From the FanPosts. Wanted to make sure everyone saw this great post. -Joel
The weeks leading up to NFL draft always prove to be fertile ground for thoughtful discussion and excitement among fans regarding which prospects the Chiefs might target in the draft. Countless mock drafts are posted, and conversation generally always seems to revolve around two questions: 1) who should the Chiefs target with their first round pick?, and 2) should the Chiefs target a QB at some point in the draft?
In furtherance of this topic, I thought it might be interesting to start a series of articles focused on a position-by-position quantitative breakdown chronicling the lottery that is the NFL draft. My initial goal was simple: to determine whether those positions seen as "high-value" picks (i.e., QB, LT, DE) justified thier lofty draft status. In other words, you hear pundits talk about picking a LT in the top 5 as a much safer and smarter propsition than selecting a top-5 Safety generally. I wanted to try to shed light on success in the draft, based primarily on position and round selected. As I progressed, I couldn't help but dig deeper into the results, breaking down results by team, year, number selected, etc. This week, I'll post the results for the QBs. I mostly have the other positions finished as well, so if the feedback is positive I can try to make a weekly series out of this. It's a long post, but if you stick with it, it yields some interesting results.
To start, let me outline the process I used in finding the results...
The task is simple: identify every player chosen in the draft since 1989 up to 2008, determine which round the player was selected, and assign the player a rating based primarily on their career stats in relation to other players of the same position. 1989 was selected as a starting point because it provides about 20 years of data (and is the first year I really got into the NFL as a kid). 2008 because it generally takes around 3 years to judge a player.
The rating system is on a scale of 0-5 stars (stars denoted with *), with the general breakdown determined as follows. Note stat guidelines are not concrete, as some players threws loads of INTs to go with their yrds/TDs and had to be adjusted accordingly (cough Kerry Collins, Jake Plummer). Also for those younger players drafted just recently, their ratings are based on current data and projected career data. Aaron Rodgers obviously doesn't have the stats of a Brett Favre yet, but we can all agree he's a 5-star guy.
- 0: These players made almost no impact in the league, with most never playing a snap; Example: too many to name; Stats guidelines: accumulated less than 200 total yards 3 TDs in their careers,
- * Generally had a very minimal impact in their careers, generally either never serving as a team's starting QB or starting for only a year; Example: Brody Croyle; Stats guidelines: 200 yrds/3TDs to 4,000 yards/20TDs,
- ** Below-average players who mostly served as career back-ups in the league, many having starting experience between 1-3 years total; Example: Seneca Wallace; Stats guidelines: 4,000 yards/20TDs to 9,000 yards/50TDs,
- *** Definition of a league-average QB, generally started for around 5 years with average results; Example: Matt Cassell; Stats guidelines: 9,000 yards/50TDs to 25,000 yards/125TDs,
- **** Above-average QBs with several productive years, usually making a few Pro Bowls; Example: Trent Green; Stat Guidelines: 25,000 yards/125TDs to 35,000 yards/200TDs,
- ***** These are Hall of Famers or near HOF players, best of the best with several Pro Bowls and All-AP teams; Example: Peyton Manning; Stat Guidelines: over 35,000 yards/200TDs.
Some of the newer players may need adjusting (likely just a point) as years go on (these still open ratings are in yellow). A perfect example is Eli Manning, who I just bumped up to a 5-star after his second SB win...could still fall back to a 4 dpending on the rest of his career, though. Matt Flynn will be another interesting guy to keep an eye on (I have him as a 2 even with only 1,000 total yrds, because you know he's going to get a shot to start, but I'm still skeptical he'll hit 3 status). Some players are definitely borderline and tough calls, so debate in the comments if you think a rating is wrong.
Here are the edited results for all players with career total yardage exceeding 5,000.
|Passing||Pass + Rush|
|1989||T5||DAL||Troy Aikman HOF||6||12||32,942||165||141||33,958||174||5||2.1|
|1989||2||SDG||Billy Joe Tolliver||0||3||10,760||59||64||11,080||64||2||0.9|
General Results by Round:
Note I use a baseball-type replacement level theory, as basically 0-2 star players are very marginal back-ups and can likely be found at any time through free agents off the street. The real goal of the draft is to find IMPACT players. Thus, the "S-Rate" is Success Rate in finding impact players (4-5 star players) and "A-Rate" is Acceptable Rate (3 star players). Bust-Rate measures those 0-2 star players. A value is assigned to give an overall number that measures the weighted sucess of 3-5 star players in relation to # of players selected overall (3, 4, and 5 points for 3, 4, 5 star players).
|0||*||**||***||****||*****||S Rate||A Rate||Bust Rate||Value|
- 258 total QBs have been drafted since '89.
