A case for Crabtree
I cannot take credit (or blame) for this one. The following is a detailed, well-researched suggestion from Chiefs fan Shawn Siegele, that Kansas City should draft former Texas Tech wide receiver Michael Crabtree with the No. 3 overall pick instead of linebacker Aaron Curry.
We like Curry an awful lot, but Shawn makes some very good points. I'm not sure what our guest contributor does for a living, but he talks an awful lot like an economist or statistician. Because I talk like a sports writer, I'm deferring many of these points to him. And instead of rewriting what he said in my own words, I'll step aside. Mr. Siegele, the floor is yours:
I went back through every draft since 1999 and looked at the top 6 draft slots. During those years, only 3 LBs have been taken in the first 6 picks. Of those three, two have been a solid players and one appears to be a bust. For that same time period, seven WRs have been taken. Of those players, 5 have become stars and two have been busts. Neither LBs nor WRs are taken in the top six as often as QBs or LTs, but WRs are twice as likely to be drafted that high and they are much more likely to become big time NFL players. You've often pointed to Charles Rogers as a reason to avoid WRs, but WRs taken that highly in the draft actually have an incredible rate of success.
In fact, here are the rates of success at all positions:
QB 3 Stars 3 Solid 7 Busts 13 Total
LT 3 Stars 2 Solid 3 Busts 8 Total
DT 1 Star 1 Solid 4 Busts 1 Uncertain (Dorsey) 7 Total
RB 2 Stars 4 Solid 2 Busts 1 Uncertain (McFadden) 9 Total
DE 2 Stars 2 Solid 1 Bust 1 Uncertain (Long) 6 Total
DB 2 Stars 2 Solid 1 Bust 5 Total
If you rank the positions in terms of most top 6 selections you get this:
1. QB 2. RB 3. LT 4. (tie) WR and DT 6. DE 7. DB 8. LB
If you rank the positions by percentage of NFL stars they produce (eliminating the three players from last year's draft for whom the jury is still out) you get this:
1. WR 71% 2. (tie) DB and DE 40% 4. LT 38% 5. QB 23% 6. RB 22% 7. DT 14% 8. LB 0%
If your rank the positions by the unlikelihood of drafting a bust you get this:
1. (tie) DB and DE 20% 3. RB 25% 4. WR 29% 5. LB 33% 6. LT 38% 7. QB 54% 8. DT 67%
While any sample limited to the top 6 players over a 10 year time period is going to have blips in the data that don't hold up long term, the anecdotal evidence strongly supports three conclusions: 1) WRs are drafted at the top of the draft more often than linebackers, suggesting they are valued by NFL teams more highly. 2) Contrary to conventional wisdom, WRs produce the highest percentage of superstars of any position. 3) WRs produce a relatively low percentage of busts, significantly less than the two most highly sought positions in the draft (WR, LT).
Even after the signing of Bobby Engram, WR still joins LB and DE as the positions of extreme weakness on the Chiefs roster. I think in order to draft Aaron Curry over Michael Crabtree, one would have to believe Curry is a far superior NFL prospect. Considering Crabtree's unparalleled college career and very close physical similarity to NFL stars Andre Johnson and Terrell Owens, I think that is a difficult argument to make.
There are two main myths I’m most concerned with related to drafting Crabtree. The first is that WRs are a risk early in the draft, which I sent you info purporting to debunk earlier. The second is that having two elite WRs is a luxury NFL teams can’t afford. The biggest issue with drafting either Curry or Crabtree is obviously the opportunity cost of not getting to draft the other guy. To address this, I wanted to look at the success rate of teams with two elite WRs compared to those with one elite WR and those with zero elite WRs. (Though the label “elite” is somewhat subjective, I tried to define elite WRs as those with a reasonable expectation of 1000+ yards and/or 10+ TDs; receivers who would otherwise qualify but did not play or played with serious injuries I did not consider elite for the year of their injury.)
The results are somewhat staggering and surprising even for someone like me who is a big believer in the value of multiple elite WRs.
In order to get a large enough sample size, I used the 2004-2008 seasons. (It was unbelievably difficult trying to make all the numbers add up.) During those seasons, I charted 27 teams with 2 elite WRs, 67 teams with 1 elite WR, and 66 teams with 0 elite WRs. I didn’t find it surprising that teams with 2 elite WRs were the serious minority, but it was a mild surprise that so many teams opted to start a season with no elite WRs.
During that time period, those teams finished with the following records (ignoring the one tied game).
2 elite WRs: 273-157 (63%)
1 elite WR: 511-560 (48%)
0 elite WRs: 495-562 (47%)
During that time period, teams with 2 elite WRs had a cumulative winning record every year. Teams with 1 elite WR had a cumulative losing record every year except 2005, and teams with no elite WRs had a cumulative losing record every year except 2006.
Obviously these records are likely to influenced by outliers. The Colts with Harrison and Wayne bring up the overall record of teams with 2 elite WRs and that could be attributed just as much to Peyton Manning. Meanwhile, the Titans and Ravens bring up the record of the teams with no elite WRs. As a result, a secondary analysis is important.
