First – An Apology
Well, I was finally able to drag The Drinking Guy out of the bar and back into the writing nook. Please accept our apologies for how long this article took to publish and we will endeavor to put out articles/posts at a more frequent rate than this. On the bright side, we’ve come up with something that we think you’ll appreciate and can use to shut down the Steelers haters out there.
The Premise – The Greatest NFL (Statistically) Receivers of the last 40 Years
In the 2012 NFL season, Calvin Johnson broke Jerry Rice’s 1995 record for most receiving yards in a season. In 2007, Randy Moss broke Jerry Rice’s 1987 record for most receiving touchdowns in a season. In 2002, Marvin Harrison broke Herman Moore’s 1995 record for most receptions. Where do these feats place these players in the annals of recent NFL history? What about the oft-attacked Lynn Swann, whose detractors include ESPN senior writer Sal Paolantonio? In SalPal’s book/incendiary fuse, The Paolantonio Report: The Most Overrated and Underrated Players, Teams, Coaches, and Moments in NFL History, he names Swann the most overrated wide receiver in NFL history.
When one explores the raw numbers (total receptions, total yards, etc.) one may think that SalPal and Swann’s other detractors are correct. But, like all good things in life, you have to work a little harder to get the real answer. We decided it was time to compare the receivers of the NFL (since the merger) against each other, using numbers normalized to the modern, pass-happy, era of the NFL. What follows is an explanation of our assumptions/disclaimers, our method and the results. If you don’t care about our method, we advise you to at least read our assumptions/disclaimers.
The Players (assumptions/disclaimers)
We’re sorry to say that we had to leave some pretty good players out of the analysis. The reason, we must admit, is a sheer lack of time and availability to really delve into the old American Football League (AFL) stats. The NFL and AFL merged in 1970, and by then had already played 4 Super Bowls. Once the leagues were merged into one, statistics become more reliable and easier to find and use. Our chief source of anguish are the pre-1970 AFL stats, where running backs are listed alongside receivers and The Guys must admit that it was going to take way too long to determine who the running backs were vice the receivers. Those of you who watched football back then, or those that watch a lot of NFL Films products (as we do), recall that running backs were a greater focus of the offense than receivers, who usually didn’t get a ton of opportunities in a run-heavy time. Thus, when all of the stats are jumbled together, it got too messy for us part-timers. So, unfortunately, great receivers like “Bullet” Bob Hayes, Don Hutson, and Fred Biletnikoff were left out. Sorry.
So who did we include? We didn’t want to look at every receiver of the merged NFL era, because you don’t need to adjust numbers to know that Mohamed Massaquoi isn’t going to appear as a great receiver. A key assumption we made was that the merged NFL era greats will either have made the Associated Press (AP) All-Pro team and/or have been elected to the Hall of Fame. We have great respect for both institutions and we deferred to their selections to form our list of receivers. It should go without saying that we’re not talking about Pro Bowl players. The All-Pro team (All-Pro Wiki) gets our nod over the Pro Bowl and the other journalist-nominated teams due to its longevity and reputation (i.e., didn’t include Javon Walker once). Here’s the 2012 team, for example: 2012 All-Pro Team. The All-Pro players are the ones with (AP) next to their names.
All good Steelers fans know this list includes Lynn Swann and John Stallworth. It does not include Hines Ward or Louis Lipps. Elbie Nickel and Buddy Dial are in the top 10 in Pittsburgh career receiving, but they played before 1970. Heath Miller is in the top 10, but he’s a TE. We can do TEs in a future article, but for now we’re focusing on WRs. So, to get another Steelers WR in there, we threw in local fan favorite Plaxico Burress. Surprisingly, he’s #8 in receptions and yards all-time for Pittsburgh.
So, to reiterate, the included players must fit the following critera:
- Did not started their career until 1970
- Have been selected to the AP All-Pro Team since 1970 or have been elected to the Hall of Fame.
- or be named Hines Ward, Louis Lipps, or Plaxico Burress.
Overall, this is 48 players (45 All-Pro or HoF and Ward/Lipps/Burress). For the rest of this article, we will just refer to these 48 players as “the receivers.”
The Method – Quick Edition
As stated above, if you don’t care about our approach and just want to skip to the good stuff, skip this section and head straight for the section labeled The Results. Just know this: at a high level, we adjusted each season of the receivers’ key stats (receptions, receiving yards, and touchdowns) to modern NFL statistics. We then scored (via an index) the receivers in six key statistics:
- Best single season adjusted receptions
- Best single season adjusted receiving yards
- Best single season adjusted touchdowns
- Average adjusted receptions in the prime years
- Average adjusted receiving yards in the prime years
- Average adjusted touchdowns in the prime years
We didn’t count stats for a season for the receiver if the receiver averaged less than one catch per game (16 catches per season post 1978 and 14 catches per season pre-1978).
