≡ Menu

We’ve been talking a bit about Charlie Joiner over the past few days. Here’s a good comment from Brad O, where he called Joiner “the best receiver on the best passing team this side of Dan Marino.”

Brad is right in that Joiner generally played on very good passing teams. That wasn’t the case during his years in Houston, but beginning in 1974, Joiner generally played on top-5 passing teams for over a decade. With the Bengals and Ken Anderson, Joiner’s team ranked 4th in value added over average in 1974, defined as (ANY/A minus league-average ANY/A) multiplied by team pass attempts. The next year, his Bengals led the league in passing Value.

In ’76 and ’77, Joiner was with the Chargers, and San Diego was middle of the road then, ranking 12th (in a 28-team league) both seasons, and the team ranked 18th and 19th in points scored those years as well. But things changed in ’78, when San Diego began its famous ran under Don Coryell.

San Diego ranked 5th in Value in ’78, then 3rd in ’79 and ’80, and then 1st in ’81 and ’82. In ’83, the team was 7th, and rebounded to 4th and 3rd over the next two years. In ’86, Joiner’s final year in the league, the Chargers dropped to 22nd.

A quick count shows nine seasons where Joiner’s team — always with Anderson or Dan Fouts, of course — ranked in the top 5 in passing Value, including three seasons at #1. It’s tough to say exactly how much this was on Joiner; by way of comparison, I wonder how future generations will look at Wes Welker who was on the ’07 Patriots, the ’13 Broncos, and lots of great passing teams in between. But it is one way to craft a pro-Joiner case.

One thing we could do is compare how effective at passing Joiner’s team was in any given season to how large a percentage of the pie Joiner was taking. His best season, by far, in terms of percentage of team receiving yards was in 1976. That season he had 39% of the Chargers receiving yards, and he never topped 26% in any other season. That 39% mark was the 2nd best in the league, so you could say he was a very big part of the Chargers ranking 12th in passing Value. How does that compare to say, 1981, when he ranked 20th in percentage of team receiving yards (at 24%) for the top-passing team in football? I’m not quite sure, but let’s plot the information.

In the graph below, I have plotted in orange where Joiner ranked among all players in football in percentage of team receiving yards, all against the left Y-Axis.  Then, I plotted in blue, where Joiner’s team ranked in passing value, measured against the right Y-Axis.  Note that I excluded his ’69, ’72 and ’73 seasons.1  Presumably, the place to be is with both lines as high as possible on the graph, although it is up to each of you to determine which line is more important:

joiner rec yds perc

Taking a look at this, his ’76 season certainly stands out as noteworthy, and you can see what Brad is talking about with respect to him being a big part of a great passing attack, particularly in’79, ’80, and ’81. Here is the same information in table form:

YearTeamPass Val RkPerc RkPerc of Tm Rec Yds

So, what do you think of this as a way of evaluating a wide receiver’s career? Is there a way we can turn this into a metric to evaluate all wide receivers? Please leave your thoughts in the comments.

  1. Why? I only wanted to do this post-merger, and since Joiner had just 77 yards as a rookie, I thought that season was meaningless.  I wasn’t quite sure what to do with his ’72 season, since he split his time with two teams, and for ’73, he was hurt and had just 214 yards.  It would have distorted the graph to include what is obviously not an important season for his career. []
  • For 1972, I think you could prorate his time with each team to 16 games and find those fake rankings. Alternatively, you could just calculate the team rankings over the time period he played for those teams. A third option is that you could just treat it like it isn’t that important and ignore it completely.

    • Yeah. If it was a big season for him, I’d give it more thought, but I ignored for ease here. More concerned about the general process.

  • Thinking about this some more, I don’t know if this really boosters Brad’s argument. Based on my eyeball, Joiner’s most impressive seasons here are ’76, ’79, ’80, and ’81; those are also his four best years by raw receiving yards. And again, even ignoring the pass-happy teams he played on, the fact that he had only 20 points of Gray Ink is a knock against him, IMO. Now, I think Brad’s point is that he had a lot of quality seasons even if they weren’t dominant, and I don’t think I’d disagree. I guess it brings us back full circle as to how much we value those non-dominant but quality years.

  • sacramento gold miners

    I think future generations and historians will look at Wes Welker as a terrific receiver who ran a very limited route tree, was very productive, with some key drops in big games. He was excellent behind the line of scrimmage, and those five yard patterns in which he gained additional yardage. Tremendous career, but not at the HOF level.