- 3.9% of QBs drafted have been 5-star players, wither another 5% registering 4 stars.
- 21% of all QBs drafted have been "acceptable" (3-5 star players). This means the total bust rate is 79%, or basically 8 out of every 10 drafted.
- Top-5: QBs selected have a 50% S-Rate. Thus, 11 of the 22 QBs drafted since '89 have turned out to be a 4 or 5 star, franchise-level QB. 50/50 shot of getting the stud QB your team needs. Six more QBs have been acceptable, so only 23% Bust-Rate.
- Rest of 1st Rd: Bust Rate skies to 55% for QBs selected and Total Value of QBs selected falls to almost half.
- 2nd Rd: Good luck finding a franchise QB here: only a 9% S-Rate. Value of QBs selected decreases slightly from the 1st round.
- This shocked me: There was virtually no variation in results for QBs selected in the 3rd and 7th+ rounds. Basically falls off a cliff after the 2nd Round. The Bust Rate stayed between 88% and 96%, with values staying mostly contstant. Overall, from Rounds 3-7 combined, the Success Rate for all QBs drafted was only 2.6%. This a HUGE HUGE contrast with other positions, which follow a more steady decline in value as the rounds progress as we'll see later. This makes sense as teams like to "reach" for QBs early in the draft.
- Only 1 5-star QB was drafted after Round 2: Tom Brady. There have been 214 total QBs taken in these rounds. So even though pundits say "look at Brady, you do't need to take a QB in Rd 1"...yeah...no.
- Worst Year for QBs based on Value--2007: none of the QBs have even hit the 3 star mark. The only hopeful appears to be Kevin Kolb (unless you're still a huge Thigbone fan).
- Best team at drafting QBs based on Value +/- (this is based on how much value you got compared to expected value from the slot where the QB was picked--can discuss in the comments if you're interested in knowing where I got "Expected Value"): Green Bay Packers. They've racked up +15.45 points in QBs drafted, nearly 5 times the league average. And they didn't even get credit for Favre! (see Falcons). But they did draft Mark Brunell, Hasselbeck, Aaron Brooks, and Aaron Rodgers. Notice all but Rodgers were drafted after the 4th round and made their names with other teams, with GB getting good trade compensation for these guys. Seems to be Pioli's strategy with the "draft a QB late every year talk."
- Worst team at drafting QBs: take a guess...I'll give you a second. That's right, your KC Chiefs!! No team has been worse at drafting QBs in the last 23 years. This doesn't even include the Todd Blackledge clusterfuss! Just edging out the dreaded Raiders, we're actually only 1 of 5 teams to return negative value of our QB selections (take a bow Mike Elkins and Matt Blundin).
- Biggest Bargain based on Value +/- : Tom Brady obviously. Next, Brett Favre (only other QB taken in Round 2 and drafted after 2 other QBs).
- Biggest Bust: Jim Druckenmiller, SF. First QB taken of the '97 draft, only 1st Rounder to register a ZERO rating.
- Looking at every Super Bowl from 1990 to current, ALL TEN 5 STAR QBs ON OUR LIST STARTED THE BIG GAME. Of the 4 and 5 stars, 61% played in the big game (Trent Green being the second best never to make it behind Brunell, and at least he played in a conference title game). Amazingly, only 1.3% of all other QBs (0 to 3 star) played in the SB. If this doesn't already demonstrate the importance of the QB in winning the Lombardi trophy, just wait until we compare these numbers with other positions.
|SB Players # % of class|
|QB||0||*||**||***||****||*****||S Rate||A Rate||Bust Rate||Value|
The overall conclusion: QBs are important. Franchise QBs basically have around a 2/3 shot of playing in the Super Bowl at some point during their careers, while all other QBS have a 1/100 chance. So if you don't have a franchise QB, you need one. If you plan to draft one, take him in the top-5 and you've got a 50/50 shot. If you don't have a top-five pick, you better hope you can land on the right side of the 23% chance in the rest of Round 1. If you don't pull the trigger in the first two rounds, you might as well not draft one.
Anyway, there's a lot of information I didn't post. And I don't know how to post all the pretty graphs I have. This stuff really gets interesting when you add other positions to compare and contrast.
Note: all data taken from pro-football-reference.com