Teams with 2 elite WRs:
Number of times a team had a 10-6 record or better: 15 (55%) Number of times team had a record of 7-9 to 9-7: 10 (37%) Number of times a team had a record of 6-10 or worse: 2 (7%)
Teams with 1 elite WR:
10-6 or better: 18 (27%)
7-9 to 9-7: 28 (42%)
6-10 or worse: 21 (31%)
Teams with 0 elite WRs:
10-6 or better: 18 (27%)
7-9 to 9-7: 19 (29%)
6-10 or worse: 29 (44%)
Almost all pundits will suggest that having two elite WRs is unnecessary and keeps you from addressing other aspects of your team. This is exactly the argument so many people are currently advancing in regard to the Cardinals possible trade of Anquan Boldin. In fact, this couldn’t be further from the truth. These stats would argue that the single most important thing you can do to create a perennial playoff team is to pair your elite WR with a second elite WR. An unbelievable 55% of teams with 2 elite WRs have been a playoff-caliber 10-6 or better during the last 5 years. Even more shockingly, only 7% of those teams have been a non-competitive 6-10 or worse. (One of those teams was an early version of the Fitzgerald/Boldin Cardinals, a team quarterbacked by the trio of Josh McCown, Shaun King, and John Navarre; if one credits being three minutes from a Super Bowl championship a fairly successful season, things have turned out pretty well for the Fitzgerald/Boldin combination.)
The numbers for the remaining profiles stand in stark contrast. Teams with 1 or 0 elite WRs put out playoff squads a paltry 27% of the time. For teams with 1 elite WR, the numbers suggest the likelihood for mediocrity. Teams with no elite WRs are simply bad a disturbingly high percentage of the time.
Most fans and pundits view an offense with multiple elite WRs as a luxury that is fun for the fans but counter to winning. I hope this proves otherwise. While teams with multiple elite WRs represent only 17% of the team seasons in the last five years, they represent 29% of the teams with records of 10-6 or better. They are also disproportionately represented on both the winning and losing sides of recent Super Bowls.
While the track record of top 5 WRs suggests that they are very likely to become elite contributors, WRs in the 7-15 range have had a very high bust rate over the past 5-10 years. At least recently, LBs drafted in that same range have been very big contributors. If the Chiefs had a pick in that range and the choice was Jeremy Maclin or Aaron Curry (or whomever the second best LB might happen to be), then I’d say history favors taking a LB. However, if the Chiefs have a chance at a guy like Calvin Johnson, Andre Johnson, Larry Fitzgerald, or Torry Holt, I don’t think there’s a way Curry could possibly match that contribution even if he became – as many project – the best LB in the NFL.
Considering the monster Crabtree was in college even on a badly injured ankle, I think he’s the best player in the draft (which I believe was the consensus before all the irrelevant stuff with the minor foot injury came up). A small aside: Pioli was involved in a deal to bring an unproven former Texas Tech WR to the Patriots and gave up a 2nd and 7th round pick to acquire him. Since the Pats acquired Wes Welker, they’ve gone a combined 29-6. And while Dwayne Bowe is no Randy Moss, their respective Red Raider careers suggest Michael Crabtree has a much higher ceiling than Wes Welker.
Submitted by Kent Babb on March 18, 2009 - 9:28am.
login or register to post comments
Possible answers for critics of the Crabtree pick
Submitted by S Siegele on March 18, 2009 - 6:35pm.
I'd like to give a huge thanks to Kent for posting my analysis and thank everybody for commenting. I wanted to address a few of the best and most prevalent rebuttals.
1. It was pointed out that I didn’t specify a criteria for the ‘star’ ‘solid’ and ‘bust’ labels and that as far as anyone knows I don’t have any special football knowledge. That’s a fair criticism and I admit openly that my bias toward the Chiefs drafting Crabtree is likely to skew the data. That said, assigning those labels was actually much easier than deciding which receivers qualified as “elite” even though I tried to use a fairly objective measure for that. The players drafted at the top of the NFL draft are high profile players. If they have appeared in multiple Pro Bowls – and for the more recently drafted players, if the consensus about their career trajectory has them projected to multiple Pro Bowls – then they qualify as stars. If they were unable to hold down starting positions, then they are busts. If they start but are not considered much if any above average for their positions, then they are solid. I expect if most people did their own analysis they might come up with something different, but based on how few borderline calls there were, I think the change would be slight.
