When should the Lions have given up on Joey Harrington?
Posted by Jason Lisk on September 19, 2008
In June, Chase Stuart wrote a series of posts about quarterbacks, including the worst quarterbacks of all-time. In that post, he had this to say about Joey Harrington:
There you have it — no QB has performed so far below the league average for so long as Joey Harrington. To be clear, Joey Harrington probably isn’t the worst quarterback of all time in an absolute sense. But in terms of being so far below average, but far enough above miserable to earn more playing time, Joey Harrington hurt his team more than any other QB in NFL history. If Harrington had been worse, he would have played less, and he wouldn’t have set back the teams he played on.
So that got me thinking. At what point should the Lions have given up on Joey Harrington? Let me define what I mean by "give up". It could mean releasing or cutting the player, but I don't necessarily mean waiting until that point. I more consider it the point at which the team should bring in a veteran quarterback, or another high draft pick, to legitimately compete as the starter and potentially beat out Harrington--and I don't count Mike McMahon as doing that. It's just hard to say all of that in a quick and easy way.
Chase opined that the reason that so many of the "bad" quarterbacks were recent high draft picks is because teams give them many opportunities to fail. I think that's right. And I'll go so far as to say that teams are far more likely to commit errors of holding on to a quarterback for too long, while rarely giving up on a quarterback to early--once they have seen him play any amount of time in a real NFL game. I can think of examples of quarterbacks who were drafted, never started for their original team, and found success elsewhere, but its relatively rare to find a quarterback who started but never had success with his original team, and moved elsewhere to have his first breakout.
But I think NFL teams who hold on to a bad quarterback for too long are compounding their problems, and committing a new and independent error. Drafting Joey Harrington may have been a mistake, but having him as the best quarterback on the roster, and starting him for four years, is a bigger one. All NFL teams make drafting mistakes or get unlucky, but the good teams move on quicker and do not compound their mistakes.
We can probably think of examples of young quarterbacks struggling, but was there a point at which Harrington's career path and numbers diverged from the quarterback successes? After all, he was the primary starter for four seasons--surely the decision could have been made before then.
I used Chase Stuart's database of all quarterbacks who had an above average season using the formula from this post, which included both rushing data and sacks. I pulled every quarterback who entered the league since 1970, and who had at least one above average season. Then, because there were backups who played a few games in a season who showed up as above average, I set a lower limit of +500 of value added from all of the above average seasons combined.
This isn't a very high standard. All of your recognizable hall of famer and all-pro quarterbacks of the last 38 years appear on the list, as do guys who had up and down careers, so long as they had some good seasons (or one really good one). Your new starters in Tennessee and Tampa Bay, Kerry Collins and Brian Griese? Check. Don Majkowski? You bet. Jay Schroeder, Scott Mitchell and Kordell Stewart? All on the list. Every quarterback who started for the Chiefs in the 1990's, from DeBerg to Grbac? Yes.
The three "best" quarterbacks to not make this list are probably Jon Kitna, Bubby Brister, and Trent Dilfer, after that, it really drops off (Oh, and you can add Super Bowl winner Eli Manning, though I suspect he would qualify for this list after this season). That's right, the Lions have followed up four years of Joey Harrington with three consecutive seasons with an aging journeyman who has still never had a single season where he was significantly above average. (Must . . . avoid . . . urge . . . to go on . . . Matt Millen rant. . . .)
The end result was a list of 98 quarterbacks. But then, it wouldn't be fair to compare how guys like Kurt Warner, Warren Moon, and Trent Green played in their first starting opportunities, since they came at a later age. Harrington started as a true rookie, albeit a relatively old one at age 24. So I kicked out everyone who threw 10 or more passes per game for the first time at age 26 or later. Let's dig in to the rest and see how they started their careers or played at similar age and experience level with Harrington.
