Night Train Lane, Xavien Howard, and the best ball-hawking seasons in NFL history

In 1952, in his first season as a professional football player, Hall of Fame defensive back Dick "Night Train" Lane burst onto the scene as one of the most ferocious tacklers in NFL history while also setting a still-standing record for single-season interceptions with 14. Lane's record is made even more impressive by the fact that he caught those 14 picks in just a 12-game season, averaging 1.17 interceptions per contest in a year that saw him become one of only four men ever (or since 1940, as far back as pro-football-reference.com's yearly defensive leaderboards are available) to win the league interception crown while averaging over one pick per game (with Lane's '52 per game rate the highest among all seasonal interception leaders since 1940). 

Since those early days of the NFL, the aerial game has steadily increased its presence in professional football. In that 1952 season, Night Train's Los Angeles Rams defense faced 360 pass attempts from opposing quarterbacks while the league's twelve teams threw the ball 27.9 times per game on average; by 2013, a year in which pick leader Richard Sherman's Legion of Boom Seahawks faced 524 pass attempts, NFL teams were collectively throwing the ball over 18,000 times a year, letting it fly an average of 35.4 times per game each. 

Perhaps counter-intuitively, this passing explosion has not really led to an increase in interceptions thrown. Of course from a literal standpoint there are significantly more picks caught nowadays than there were in a given season in the 50s, but 32 teams playing 16 games a piece will naturally accumulate more counting stats of all sorts than will 12 teams playing 12 games each; however, while the average team in 1952 threw 24.8 interceptions, the mean 2013 squad threw just 15.7 picks despite playing in 33% more games and attempting more than 25% more passes in those contests. To further illustrate the point, we can contrast Sherman's league-leading 8 interceptions in '13 (nobody has snagged double-digit picks since Antonio Cromartie's 10-interception 2007 campaign) to the 14 that Lane had in '52 (during a stretch in which the NFL's pick leader caught at least 10 interceptions in thirteen straight years, from 1946-58). 

This, of course, is all evidence of something we already know: NFL teams are passing increasingly more often as well as increasingly more effectively over time. That increased effectiveness in the passing game has led to a league-wide interception rate that has gone from 7.38% of all pass attempts leading to picks in Lane's great 1952 (after the league's all-time high-water mark of 11.07% that defenses enjoyed in 1944), to defenders now snagging passes at a rate not exceeding even a third of that number in any season since 2014. One consequence of this decrease in aerial turnovers is a disproportionate amount of modern quarterbacks at the top of the all-time leaderboards in things like interception rate and touchdown-to-interception ratio. You have to go down to 11th on the all-time int. rate list and to 27th on the TD/Int. ratio list before you find passers who never took a snap in the 21st century, and on the int. rate list, we find old school Hall of Famers like Roger Staubach, Fran Tarkenton, Dan Fouts, and Johnny Unitas occupying much the same space as many nothing-to-write-home-about modern guys like Tim Couch, Kyle Boller, Rex Grossman, and Trent Dilfer. 

And yet, in our appreciation of quarterbacks, we tend to do a pretty good job of automatically calibrating the necessary adjustments for comparing players between eras. It's well known that guys threw more picks back in the 60s and 70s, so, aside from the occasional hot-take artist firing off silly "Joe Namath wasn't actually good" -type takes on Twitter, we as football people know that Sonny Jurgensen and Terry Bradshaw were better than are Case Keenum and Brian Hoyer despite the latter having posted far superior efficiency numbers during their careers, and smart folks have even developed metrics that adjust things like passer rating and TD% for the era in which a quarterback played. 

Where I think we are less good at this is in evaluating the performances of the defensive players who catch those interceptions. I know that Night Train Lane is one of the best defensive backs of all time, and I'm aware that he holds the single-season interception record; what I don't know, given the changes in the game between the end of Lane's career in 1965 to today's NFL, is how he compares to greats like the aforementioned Richard Sherman, or Deion Sanders, or Champ Bailey, or even to (perhaps) slightly lesser players like Asante Samuel or Marcus Peters. He had more interceptions in 14 seasons than Sherman and Peters do collectively in their combined 14 seasons (as of 2019), but I have no idea what that means about the greatness of any of those three players. 