If you are interested in the details, keep reading dear friends. Otherwise skip ahead to The Results.
The Method – Long Edition
Below are the laborious steps we followed to eliminate, aggregate, and analyze the receiving data:
- Gather data for each player, by season. We were only interested in the basic receiving stats: Receptions, Receiving Yards, and Touchdowns.
- For each player, eliminate all seasons where the receiver averaged less than one reception per game. Yes, we are aware that between 1970 and 1977, they only played 14 games.
- Now the tricky part, we need to adjust each year’s stats to the modern era (which includes playing 16 games). Basically, we need to divide the older year by the modern year to get the coefficient to adjust each stat. To start, we gathered the top 20 values for receivers for each of receptions, receiving yards, and touchdowns for each season (1970 – 2012).
- Find the mean value of the top 20 values in receptions, receiving yards, and touchdowns for each season (1970 – 2012). We are treating these numbers as the representative values of receiving for that season.
Notice the trend creeping upwards from 1970. The rows in green are 14 game seasons. The rows in yellow are strike shortened seasons. We’ll talk about these later.
- Now that we have our numerator in the adjustment fraction, we need the denominator. The denominator is our modern representation of what passing looks like today. You could just use 2012, but we didn’t feel that would be enough data. Typically, a record setting year (Calvin Johnson) is going to skew the data a bit, so we opted for finding the mean of receptions, receiving yards, and touchdowns from 2010 – 2012. This yielded the following table, which forms the denominator of all adjustment calculations.
- Then, for each season, divide the three statistics by the 2010 – 2012 mean for the statistic to receive the adjustment coefficient. This yields the following adjustment chart, which is key to leveling the playing field for the receivers.
- Click on the chart to see it larger. Notice the larger coefficient for years 1970 – 1977, 1982, and 1987. In 1970 – 1977, the NFL season was only 14 games long (yeah, real impressive there 1972 Miami Dolphins) and in 1982 and 1987 there were players strikes.
- We couldn’t help but notice how large the coefficients were for the 1982 and 1987 strike years. In 1982, the year was only 9 games long. In 1987, most players crossed the picket line in time to play 12 games. We ditched these seasons (1982 and 1987) due to their effect on the others. Don’t worry, we’ll show you later why this is no big deal.
- You can guess what comes next. For each player and for each year, we multiplied the appropriate coefficients by the players’ actual stats for that year to get the adjusted statistics. For example, Lynn Swann caught 61 passes in 1978. As shown in the above adjustment chart, the adjustment coefficient is 1.644. Thus, to adjust Lynn Swann’s 1978 receptions to modern day statistics, we multiply 61 by 1.644 to get 100.3 adjusted receptions. What this statistic tries to predict: because from 2010 – 2012, the top 20 receivers (which Swann was in 1978) caught more passes than in 1978, he would have racked up 100.3 receptions in the modern era. Cool? We think so.
- If we left it at this, we could try and compare the single greatest seasons since the merger. However, to compare receivers as a whole, we must take into account both the peak of production from the receiver as well as their longevity. For example, David Boston had a truly great year in 2001. However, that was really his only great season, so we can’t say that he’s amongst the greatest receivers ever. On the flip side, a very consistent, but unspectacular receiver provides value, but not as much as a receiver that hits a higher peak production for a shorter period of time. To try and include both of these timeframes for each receiver, we also identified the receiver’s “prime years” and determined their average adjusted statistics over these five seasons. To get this, we indexed (i.e., measured the statistic against the maximum) each receiver’s adjusted receptions, yards, and touchdowns per season and then averaged each year. The five years where the receivers posted the highest average indexed score were considered the “prime” years. We then average each adjusted statistic over the 5 years for each player to get their prime statistics.
- This left us with six numbers for each receiver: Peak single season adjusted receptions, yards, and touchdowns, and the average adjusted receptions, yards, and touchdowns over their five prime years. We then sorted each statistic and assigned the highest player in each category a 1, as in 100% of the leader. Each subsequent receiver received a score that was a percentage of the leader in that statistic (e.g., an index). For example, if the highest adjusted touchdowns in a season was 25, that receiver would get a 1.000 in that category. If the second receiver had 24, that receiver would get 24/25, or 0.96.
- Finally, we averaged these six numbers to determine a final percentage of maximum output in comparison to the receiver’s peers. This puts every receiver on a single scale incorporating their adjusted stats from their peak season and their five prime seasons that ranges from 0 to 1, where 1 is basically Jerry Rice and 0 is basically Charles Rogers. Too soon, Detroit fans?
Whew, that’s an awful lot to digest. I hope we did a good job of explaining it. If you have specific questions, please feel free to post them or email them to us and we’ll be happy to answer.