Thirty quarterbacks on the Success list started as true, straight out of college rookies at age 25 or younger, and threw at least 10 passes per team game in their rookie season. It is common knowledge that rookie quarterbacks generally struggle. But there are different types of struggling, so I broke down each season by looking at completion percentage, yards per attempt, touchdown percentage, and interception percentage. A quarterback who throws for yards but throws alot of interceptions may be different than a quarterback who avoids interceptions but throws for few yards. Here are the bottom 10, along with Joey Harrington's numbers, for each of those categories, among our rookie starters. The last column is the raw number. The "adj" number is the adjusted number in relation to the team wide league average for that season. For example, interception rates were much higher in the 1970's, so a failure to adjust to league average would result in a list of guys almost exclusively from the 1970's on the interception list.
successes who threw 160 or more passes as rookies, worst completion percentage
first last year age adj comp% Doug Williams 1978 23 -15.5% 37.6% Terry Bradshaw 1970 22 -13.0% 38.1% Richard Todd 1976 23 -12.1% 40.1% Vinny Testaverde 1987 24 -11.8% 43.0% Don Majkowski 1987 23 -11.5% 43.3% Joey Harrington 2002 24 -9.5% 50.1% John Elway 1983 23 -9.4% 47.5% Kerry Collins 1995 23 -8.7% 49.5% Donovan McNabb 1999 23 -8.0% 49.1% Drew Bledsoe 1993 21 -8.0% 49.9% Joe Ferguson 1973 23 -7.5% 44.5%
successes who threw 160 or more passes as rookies, worst yards per attempt
first last year age adj ypa Donovan McNabb 1999 23 -1.91 4.39 Joey Harrington 2002 24 -0.95 5.35 Troy Aikman 1989 23 -0.63 5.97 Richard Todd 1976 23 -0.43 5.37 Drew Bledsoe 1993 21 -0.39 5.81 John Elway 1983 23 -0.08 6.42 Joe Ferguson 1973 23 -0.07 5.73 Jeff George 1990 23 -0.06 6.44 Bernie Kosar 1985 22 -0.04 6.36 Kerry Collins 1995 23 -0.01 6.29 Dan Fouts 1973 22 0.00 5.80
successes who threw 160 or more passes as rookies, worst touchdown percentage
first last year age adj td% Neil Lomax 1981 22 -2.5% 1.7% Richard Todd 1976 23 -2.3% 1.9% Joe Ferguson 1973 23 -1.9% 2.4% John Elway 1983 23 -1.7% 2.7% Terry Bradshaw 1970 22 -1.6% 2.8% Rodney Peete 1989 23 -1.5% 2.6% Vinny Testaverde 1987 24 -1.5% 3.0% Jim Zorn 1976 23 -1.5% 2.7% Dan Fouts 1973 22 -1.2% 3.1% Joey Harrington 2002 24 -1.2% 2.8% Troy Aikman 1989 23 -1.0% 3.1%
successes who threw 160 or more passes as rookies, worst interception percentage
first last year age adj int% Terry Bradshaw 1970 22 +5.8% 11.0% Richard Todd 1976 23 +2.6% 7.4% Troy Aikman 1989 23 +2.2% 6.1% Jake Plummer 1997 23 +2.1% 5.1% Peyton Manning 1998 22 +1.6% 4.9% Dan Fouts 1973 22 +1.4% 6.7% Jim Zorn 1976 23 +1.4% 6.2% Kerry Collins 1995 23 +1.3% 4.4% Steve Grogan 1975 22 +1.3% 6.6% Chris Chandler 1988 23 +1.3% 5.2% Joey Harrington 2002 24 +0.6% 3.7%
Okay, so let's summarize. Most rookies do struggle relative to the league average, at least in most categories. I certainly wouldn't exclude a player just because he threw alot of interceptions as a rookie. Alot of really good quarterbacks struggled with interceptions as rookies, moreso than Harrington. Only 36.7% of the successes were at or below league average in interceptions as rookies. Though I've previously noted that throwing for a high touchdown percentage at a young age is a positive indicator, a below average touchdown rate as a rookie is not a reason to exclude a player, as most of the rookies who struggled simply played on bad teams. 30% of the successes were at or above league average in touchdowns as rookies. Virtually all of our rookies struggled with completion percentage, with several posting seasons significantly below 50%, and only 20% finishing at or above the league average.
There is one category, though, where the rookies who became successes did pretty good as a group--yards per attempt. 70% were at or above the league average in this category. Even most of those that were below average were not significantly so. Most of the young quarterbacks who became good starters in the league may have struggled in completing a high rate of passes or in throwing interceptions, but they showed an ability to make plays. Joey Harrington ranked ahead of only Donovan McNabb in season-adjusted yards per attempt. However, I'm not going to exclude anyone based on their rookie year. This is different than saying that the rookie season doesn't matter--because it does. I would much rather have my rookie start his career like Dan Marino or Ben Roethlisberger--I would feel more confident in my choice right away. It's just that with the variety of circumstances (most rookies are playing for really bad teams) we can find rookies who struggled early and later became successful.