So, I decided to formulate some statistics that would offer era-adjustments for the best interceptors of all time. 

My thought process in doing this was fairly simple: in any particular season, we can approximate the amount of interceptions a team could've been expected to catch on defense given the total amount of pass attempts the team faced over the course of the year in addition to the league-wide interception rate in that season (we'll call this Expected Interceptions). To give a quick example: given that 2.77% of all passes thrown in 2013 resulted in interceptions, the 2013 Seahawks could've been expected to pick off 14.5 of the 524 passes they faced that year (they actually snagged 28, maybe they were pretty good). From this Expected Interceptions number and its relationship with a player of interest's actual interceptions total, we can derive a few "market share" -style metrics that -- by judging players based on their performance relative to league average -- can theoretically be used to compare players between different eras. 

The first of these is something I have yet to come up with a catchy name or handy acronym for, but I'll call it Vacuum Rate for now, just for the hell of it; Vacuum Rate describes the percentage of a team's Expected Interceptions total that a player actually intercepted in a season, or essentially how much of the "available" picks (again, based on an estimate generated by league-average Int.%) the dude vacuumed up by himself. To continue using our mostly arbitrary example from earlier, Night Train Lane caught over half of his team's 26.6 Expected Interceptions in 1952 for a Vacuum Rate of 52.69%, while Richard Sherman's 8 picks in 2013 represented 55.16% of the Seahawks' expected total of 14.5 (this is where I smash the sideways eyes emoji at Sherman outperforming the most prolific interception season of all time per this stat I just made up).

The second metric I derived from Expected Interceptions is something I'm calling Expected Picks-x, or EPx, mostly because I can't think of anything else but also because it just makes sense to call it that and throwing a lowercase "x" next to two uppercase letters in an acronym looks dope. 

Expected Picks-x is essentially the degree to which an individual player's actual interception total exceeded his personal Expected Interceptions total. You'll notice now that the Expected Interceptions metric I am using on the team level does not have expectations for individual players built into it; in order to formulate those, I (almost completely) arbitrarily decided that -- in our theoretical world of expectations in which league average applies to all situations -- it would make sense for all the interceptions made by a team to be (sort of) evenly distributed among a few defensive entities: 1/7th for each of four starting defensive backs, 2/7ths for the linebacker group as a whole, and 1/7th divvied out between everyone else (linemen, backup DBs, etc.). Essentially, if a defense has an Expected Interceptions total of 21 in a given season and all their players are league average-quality players, it seems reasonable that the starting cornerbacks and safeties might have 3 each (for a total of 12), the starting linebackers might have 6 among them, and then 3 more picks might be made by various other random members of the unit. Given this assumption, we now have Expected Pick totals for individual players on defense (at least for mainstay players in the defensive backfield, which is who we care about for the purposes of this article).

You may disagree with the assumptions I'm making about what percentage of picks a given player or positional group might be expected to snag given a certain amount of total Expected Interceptions for their entire defense, and that's ok. There's certainly work that could be done with the historical data that could give us actual such distributions for interceptions, but a) that would be a lot of work that I don't really want to do right now, and b) such work is not necessary for the analysis that I'm doing to make sense. It doesn't really matter if starting defensive backs have historically caught 1/7th or 1/8th of their team's total picks -- all that matters is that we judge all of them using the same starting point. Given that Night Train Lane's 1952 Cardinals would've been expected to come away with 26.6 interceptions over the course of the year, and given that Richard Sherman's 2013 Seahawks would've been expected to grab 14.5, if we assign Lane and Sherman the same expectation as each other -- catch 1/7th (or 1/8th, or 1/9th, or whatever) of those team Expected Interceptions -- we can then look at their actual interception totals and see which of them snagged picks above expectation to the greater degree. Lane's Expected Picks total (using 1/7th of Expected Team Interceptions) in '52 was 3.80, while Sherman's was 2.07; Lane grabbed an actual 14 picks, representing a total 3.68 times his expectation, and Sherman's 8 real-life picks were good for an EPx of 3.86. 