    In terms of percentage of the pie for Charlie Joiner, the Chargers had so many weapons, the diversity made them so tough to defend. On a different team, Joiner may well have produced greater numbers, but that offense may not have been as productive.

  • FYI, I was reminded today of an old post on WRs that is interesting:


    It has Joiner 34th all-time. My guess is Brad will like that one.

    • Haha, while I appreciate seeing Joiner closer to where I think he should rank (which is top-30), that’s also a quick-and-dirty method, so there are some obvious hiccups.

      I think my larger problem is approaching WR stats as a percentage of team receiving yards. I just frankly think it’s the wrong approach. We’re much better off using raw stats and making subjective adjustments for team context than using team statistics and adjusting for individual context, or — even worse — assuming the new stats are perfect and don’t require any contextual adjustments.

      • If we assume there’s a realistic max/min on how many yards each team is going to pass for, the % of team yardage approach artificially inflates the one decent receiver on a bad team. For example, let’s take the 2015 Arizona Cardinals, who passed for 4,616 yards, and the 2015 St. Louis Rams, who passed for 2,805 yards, which is probably pushing that realistic minimum. The Cardinals gained 64.6% more passing yards than the Rams; the Rams gained 60.8% of Arizona’s total. But does that mean the Cardinals’ receivers were only 64.6% better than the Rams’ stable of trash?

        Larry Fitzgerald got about 25% of Arizona’s yards, and Kenny Britt got about 25% of the Rams’. Should Fitzgerald score 64.6% higher than Britt? I think that difference is much too small. Fitzgerald was elite and Britt was slightly above replacement level. We shouldn’t reward him just for being better than Stedman Bailey and Bradley Marquez.

        Also, unless I’ve misunderstood something, this approach underrates possession receivers while overrating deep threats. If we want to bring this back to Joiner, his role in the ’80s was possession receiver. He’s really a perfect test case for all of my objections to your WR-rating formula 🙂

        * Long, productive career without exceptional peak seasons
        * Played on pass-oriented offense with HOF-caliber teammates in receiving corps
        * Possession receiver in his best seasons, with more value than his yardage totals imply

        • Heh, I started writing my comment, got distracted, and by the time I came back and posted it, you had already written another comment. Whoops!

          The Cardinals ranked 1st in passing Value, and the Rams presumably ranked 31st or 32nd (I don’t have the #s in front of me). So that’s a massive difference: we are literally comparing apples to oranges here, but I’m comfortable with these examples. Fitzgerald was a big (but not enormous) part of the best passing attack in the NFL — that’s good evidence he’s very good. Britt was a big (but not enormous) part of a terrible passing attack — that’s good evidence that he’s starter quality, but little more. I see a massive difference between those two statements.

          It underrates possession receivers to the extent “yards” underrates possession receivers. A first down bonus would help, and Joiner may be harmed by using yards (tho ACY may solve that problem). As for the Joiner test case;

          ***Lack of exceptional peak seasons is kind of a big deal to me. In theory, someone could be really good without ever benefiting from variance, but just about every great WR has one or two huge years. Joiner didn’t, and it’s not like he had 8 really good ones, either. He had 4.

          ***Again, most people overrate players on pass-oriented offenses with HOF-caliber teammates. Or, most people do if you look just at stats.

          ***Maybe, on this point.

          • My point with the Fitzgerald/Britt comparison is about degree more than kind. Sure, you can broadly paint Fitz as very good and Britt as nothing special. But I think you’re underestimating how good Fitz may have been and how Not-special Britt was. I don’t even believe he was starter quality. Granted that it’s hard to judge anyone in the Rams’ impotent passing game, how many teams do you think Britt could have started for last season? I think it’s entirely possible that the answer is 1. I don’t think he would have made the Cardinals’ roster, and I’m certain he wouldn’t have been one of their top three WRs. I believe this approach leads you to underrate players like Fitzgerald and overrate players like Britt.

            I feel like the method outlined in this post, looking at individual production vis-a-vis team production, leads you back awfully close to the player’s yardage. Perhaps I’ve misunderstood something, but I can’t see this leading to any new conclusions.

            You’re right that my comment about possession receivers was really about using “yards” without any other stats. I think quick-and-dirty methods can be illuminating, but I wouldn’t want them anywhere near a pursuit-of-truth ranking.

