the rest of his career, which you seem inclined to do.

    42 pass protection snaps is also simply too small a sample size to identify clear improvement and differentiate that from an anomaly. For example, if you add two more hurries to his 2011 season, it accounts for nearly a 4% swing in PBE, and a greater than 5% swing in success rate.

    There are such things as aberrations. Using Michael Turner as an example, his PBE in 2012 was 97.2. In 2011 it was 99. But in 2010 it was 93.5. So did he get better, or was 2010 the outlier? Well, in 2009 his PBE was 98. You can jump to the conclusion that the data dictates Mendenhall improved by about 6% from 09-10 to 11, but I will take the more cautious approach by not making that assumption.

    Comparing individual backs to whole teams is not really impressive, or indicative of anything. Mendenhall’s best season may have ranked him fourth (not third) in PBE by team a year ago (his success rate would place him fifth), but you would have to apply that equally to the 20 running backs who played at least 25% of the team’s passing snaps who had a higher PBE than Mendenhall in 2011. I am not taking Mendenhall’s strong 2011 season in pass protection away from him. What I’m attempting to do is put it in context with the rest of his career.

    With the exception of his 4 snaps last year, Mendenhall has been below the team average every year. The team average has improved steadily in recent years, yet Mendenhall’s average has still lacked in comparison. The team’s PBE last year was 97.57, so more than a full percentage point better than Mendenhall’s best season, not the 0.3% statistic you mention; you compared Mendenhall’s PBE to the team’s success rate; they’re not the same statistic. Mendenhall’s success rate was 95.2 in 2011. Mendenhall still lagged behind. Therefore, it is still true that Mendenhall’s PBE brought the team’s running back PBE down, and therefore, my point in the article holds true, even using Mendenhall’s best season as the barometer.

    For the record, your Keenan Lewis comparion holds no water because you are assigning fictional snaps to Lewis. He played very sparsely in his first two seasons, so his career averages would not be far off from his averages over the last two years. Mendenhall played the snaps that he played, and I did not even use any of his data as a rookie. He should have been able to know what he was doing by year two, but the fact of the matter is that he did even worse in year three. And he had no excuses, being tutored by backs who were strong in pass protection like Willie Parker and Mewelde Moore.

    Additionally, it is widely understood that running back is a much easier position to grasp earlier on than cornerback. I guarantee you that the number of rookie starting running backs exceed the number of rookie starting cornerbacks despite the fact that each team starts two corners.

    For the record, Lewis logged 29 snaps in his first two seasons, and his quarterback rating against was 64.6. HIs rating against last year was 80.7.

    There really is no argument here. You say 2011 points to a clear and marked improvement in Mendenhall’s pass protection capabilities. I say his 2011 numbers are not substantial enough to conclude whether it was an improvement or an anomaly. And as I explained, because I can’t make that judgement, I have to use a broader sample size. I hope you are understanding that I’m choosing to make this my last response on the topic. In light of that, I tried to be as comprehensive as possible.