All right, the moment you’ve been waiting for is at hand. We’ll start with the single season peak numbers first, which essentially reward players for great individual seasons. We’ll then transition into the prime years numbers, which reward a player for playing well over a longer period of time. Finally, we’ll average all six score together to get the overall receiver rating index for post-merger receiver.
Single Season Adjusted Receptions
Art “I played in the NFL for 125 years” Monk turns in the greatest single season adjusted receptions performance at 131.5 adjusted receptions in 1984. Art actually racked up 106 receptions that year, but in 2010 – 2012 passing terms, that means he would’ve caught 131.5 today. Notice that Marvin Harrison’s receptions were adjusted downwards because of the outlier nature of his receptions (143) that year (2002). The mean receptions of the top 20 players in 2002 was 92.75, which is the 2nd highest mean of the post-merger era (and very likely of all time). This results in a downwards adjustment of Harrison’s numbers as they equate to 2010 – 2012 numbers.
The highest finishing Steelers player in this category was Hines Ward in 2002. Hines logged 112 receptions, which also adjust downwards to 102.5 because there were more balls caught by top 20 players in 2002 than in 2010 – 2012. Hines’ score is good for 49th overall.
For those interested in seeing the list, click here. Fair warning, the image is huge. You can also see the full data in this Excel spreadsheet.
Single Season Receiving Yards
The next two categories are dominated by the same person. We’d like to reacquaint you with someone that we feel is surely deserving of being in the Hall of Fame, but for some unknown reason is not.
Cliff Branch had an unbelievably good year in 1974. You’ll notice how high his receiving yards are adjusted, but intuitively you know it’s what needs to be done. Think back to 1974 and do the research. The average of the top 20 receiving yards in 1974 was 684 yards. Branch recorded 1092. It was a dominant performance in an era where receiving was not prolific.
The highest Steelers performance was John Stallworth in 1984. Stallworth recorded 1395 yards which adjusts up to 1508 yards. This was good for 37th overall.
You can see the full list here, or in the same spreadsheet linked above in the discussion of single season receptions.
Single Season Touchdowns
Continuing his standout season in 1974, Cliff Branch wins this one again. Recall that we’ve excluded strike years, which did affect this category. If the strike years are included, Jerry Rice wins this one. Somehow, in 12 games, Rice managed to score 22 touchdowns. That makes our head hurt trying to explain that one.
But, given that we threw out strike years (again we’ll explain why it’s OK later), Branch takes this one with 13 touchdowns in 1974. In 1974, the NFL only played 14 games. In 1974, the mean of the top 20 receivers (not just all receivers – these are the top receivers that year!) in touchdowns was 5.7 and Cliff Branch more than doubled that at 13. Freak.
Not only did Branch win three Super Bowls with the Raiders, but he was an AP All-Pro selection four times. We’re speechless at any argument that keeps him out of the Hall of Fame. What are these voters thinking? The guy was a stud, and I think people in this city remember him from those days and are glad he’s no longer a threat to the Steelers defense.
The best Steelers results came from Lynn Swann in 1978 at 36th overall. Swann caught 11 touchdowns that year which adjusts up to 13.7. 1978 was the first year the NFL switched to a 16 game schedule.
You can see the full list here, or in the same spreadsheet linked above in the discussion of single season receptions.
Prime Years Average Receptions
We decided to provide the prime years average adjusted statistics in index form vs. the raw adjusted data (though we did include the overall table below for your musings).
In terms of prime season average receptions, this one kind of surprised us a little. It seems strange to think that Wes Welker is catching passes at a higher rate than anyone in NFL history. When you consider that he has five 100-catch seasons (the most in NFL history) though, it doesn’t seem so ridiculous. Welker’s mutation into receiving savant by that evil genius Bill Belichick is nothing short of mesmerizing. In his best five years (so far), Wes Welker has an average adjusted receptions value of 114.3. Contrast that with Jerry Rice, who’s in at 2nd place at 108.6. In case you haven’t been paying close attention to those graphs, though, I encourage you to go back and see how close Rice is to the top of all of the charts.
We included the top 20 receivers in this category but also each Steelers player for the prime averages in the graph. The charcoal colored values show the index scores of the Steelers players. Stallworth is the most respectable player here, with Lipps being the worst.
Prime Years Average Receiving Yards
We again included the top 20 receivers in this category and left all Steelers players in the graph to show you where they rate overall.
Ah Jerry, there you are. Jerry Rice’s consistency over time gives him a crushing victory here. These are indexed scores, so Jerry gets the perfect score (since he averaged 1649 adjusted yards) for prime average adjusted yards. The next closest player, Michael Irvin, weighed in at 1523 yards, or 92% of Rice’s yards (as shown in the graph above). Thus, the highest performing Steelers player in this category, John Stallworth, is lagging quite far behind at 1283.7 adjusted average yards, or 78% of Rice’s adjusted score.
