Keyshawn Giveth or Keyshawn Taketh Away?

Let's take a moment to ponder two running backs: Curtis Martin and Mike Alstott. Both will almost certainly be starting for somebody's fantasy team this year, and both will cost you a relatively high draft pick to acquire. What kind of years are these two going to have?

The link between them, of course, is Keyshawn Johnson. He was sent from the Martin's team to Alstott's, and this has prompted a good deal of speculation about their futures (and Keyshawn's -- but that's another article). Obviously, Keyshawn's presence upgrades Tampa's passing attack and his absence hurts the Jets' aerial abilities. But the question is: how does this affect the two teams' ground games?

There are two conflicting forces at work here:

So which force dominates? In general, the question is this: if a team's passing offense improves, does that usually coincide with a decline in the running game (because of a shift in focus), or an improvement in the running game (because it loosens up the defense)? Obviously, this question -- and related ones -- are relevant for many more players than just Martin and Alstott. Just off the top of my head:

I'm sure you can fill in a few more.

I want to warn you at the outset that this is an extremely hard problem to get a handle on. The interdependence between the running game and the passing game is very complex, and there's no chance of coming up with any hard-and-fast conclusions in a quick study like this one. It's a headache for statheads, but it also happens to be one of the things that makes football such a fascinating sport -- that's a trade I'll make any time. I want to be clear that this article is just a very brief look at some data to see if anything obvious jumps out at us.

The data I examined was the number of rushing and passing touchdowns scored by each team from 1995-1998 -- that's 120 teams. Now, let's find all the teams whose passing offense got better from one season to the next. For example, the 1998 Rams had 12 passing TDs. The 1999 Rams had 42, so their passing offense improved. There were a total of 54 teams that fell into this category. Meanwhile, there were 59 whose passing offense declined. (The number of passing TDs remained constant for 7 teams.) What happened to the rushing game for each group of teams?

                           Year 1        Year 2
Group                     rush TDs      rush TDs     Change
Passing game got better     13.0          12.3       -5.4%
Passing game got worse      13.0          12.7       -2.3%
Those numbers are averages. So to sum up, both groups scored an average of 13 rushing TDs in year 1. The following year, the team whose passing offenses improved (the top line) had an average of 12.3 rushing scores, while those whose passing offenses slipped (the bottom line) average 12.7 rushing TDs.

As you can see, the percentage changes are nearly identical. Thus, in the absence of other information, knowing that a team's passing attack is on the rise doesn't give you any hint about what will become of the ground game. In other words, the "he'll loosen up the defense" force and the "he's all we've got left" force seem to be equally strong. Sometimes one wins, sometimes the other, and sometimes they appear to cancel each other out. Again, I'll reiterate that a closer look or a different look at the same question might give us different results, so don't read too much into this.

When we look at things from the other angle, we get a similar picture:

                           Year 1        Year 2
Group                     pass TDs      pass TDs     Change
Running game got better     21.2          21.7       +2.4%
Running game got worse      21.2          20.7       -2.4%
Again, the overall change in the passing game was virtually the same for teams who beefed up their running games and teams whose running games took a nosedive.

So what does this say about Martin and Alstott?

From what I've read, the general consensus is that Keyshawn's absence will hurt Martin rather than help him. I'm not sure I agree. If a struggling passing game automatically spells doom for that team's leading ground-gainer, then that fact would have shown up in the study. But it didn't. Plenty of teams have suffered through ugly dips in their passing game while their running game chugged right along. Just a few examples:

It doesn't happen that way every time, of course. The point is that it's not clear what the effect on Martin will be.

I see a lot to like in Martin for 2000. His 259 receiving yards last year was a career low, and that should improve this year with Johnson out of the picture. Further, I think his low touchdown total from a year ago is to some extent a fluke (you can check out this article for more information on that), so I expect that to pick up a bit as well. He's got a history of relatively good health, and is virtually guaranteed 400 touches. Because he's no longer viewed as an elite back, I think he'll be a very good value pick this year -- he's likely to be a second round pick who will give you first round production.

As far as Alstott is concerned, I have to confess that, after much thought, I have no idea how Keyshawn's presence will affect him, and I'd rather be honest about that than make up an opinion and try to justify it.

That's OK, though, because my opinion on this isn't any more valuable than yours. One of the main goals of this article (and most of my articles) is to get you to question the things you read or hear. If some website (even this one) tells you that the threat of Keyshawn will start opening up gaping holes for the Buccaneer RBs, don't believe it. Think about it and decide for yourself -- and be aware that it doesn't always turn out that way. If you read in a magazine that Keyshawn's presence will be a drain on Alstott's fantasy production, don't believe that either -- history has shown that it's not always the case.

This study doesn't prove anything, but it is enough to cast some suspicion on anyone who claims to know what the addition of Keyshawn Johnson will do to Mike Alstott.