2. Some posters pointed out that in the 7-15 range there have been a lot of busts for WRs and that first round LBs have done well. I agree with this and even note it in my original statement. The fact is that there is a vast difference between that first tier and the second tier. I have tracked the WR drafts back to 1999 and the likelihood of drafting a good WR plummets with each subsequent draft slot. While I won’t bore you with the long, tedious numbers, WRs in the Top 6 are nearly surefire superstars. WRs in the first round have a high success rate as starters or above. WRs in the second round have a decent success rate, and WRs after round 4 have almost no success rate at all. In fact, since 1999, WRs taken in the 7th round and as undrafted free agents have vastly outperformed those taken in rounds 4-6. If the Chiefs pass on Crabtree, they might as well look at those guys who don’t get a call on draft day as waste a draft pick. (I also don’t buy the argument that there are often multiple elite WRs and that messes with the calculus. The WR who just misses my arbitrary cutoff, Troy Williamson, was not considered an elite prospect by most scouting services, and was seen as a huge reach by most people during that draft. It was a case of drafting for need and trying to appease a fan base who had just seen Randy Moss traded. I also don’t buy the argument against the Detroit Lions. Their problem was not drafting WRs, but being so awful in player evaluation that they kept taking bad ones. They do have the misfortune of drafting one of the only two top WRs to ‘bust’ during the time period I looked at, but they also used picks in the 7-15 range to draft a mediocre WR in R. Williams and a massive bust in M. Williams. Had they followed up on the pre-draft red flags regarding Rogers drug use and drafted A. Johnson instead, they could now theoretically be in the position of having the two most gifted WRs in the NFL along with having the draft picks Jerry Jones mysteriously gave them for R. Williams. Of course, if they had taken Andre Johnson, they wouldn’t be drafting #1 right now either.)
3. The strongest counterargument against my contention that having 2 elite WRs is important is simply the list of recent Super Bowl winners. I don’t expect die-hard ‘defense wins championships’ folks to buy this argument, but again the important thing is likelihood of victory in any individual playoff game. Using the numbers for 2004-2008 again, teams with 2 elite WRs were 16-13, teams with 1 elite WR were 19-22, and teams with no elite WRs were 20-20 (that's hardly the statistically significant difference I'd like to see, but it is still better than the other two profiles). It’s also important to keep in mind that while teams with two elite WRs are a limited minority of all NFL teams, they represent nearly a third of playoff teams. However, if 71% of playoff teams do not represent the suggested profile, the non-profile teams are still likely to have a lot of total playoff victories (and thus Super Bowl victories) because of their sheer number of horses in the race. It’s also important to keep in mind that what did happen is possibly not as important in preparing for the future as what was most likely to happen. I think most fair-minded people would grant that if the 2007 Super Bowl were played 100 times, the Patriots would have lost only a handful of times. It’s also worth noting that the 2008 Steelers finished 29th in the NFL in yards per carry and had an anemic 2.9 ypc in the playoffs. Despite their vaunted defense, they found themselves behind with roughly three minutes to play in the Super Bowl. On the other hand, they do have Hines Ward, a player who in the last 8 years has averaged over 1000 yards receiving with numerous double digit TD years, and Santonio Holmes who averaged 1000 yards per 16 games played over the last two seasons while finishing in the top 5 in yards per catch during that time. Lucky for the Steelers they made that WR investment as it led directly to a Super Bowl title on the final drive.
4. The other compelling counterargument is the presence of Tony Gonzalez. For whatever reason, having an elite receiving TE to pair with an elite WR does not seem to convey the same advantage as having multiple elite WRs (although there aren’t many so that creates a sample size problem). In the Chiefs specific case, there are two more problems with relying on Gonzalez instead of drafting a WR. First, splitting Gonzalez out wide in multiple receiver formations robs the Chiefs of a good blocker and the important element of disguise when he releases from the line into a pattern. Second, Gonzalez is 33. Unless we’re closer than most dare hope, neither Gonzalez nor Engram will still be on the roster when the Chiefs are ready to compete for Super Bowls. In a couple of years, the Bowe-Crabtree combination would just be entering the peak of their careers.
5. And finally, the argument that there’s no way the Pioli/Haley braintrust will take a WR because they’re about substance over style and will build the team from the inside out. At this point, even casual NFL fans realize the extreme importance of offensive and defensive linemen (although the Super Bowl champion Steelers had a terrible offensive line in 2008). That doesn’t mean the skill positions represent style over substance. In fact, that’s the whole point my admittedly rough analysis is supposed to make. And this is a fact not lost on the top NFL minds. Even after having won 3 Super Bowls, the Belichick-Pioli tandem made an obvious decision to move in the direction of a multiple-WR juggernaut in 2007. Since that time, they’ve gone 29-6 (despite playing a lot of those games with a QB who hadn’t played since high school). That they haven’t won a Super Bowl with that approach is largely a historical accident and the fact that in any given year the best team still has odds of being Super Bowl champ only in the 25-40% range. The most successful regular season team during this decade is Indianapolis, and Bill Polian has suggested WR is his top target in the offseason despite the presence of Reggie Wayne and Anthony Gonzalez (a first round pick who has been the 1st and 3rd most efficient WR on a per play basis over the last two years according to Football Outsiders). Considering the Chiefs have used a disproportionate number of draft picks on defense over the past 8 years and drafted their LT of the future last year, the presence of a talent like Crabtree at the top of the draft presents the Chiefs a unique opportunity to employ a strategy recognizing the fact that in this case style is substance.