(note: the discerning reader will realize here that Vacuum Rate and EPx are really just two sides of the same coin. If you take all player seasons and sort by one of them, you'll get the same result as if you took all player seasons and sorted by the other. Use one, use the other, don't use either, whatever; I like both of them because they are framed differently: Vacuum Rate is sort of like the Dominator Rating of interceptions, while EPx is a more intuitive and straightforward way of answering the question "how good was this guy's performance relative to the performance of the rest of the league at the time he was playing?")

(another note: these stats are obviously not flawless representations of what happened on the field. A team or player may have been expected to intercept Z amount of passes in a given season based on league-wide interception rates and total pass attempts faced when they actually dealt with a much easier or harder group of quarterbacks than league average; diving deeper into the schedules of each player/team in this study and calculating expected interceptions based on the interception rates of individual quarterbacks faced would likely produce numbers closer to what actually happened, but I think we'd be looking at a diminishing returns situation there for the amount of time it would take; a PFF-style charting of each pass attempt going back to 1940 would be the best way to calculate how many interceptions a defender had over what they could or should have been expected to grab, but I don't have access to all-22 for every game ever and such a process would take my entire life anyway; I'm one guy with a laptop, and while I know the way I'm calculating Vacuum Rate and EPx is not perfect, I do think it's a) interesting, and b) perhaps better than the way we currently/typically think about defenders in historical contexts, so there we go.)

Now that I've laid out exactly what I did with the historical data available to me, I can move on to what I initially set out to uncover with this analysis: how do the most prolific interception-grabbing seasons compare to one another when you look at them through a lens of historical context, and, ultimately, what is the greatest ball-hawking season in NFL history? I pulled the numbers for every league-leading interception season since 1940 in both the AFL and NFL, including ties (for a total of 129 individual player campaigns), to find out. 

Here, sorted by EPx and Vacuum Rate and also listed with raw interceptions per game and interception rate based on total attempts faced (divided by games played), are the "bottom" 20 league-leading interception seasons since 1940: 

While I haven't done any research into what a "normal" EPx or Vacuum Rate performance would be, I think it's safe to say that while grabbing around 65% more picks than expected and accounting for about a quarter of your team's total interceptions does not make for an especially impressive season historically, those marks are surely impressive relative to the entirety of all players and seasons -- after all, posting the "worst" league-leading interception campaign in NFL history is sort of like being the worst Super Bowl winner ever: you're still pretty damn good. 

One thing you'll notice here is that these lower-rated seasons skew early; three guys who tied for the interception lead in 1999 all find themselves here, with James Hasty boasting a more impressive performance that year than either Donnie Abraham or Hall of Famer Rod Woodson, and then we have Mark Murphy in '83 and Thom Darden's '78 campaign as the only other post-merger seasons in the bottom 20. 

The next set of 20 interception champs also features more pre-merger seasons (12) than it does post-:

Also, we see Troy Vincent from 1999 pulling up to outpace Hasty, Abraham, and Woodson (though even Vincent can't claim the top spot from that '99 season when five dudes shared the interception title), and more Hall of Famers in Willie Wood, Night Train Lane, and Charles Woodson make cameos. We also have the first instances of guys averaging more than one interception per game (a number that is not era-adjusted) in Bob Nussbaumer and Spec Sanders. In this group are also the first seasons in which players outdo their Expected Picks output by at least 3x, as well as the first seasons in which players collect at least 40% of their team's Expected Interceptions total. 