So, I'm not giving up on Joey Harrington after his rookie season, though it wasn't a particularly stellar one by any measure. We are dealing with only 30 comparable successes who started as rookies, and several of them who became good struggled in one or more categories.
Let's move on to year two. Here, I not only bring forward all of our successful rookies who, like Harrington, also started as second year players (all except for Steve Bartkowski and Chris Chandler) and also add in any other quarterbacks who began starting by age 25, the same age as Harrington in the 2003 season. This brings in an additional 29 quarterbacks who didn't play significantly as rookies, but started soon after, for a total of 57 successes to use in comparison to Harrington.
Here are the bottom ten performers in each category:
successes who threw 160 or more passes as a non-rookie, by age 25 or under, worst completion percentage
first last year age adj comp% Doug Williams 1979 24 -12.3% 41.8% Steve Beuerlein 1988 23 -10.2% 44.1% Jim Zorn 1977 24 -9.9% 41.4% Phil Simms 1980 26 -8.2% 48.0% Gus Frerotte 1995 24 -7.9% 50.3% Steve Deberg 1978 24 -7.7% 45.4% Jeff Blake 1994 24 -7.0% 51.0% Vinny Testaverde 1988 25 -6.7% 47.6% Michael Vick 2002 22 -4.7% 54.9% Steve Grogan 1976 23 -4.2% 48.0% Joey Harrington 2003 25 -3.0% 55.8%
successes who threw 160 or more passes as a non-rookie, by age 25 or under, worst yards per attempt
first last year age adj ypa Wade Wilson 1984 25 -1.17 5.23 Joey Harrington 2003 25 -1.00 5.20 Steve Deberg 1978 24 -0.80 5.20 Phil Simms 1980 26 -0.63 5.77 Jeff George 1991 24 -0.40 6.00 Donovan McNabb 2000 24 -0.39 5.91 Chris Miller 1988 23 -0.32 6.08 Don Majkowski 1988 24 -0.09 6.31 Drew Brees 2002 23 -0.06 6.24 Troy Aikman 1990 24 -0.04 6.46 Mark Brunell 1995 25 -0.03 6.27
successes who threw 160 or more passes as a non-rookie, by age 25 or under, worst touchdown percentage
first last year age adj td% Jim Plunkett 1972 25 -2.2% 2.3% Ken Anderson 1972 23 -2.2% 2.3% Wade Wilson 1984 25 -1.7% 2.6% Jay Schroeder 1985 24 -1.7% 2.4% Jeff George 1991 24 -1.6% 2.1% Neil Lomax 1982 23 -1.6% 2.4% Troy Aikman 1990 24 -1.5% 2.8% Steve Deberg 1978 24 -1.4% 2.6% Ken O'Brien 1984 24 -1.3% 3.0% Don Majkowski 1988 24 -1.2% 2.7% Joey Harrington 2003 25 -0.9% 3.1%
successes who threw 160 or more passes as a non-rookie, by age 25 or under, worst interception percentage
first last year age adj int% Vinny Testaverde 1988 25 +3.6% 7.5% Steve Deberg 1978 24 +1.9% 7.3% Jim Zorn 1977 24 +1.9% 7.6% Steve Grogan 1976 23 +1.8% 6.6% Jim Plunkett 1972 25 +1.7% 7.0% Wade Wilson 1984 25 +1.5% 5.6% Doug Williams 1979 24 +1.4% 6.0% Carson Palmer 2004 25 +1.0% 4.2% Troy Aikman 1990 24 +0.9% 4.5% Kordell Stewart 1997 25 +0.9% 3.9% Joey Harrington 2003 25 +0.7% 4.0%
So, when should the Lions have given up on Joey Harrington? I say after the 2003 season.