Prime Years Average Touchdowns
Here’s another category where Jerry Rice is getting hurt, statistically, by not including the strike years. Again, he racked up 22 touchdowns in 1987 (in some insane way), but that entire season is discarded. Nonetheless, he still comes in second behind Randy Moss, who broke his record. Moss averaged 16.6 adjusted touchdowns over his best five years. Moss is, simply put, a touchdown machine. His yards numbers are good, his receptions numbers are underflated because he’s a home-run guy, but his prowess and nose for the end zone cannot be questioned.
Lynn Swann, Hines Ward, and John Stallworth are in yellow in the graph above because they crack the top 20 in this statistic. Those Steelers in charcoal are included for your reference, but were not in the top 20.
The data for all prime years’ statistics can be found here (click to enlarge).
Overall – Averaging the Total Scores
To weigh all players against each other, we simply took the average of the index score for each player in each of the 6 categories to yield their total index score (on a scale of 0 to 1).
There’s probably no surprise here. Ask any casual fan or expert who the greatest receiver of the post-merger NFL era is and you’d probably be at a place where there’s total agreement. Qualitatively and quantitatively, it’s one Jerry Rice. You’ll have noticed that he comes in towards or at the top of each category and his consistency is really what put him over the top. We’ll show you a slick chart, called a radar chart below, but first, look at the gap between Rice and his nearest statistical peer (Sterling Sharpe).
Jerry Rice posted an overall average index score of 0.9350. Sterling Sharpe came in at 0.9084. This is a landslide victory, by all accounts of the definition. After Rice, the rest of the top 5 generally plateau, and then the indexed scores start to drop off again once they hit Steve Largent.
We feel we must comment on Sterling Sharpe. Before even running these statistics, we always suspected that if you removed the total career numbers from the equation that Sharpe would be a sure-fire Hall of Famer. Unfortunately, his career was cut tragically short following a neck injury suffered during the 1994 season, a season in which he had a stat line of 94-1119-18. He was dominant and would have been in the discussion with Rice, we suspect, had he turned in a long career. Sharpe only played from 1988 – 1994 and was forced to retire at age 29, which is a relatively brief career for a star receiver. He was so good in his brief career that we feel it’s a travesty that he’s not in the Hall of Fame. For every year that Sharpe and Branch are kept out of the Hall, a corresponding modicum of respect is lost for that institution. There is a logjam at receiver, but the numbers here show that Cris Carter was hastily rushed in and that other players statistically had better careers when their numbers are adjusted for 2010 – 2012 passing. Lynn Swann and John Stallworth, while often criticized, place a respectable 31st and 26th overall. Both deserve their Hall of Fame accolades, but so do Sharpe and Branch.
Hines Ward finishes at 32th…just behind Lynn Swann. Does this mean he’s a sure-fire Hall of Famer in four years? Who knows.
Below is a chart that we think is pretty slick. It’s called a radar chart and it delivers a lot of data in one picture. The thick black line is the overall average index score, and the 6 categories are represented by the 6 colored lines. Another way of interpreting it is that Jerry Rice appears at the top of the circle, and as you move clockwise around the circle from Rice, the overall index score is decreasing.
And the data:
We talked a lot about purging out the strike years. Those years are really statistical outliers and distort the data. Players were less likely to get injured those years because they played significantly fewer games and there was little preseason training. The numbers only affect the top 5 players in this outcome slightly. The top 3 are untouched: Rice, Sharpe, and Moss. Without strike years, Cliff Branch finishes 4th and Marvin Harrison is 5th. With strike years, these two change places.
But…if we included the strike years, Jerry Rice’s domination continues and he ends up being 0.1 index points higher than Sterling Sharpe!
We must readily identify the weaknesses/assumptions of our approach. First, it cuts out some really good receivers due to the lack of massed AFL data. Second, it rules out strike years, which isn’t a weakness, but ideally you’d want to include every data point.
So, given the adjustment assumptions, what does the data tell us? It confirms that Jerry Rice is the greatest wide-out the league has ever seen (well, at least since the merger but come on…) and it really shows his separation from his peers. It shows that if you remove aggregated (e.g., career totals) stats and focus on the value added by that receiver in terms of stellar seasons and prime year contributions, that there are some great players that are not in the Hall of Fame.
On the local front, it does give Swann and Stallworth a bit of support, given that they are often attacked as examples of overrated players in the Hall of Fame. These numbers show they had respectable careers and when you factor in their Super Bowls (The Drinking Guy would like to note that he doesn’t factor that in to evaluating individual players like Swann and Stallworth but The Thinking Guy insists it matters), it’s obvious they should be in. Perhaps the most interesting revelation of this analysis is, although Louis Lipps ranks very highly in Steelers receiving statistics, his numbers just don’t hold up against top receivers in the league.
Use these as you will Pittsburgh. If you have any questions, hit us up and make sure you tweet us out. Cheers.