            Lack of exceptional peak seasons is a big deal to me, too, but:

            1] not as big as it is to you, and
            2] my bar for exceptional is lower than yours

            I’d be interested in your reaction to this comparison: Charlie Joiner is the Curtis Martin of wide receivers. Never the best, but he was very good for a long time.

            • Analyzing receivers is tricky; wide receivers are often helpless in many ways, as Fitzgerald himself could testify. In some ways, being better than your teammates is all you can do. So I do think it’s somewhat meaningful to see what percentage of his team’s yards someone had. After all, Kenny Britt couldn’t make the Rams passing offense good — and guess what, neither could Fitzgerald — but he could be better than the other Rams receivers. There’s a balance, but the more you go towards team success, the more you give credit for playing with great QBs. I’m hardly saying this is my preferred methodology — you know it’s not — but I do think it’s another interesting lens to use.

              I agree that our bar for exception seems pretty different. I’m not sure there’s much to be figured out there, but it’s good that we’ve identified that!

              The Martin one rings hollow to me since the comparison only makes sense on a surface level (both HOF, both more known for their consistency than peak greatness).

              — Martin is tied for 6th in 100-yard rushing games since 1960; Joiner is tied for 48th in 100-yard receiving games

              — Martin is tied for the 2nd most 1,000-yard rushing seasons; Joiner is tied for 45th in 1,000-yard receiving seasons

              — In rushing yards over worst starter, Martin ranks 6th: http://www.footballperspective.com/career-rushing-yards-over-worst-starter-baseline/ In adjusted catch yards over worst starter, Joiner ranks 60th http://www.footballperspective.com/adjusted-catch-yards-over-worst-starter/

              There may be some era adjustments that bridge the gap, but that’s an enormous gap. And keep in mind that this is looking only at rushing yards: Martin added value by being incredible at holding on to the football http://pfref.com/tiny/NdNh8

              • Agreed on the bar for exception. That’s something intelligent, well-informed people can disagree on. By the same token, I think we can agree that longevity and consistency are more important to me than they are to you. I’m suspicious of one-year wonders.

                For the Martin comparison, you’re understating the era adjustments: they bridge the gap entirely. For players whose careers began with five years of Joiner’s:

                * The only one with more 100-yard games is Harold Jackson, barely (29-27). Joiner’s ahead of Warfield, Biletnikoff, Charley and Otis Taylor, Bob Hayes, Harold Carmichael, Cliff Branch, John Stallworth, Drew Pearson…

                * No one had more 1000-yard receiving seasons, and no one had as many. Joiner’s by himself in the lead. The only remotely contemporary player with more was Steve Largent, and maybe James Lofton if nine years younger counts as remotely contemporary.

                For players whose careers began with five years of Martin’s:

                * Three had more 100-yard rushing games: Emmitt, Bettis, and Edge. That doesn’t even include Marshall Faulk, whom I imagine you’d concede was more exceptional than Martin.

                * Emmitt Smith had more 1000-yard rushing seasons.

                The enormous gap between Joiner and Martin is an era illusion. In historical context, two of the three stats you cited favor Joiner. This probably brings us back to where you point out that Joiner’s teams passed a lot, but the 100-yard games, 1000-yard seasons statistics show that Joiner was exceptional, not that he wasn’t. I wouldn’t be so quick to dismiss the comparison.

          • Adam

            Do you think dominant seasons are primarily a product of randomness, or more of a true jump in the player’s ability? If it’s the former, I agree with Brad. If it’s the latter, I agree with you.

      • (Note: To be clear, my comment here is about today’s post, not the article I linked to.)

        I like to go about it in a number of ways, which I think makes me feel more confident when I generally see the same results.

        One thing I like about this method is it’s a departure from the normal way, but it has some rational elegance to it: you would expect a great wide receiver to either make up a huge percentage of his team’s yards or — if his team is so good that his greatness doesn’t stand out — to be on a really good passing team. There will always be exceptions, but I think there’s some sound theory about it, and it goes in a completely different direction than normal methods.

        Now I haven’t exactly come up with a method to compare these two variables, but if I did, I think it would be interesting to run through the numbers and see how Joiner compares. My hunch is he won’t stand out as particularly special. That then brings the follow-up question of whether this method underrates Joiner.

        As for raw stats, I think Joiner’s raw stats are not great if you put an emphasis on dominance, and I think most subjective adjustments work against them (chief among them being playing on pass-happy teams and with great QBs).

  • Here’s Lynn Swann’s stats:

  • Here’s Swann’s stats from ’75 to ’82, using the same methodology (and the same axes):