I know it may sound drastic to suggest a team should have moved on from a high draft pick after two years, but the chances that Harrington would turn out to be a success were pretty slim. By the end of 2003, Harrington had already thrown 983 pass attempts, and had shown no improvement in his productivity as measured by yards per attempt. Donovan McNabb, the only rookie who Harrington exceeded in yards per attempt, had in contrast made drastic improvement in yards per attempt from year one to year two, made plays with his feet, and led his team to the playoffs in year two. Okay, so he did finish ahead of Wade Wilson in yards per attempt. But I'm not holding out hope that someone might turn out to be Wade Wilson, and Wilson had thrown less than 200 attempts by age 25, while Harrington had thrown over 900 passes.
What about his ability to avoid sacks? Certainly, Joey Harrington avoided sacks very well. If it were coupled with being productive on the plays where sacks were avoided, then we might have something. But it wasn't. And if we account for his sack % and generously assume there were 30 plays in 2003 where he was brilliant enough to avoid a sack and throw the ball away, avoid a grounding penalty, but not complete the pass, then removing 30 from the denominator only moves him slightly in YPA, ahead of Steve DeBerg as well.
It comes down how small the likelihood of success is necessary to exclude a player and risk making a "false negative" decision. The chances of finding a successful quarterback after the early part of the draft who qualifies for the list of successful quarterbacks used in this post is about 20%, and the chances of finding a genuine multi-pro bowl success is still close to 10%. So I don't think your standards have to be of the Lloyd Christmas variety--otherwise you will commit way too many "false positive" errors and continue to hurt your team that way. And after two seasons, the chance that Harrington turned out to produce a later career like Wade Wilson and Steve DeBerg were probably less than 10%, and the chance that he actually turned into a Phil Simms or Donovan McNabb were extremely small.
The Lions gave Harrington two more seasons after I say that the evidence was strong enough to comfortably decide to move on. I don't think Joey Harrington is the worst quarterback of all-time, far from it. I do think that his career certainly wasn't helped by being drafted by a bad organization. And Harrington's place on Chase's list is in large part due to the failures of Matt Millen, both in evaluating talent, and continuing to construct a roster without a legitimate starting quarterback. But you can also count me among the camp that doesn't buy the "The Lions ruined his confidence and he would have been a good quarterback on another team" stance. Harrington may have been Tony Eason in different circumstances, but that's about it. The talent and organization around a quarterback matters, but the eventual successes simply do more with limited talent around them than Harrington ever came close to doing.
But that's all in the past. Let's look forward. This should provide some good guidance on the minimum progress you want to see out of young starters today. For guys like Matt Ryan or Joe Flacco, I wouldn't worry too much about interceptions as rookies, or low completion percentages. Given that neither is playing with the greatest show on turf on offense, I would still like to see yards per attempt at or above 6.2 by season's end--below that and I might have some concerns. For JaMarcus Russell and Trent Edwards, the minimum expectation should be more in line with what we see in the second group of lists.
The prevailing wisdom I get is that teams simply write off the first year or two a guy starts and lump all bad play together. I suggest they use yards per attempt (to counter and/or in conjunction with their own assessment of the player) as a low end exclusionary tool to assess a player, even on a bad team, to see if he is making the minimum progress necessary.
Related posts:
ShareThis
September 19th, 2008 at 11:01 am
What would these stats say about Tarvaris Jackson? How does he compare with Joey Harrington? Should the Vikes have given up on him already? And do your stats say that Gus Frerotte counts as "a veteran quarterback... to legitimately compete as the starter and potentially beat out" the incumbent?
September 19th, 2008 at 2:57 pm
How does he compare to Harrington? not even close to being as bad.
Tarvaris Jackson's stats last year were not stellar, but he is above the bottom 20% of the successes in every category, interception percentage being the worst. In other words, there were enough eventual successes who started similarly to Jackson in year one that we can't write him off.
That said, just because a guy isn't worse than every eventual success doesn't mean he is going to be a star. Plenty of guys started merely below average, like Jackson, and never developed. My sense is it is a little early and a panic move.
As for Frerotte, he is what he is, and that is a journeyman who is now late in his career. If he was 32, maybe, but not at 36. I'm talking more like Pennington this year, or someone like Chandler or Beuerlein a decade ago, a guy that could come in, compete to be a legitimate starter, and hold the position while the next quarterback of the future is developed.
September 19th, 2008 at 7:01 pm
Well, Steve Young only had 138 passes his rookie year, but it seems criminal not to include him. He's the best "bust" ever!