This group of 20 represents the last list made up entirely of seasons that can be categorized as "below average" interception-leading campaigns. Here we finally see the '99 champ in Sam Madison, as well as two Ronnie Lott seasons alongside a whole host of other Hall of Famers: Kenny Easley, Johnny Robinson (who appeared in the first group with his AFL pick title from 1966), Rod Woodson (again), Mel Renfro, and Larry Wilson have all been honored in Canton. Bulldog Turner also is a member of the Hall of Fame, and is one of only two linebackers to ever lead the NFL in interceptions. 

Nate Odomes' 1993 season is #65 of these 129 total seasons, making his performance the median for interception leaders since 1940, while Eric Turner's 1994 and Stephon Gilmore's 2019 seasons represent the first here that are above average for historical pick champs (the average EPx for all 129 seasons is 3.41, and the average Vacuum Rate is 48.50%). 

Also, because I started pointing out Hall of Famers, Jack Christiansen and Ty Law both own gold jackets. We also have here the second and (so far) final linebacker to win the interception crown in Don Shinnick back in 1959. 

Forgive the randomly-selected cell in the bottom right of this screenshot (also please forgive that I am posting this list via screenshots lmao; I am a technological dunce that doesn't know how to embed tables so here we are). 

This grouping marks the beginning of the top 50 era-adjusted interception-leading seasons of all time (technically Eric Weddle's 2011 campaign is #50, but he got cut off at the top of the last group of 20; apologies to Mr. Weddle), as well as the first cohort that is fully made up of players who caught at least 50% of their team's Expected Interceptions. 

The 2019 interception title was shared by three guys, and here we have Tre'Davious White beating out both Anthony Harris and Stephon Gilmore (both of whom were on the previous list). I believe the only Hall of Famer here that I have yet to name is Mel Blount. 

I think the most notable season on this list is Night Train Lane's 1952. It is the 14-interception campaign that remains the single-year record and, as mentioned before, is also the 1.17-interception per game season that is the standard for all 129 seasons included in this analysis. Despite leading those categories, however, Lane's legendary year comes in at #38 all time when viewed contextually and adjusted for era. 

The above group is only made up of 19 seasons because I wanted the last screenshot to just be the top 10. Here we have our first 4+ EPx marks as well as our first 60%+ Vacuum Rates; the raw interception/game numbers from Sammy Baugh and Dan Sandifer back in the 40s are ridiculous, as is the fact that Baugh intercepted more than 1 in 20 passes faced by the entire Washington defense in 1943.

New Hall of Famers here are Paul Krause, Ed Reed, Champ Bailey, Lem Barney, and Sammy Baugh (inducted as a quarterback). 

Finally, the top 10:

Yet-to-be-named Hall of Famers here are Aeneas Williams, Jack Butler, and Bill Dudley. 

What strikes me here is a) the absolutely insane 6.17% interception rate that Dudley posted in '46, b) the amount of incomplete seasons in which guys here at the top of the list led the league in picks (only Williams and Ty Law of the "modern" guys played a full 16 games in their year atop the interception ranks), and c) that only three players ever have posted a Vacuum Rate over 70% (at least among players who led the league in picks), and only two have ever hauled in more than five times as many interceptions as they could have been expected to (same caveat applies, obviously). 

And so, the greatest ball-hawking seasons in NFL history belong to Ed Reed and Xavien Howard, with Howard slightly edging out the Ravens great for this particular crown. Night Train Lane is an all-timer, but armed with the context of historical interception rates, we see that nobody has ever done it better relative to the time in which they played than Xavien Howard in 2018. In a modern NFL in which quarterbacks throw picks not even a quarter as often as they did in 1940 (and not a third as often as they did in Lane's record-setting 1952), the Dolphins corner took more passes away relative to expectations than any interception leader ever. 

Ed Reed more than deserves his honorable mention as the only player since Lester Hayes in 1980 to grab at least 0.80 picks per game. These numbers that I made up bear out his being the GOAT safety as Bill Belichick and so many others believe.