Rookie:
52% comp, 2.2% TD, 5.8% INT, 2.8 ANY/A, 13.2% sack percentage
Sophomore:
54% comp, 2.2% TD, 3.6% INT, 3.5 ANY/A, 11.5% sack percentage.
Tampa traded him to SF after those two years. And well, you know the rest.
September 20th, 2008 at 6:23 am
I am the gangsta of the blog. No, I didn't include him in the "rookie" list, not only by drawing the line higher for pass attempts, but because he wasn't a true rookie, having played in the USFL.
However, thanks for pointing him out, because due to a programming error, I didn't have him in the second group, when he should have been. He was still 25 in 1986 (I had miss entered his age) when he started for the Bucs and threw 363 passes. He certainly put up bad numbers. He did not finish in the bottom 10 in comp%. His yards per attempt ranked him in the bottom ten, right next to Majkowski on the list above, but well ahead of Harrington. His TD% was 3rd worst, behind only Anderson and Plunkett. His Int% was actually slightly better than league average in 1986. The previous year, he did throw a high percentage of interceptions but posted a 6.8 yards per attempt. He actually fits my main points (so I wasn't excluding him on purpose)--dont exclude a guy because of interceptions early, but expect a minimum productivity in yards per attempt (roughly 6.0-6.4 even if on a bad team now).
September 21st, 2008 at 2:41 am
Do you happen to know who was available in the 2003/2004 offseason that they could have gotten? It looks like they picked up Garcia for 2005 and Kitna for 2006... I'm wondering if they were on the same page as you and just didn't see a great opportunity.
Looking at those 2002 and 2003 seasons, it looks like they had a LOT of problems - terrible defense, no running game... Maybe they just had more important areas to work on than QB....
I have this feeling that Harrington isn't really as bad as we think he is. I don't think he's good, but he's never played on a team that was better than 6-10. And I think the Lions were looking for a real gunslinger, and he's not really suited. During the Lions best year with him as QB, he had a QB rating in the high 70s and 19 TDs to 12 INTs. If he were on an average team that wasn't expecting him to win every game with his arm, I don't see why he couldn't be somewhere in the middle of the pack.
September 22nd, 2008 at 12:00 am
Harrington's record compared to other quarerbacks starting for the same teams:
Grand Total
Harrington: 15-29 (34%)
Other QBs: 3-17 (15%)
His record compared to quarterbacks before and after him:
Grand Total
Before/after Harrington: 22-58 (28%)
Harrington: 26-50 (34%)
http://luvnpeas.livejournal.com/5739.html
.
.
[NOTE FROM DOUG: p-squeek, Chase didn't delete your previous comment. I did. Chase never saw it. The reason I deleted it is because it contained nothing but insults and a link. I've got an itchy trigger finger for that sort of thing. The article in your link is interesting and worthy of discussion. Your original comment was deleted because it failed to convey that.]
September 22nd, 2008 at 4:36 pm
I'm not sure I'm following the methodology. It looks like you've sifted through all the data to find the one stat and the one year (the second) in which Joey ranks worst, and then concluded that this must be the one most important. Why is Y/A the best indicator? Why is Y/A in the SECOND year more important, except that Donovan McNabb was worst in his first year? If we believe that Y/A in the second year is the criterion, who are the QBs that we ought to give up on now? Just some thoughts.
September 23rd, 2008 at 5:43 am
Frank, I'm sorry if I gave the impression that I just cherry picked the stat that made Harrington look the worst. The stat that "makes" Harrington look the worst is actually a very important stat. I've done research in the past on later round quarterbacks, http://www.pro-football-reference.com/blog/?p=347 , and also tried to run some similarity scores on the young quarterbacks of last year, http://www.pro-football-reference.com/blog/?p=467. Quarterbacks who throw for low yards per attempt early in their careers rarely pan out. As the lists above (I hope) show, quarterbacks who have a low completion percentage, or throw a lot of interceptions, so long as they are okay at yards per attempt, sometimes do. My concept in this post was to go about it a different way, and look at only the successes, to see the way in which each started his career.
If you want, for comparison, here are all quarterbacks since 1970 who threw the required pass attempts to meet my study in year two of their careers, but failed to ever become quality starters, and had a yards per attempt under 6.0 in their second NFL season (some started as rookies, some did not, just like our successes), at age 25 or younger: (in reverse chronological order)
Brodie Croyle, JP Losman, Kyle Boller, Ken Dorsey, Tim Hasselbeck, Cade McNown, Akili Smith, Koy Detmer, Eric Zeier, Danny Kanell, Rick Mirer, David Klingler, Browning Nagle, John Friesz, Steve Walsh, Billy Joe Tolliver, Jack Trudeau, Todd Blackledge, Scott Brunner, Jack Thompson, Bob Avellini, Gary Huff, Dan Pastorini, Terry Hanratty
September 23rd, 2008 at 6:11 am
p-squeak, you won't get any argument from me that Mike McMahon was a worse quarterback than Joey Harrington. Joey Harrington started from 2002-2004 precisely because he was the best quarterback on the roster. But that doesn't mean Harrington was good or that my point is invalid. Garcia, in five games, put up similar stats as Harrington against a little tougher competition.
But let's look at the W-L record of Harrington. You give him credit, for example, for wins. Does he really deserve credit for a 9-7 Detroit win over Dallas in 2002 when the Lions had under 150 yards of total offense and Harrington was 14-33 for 104 yards, and 2 interceptions?
Let's break down defensive performance in 2002 by QB start. The Lions were 0-4 when McMahon started. The defense gave up 439 yards per game and averaged 1.0 turnovers forced. In Harrington's starts, the defense gave up 363 yards per game, and averaged 1.8 turnovers.
Or I can point out that while Joey Harrington was 4-7 while Garcia was 1-4 in 2005, the Simpson paradox explains the size of the difference in winning percentage. Neither beat a team that finished the year with 8 or more wins. Against teams with 10 or more losses, Detroit (w/Harrington) was 4-0, including a home win over GB when the offense scored 17 and the defense allowed 3 pts, and Garcia was 1-1, including a road OT loss at GB when the defense gave up 16. Garcia got to start against the playoff bound Bengals, Harrington got to come back in the following week to play the 3-13 Saints. Again, do we really want to give Harrington credit for a 13-12 win over the 3-13 Saints in 2005? Or a game where the defense had 4 turnovers (Baltimore) while Harrington finished 10 for 23 for 97 yards and 2 interceptions?
September 23rd, 2008 at 9:39 am
Finally, Pastorini turns up in this post. He's always been my standard for a long term lousy QB - 122 starts, almost all with Houston.
And it looks like he meets all the expectations for the role - 1st rd #3 draft pick, played for a bad team with an active owner, looked good failing.
I think thats the complete package.
September 23rd, 2008 at 11:20 am
Sorry, I wasn’t aware of your earlier posts. Okay, I can assume there’s a correlation. But what’s the reason for this? After all, Y/A itself is not a better indicator of QB greatness than many other measures. (Tom Brady's career figure is well below Tony Romo's, and Donovan McNabb is even with Gus Frerotte.) It also seems to me to be much more dependent on the play-calling and the ability of receivers to make yards after the catch than, say, interceptions and completion percentage. Tom Brady’s Y/A in his first full year was 6.9. He fell to 6.3 in his second year, and bounced back to 6.9 in his third year. He was again at 6.3 in 2006. (Joey had a better number in 2007 than Brady was in 2006.) In 2007 Brady found Randy Moss and Wes Welker under his Christmas tree and Belichick opened up the offense, so he went to 8.6. I don’t know what any of those figures tell us about Brady. So I’m not sure WHY they should tell us anything about Joey.
But it looks like you’re predicting that Brodie Croyle and J.P. Losman will be long-term flops. If they are, we’ll have another piece of evidence that there’s SOMETHING about that figure that’s important.
September 25th, 2008 at 2:20 pm
What about the three year mark for evaluation a player. Both Bill Parcells and Tom Landry were big on this--at three years you had progressed or were on the way out. In fact, Landry considered 5 years the mark for a QB (though Staubach won a SB at year 3 in the league). Just a thought to consider.
September 25th, 2008 at 7:40 pm
Good stuff, JKL. Just got back from vacation and having a few minutes to read it.
I think Y/A after two years is a good stat, but I'd prefer NY/A. With Joey it doesn't matter, but net yards is important because QBs shouldn't get credit for getting sacked instead of throwing incompletions.