But back to Howard: incredibly, he is currently in the midst of outdoing himself, as his ball-hawking pace 11 games into this 2020 season (where he leads the league in picks) would supplant his 2018 campaign as the greatest of all time if he were to keep it up. 

He has already matched in 11 games the 7 picks he caught in 12 games back in '18, and his posting a Vacuum Rate over 80% is absolutely nuts. Xavien Howard is really damn good. 

That's all I got for the pick title stuff. Thanks so much for reading, and if you liked whatever this was, consider subscribing to the blog or following me on Twitter at @noahmoreparties. I usually write player evaluation-type stuff in the context of fantasy football (I also publish that sort of thing over at breakoutfinder.com), but for my next article, I'll be going through the all-time interceptions list and producing career numbers in both Vacuum Rate and EPx for the NFL's most prolific pick artists in order to crown someone the greatest ball-hawk of all time. Maybe Xavien Howard will be there someday. 

curiosities about seasonal pick leaders:

- The 1947 interception title was shared by two guys named Frank: Reagan of the New York Giants and Seno of the Boston Yanks each had 10 picks that year. Reagan came out ahead from a rate perspective as he hauled in 52.02% of expected team interceptions to Seno's 43.07%.

- 1946 interception king Bill Dudley of the Pittsburgh Steelers also led the league in rushing that season with 604 yards. 

- Roy Zimmerman won the 1945 interception title by picking off 7 passes in 10 games. As the Philadelphia Eagles' quarterback, Zimmerman also threw 8 picks, though his 6.3% interception rate was well below the league average rate of 9.16%. 

- In back-to-back years in 1942 and 1943, Hall of Fame center Bulldog Turner of the Giants and Hall of Fame quarterback Sammy Baugh of Washington led the league in defensive interceptions with 8 and 11, respectively, while Hall of Fame wide receiver Don Hutson of the Packers won the interception title (in a three-way tie) with 6 in 1940. 

- Great names: Bob Nussbaumer, Spec Sanders, Otto Schnellbacher, and Night Train Lane won the interception titles in consecutive years from '49 through '52, and then we had Goose Gonsoulin as the AFL interception champ in '60 and Dick Westmoreland in a three-way tie for the AFL lead in '67. 

- Since 1940, 10 men have claimed multiple seasonal interception crowns: Night Train Lane, Jack Christiansen, Johnny Robinson, Emmitt Thomas, Bill Bradley, Everson Walls, Rod Woodson, Charles Woodson, Ed Reed, and Asante Samuel. Of this group, only Bradley in 1971 and '72 and Walls in '81 and '82 have won the league title in back-to-back years. Only Ed Reed from this cohort can boast three pick titles, in '04, '08, and '10. Johnny Robinson and Emmitt Thomas, both of the Kansas City Chiefs, hold the distinction of having won interception titles in two different leagues, as Robinson led the AFL in picks in '66 before pacing the NFL in '70, while Thomas was the AFL leader in '69 and the NFL champ in '74. 

- In Robinson's AFL-leading 10-interception season in 1966, he shared the pick title with teammate Bobby Hunt. 

- Ties for the interception crown are not uncommon: there have been 26 seasons without a single, outright leader in either the NFL or AFL since 1940, including seven instances of a three-way tie (1940, 1957, 1959, 1967, 2011, 2018, and 2019), one four-way tie (2009), and one five-way tie (1999). 

- Kent Ryan of the Lions was 25 years old and in his third season when he won a share of the league's interception title in 1940. That season would be his last as a professional football player, as he left the NFL to serve in the Army in the South Pacific in World War II starting in 1941. 

- 1960 interception co-champ Dave Baker also left football for the Army after his third season in 1961.

- The longest span of time between league-leading interception seasons is the 9 seasons between Darren Sharper's first pick title in 2000 and his second in 2009. The shortest amount of time are the zero seasons between the back-to-back pick titles won by Bill Bradley (1971-72) and Everson Walls ('81-82).