K.C. Joyner has a mission. He wants to change the way people understand football, in the way that Bill James and his disciples have changed baseball analysis over the last 30 years. K.C. has been analyzing games to better his own understanding for close to 20 years and this summer has published Scientific Football 2005, the most thorough breakdown of the NFL passing game available. Sports Illustrated's NFL writer Paul Zimmerman profiled K.C. last week, and has brought him some welcome attention. K.C. was kind enough to find time for Cowboys Blog.
I solicited questions from the community and received many worthy ones. Given K.C.'s busy schedule, I submitted thirteen, to give readers a feel for his methods. I am hopeful that we'll get more opportunities to pick his brain later this summer.
1. When did you get involved in the the analysis business?
Joyner: I've been analyzing games since 1986 using a lot of different systems, most of which are based on actual scouting systems used by NFL teams. My football friends thought it was a great thing to get scouting information from the tapes, so I knew there was an audience out there for game tape based player analysis. I had always wanted to write a book on the subject, so I finally got the nerve to do so in 2003.
2. This is your first book. Have you been publishing articles, working on radio, doing TV in the meantime?
Joyner: I had some work published by the Pro Football Researcher's Association and had also submitted my work to a lot of places. Football research isn't seen as being as popular as baseball research, and it's one of my goals to change that perception.
3. The SI story mentioned you use tape. What, in brief, is your "method?"
Joyner: I chart every passing play using a number of play-tracking and player-performance metrics. The play-tracking metrics include depth of drop, type of route run, area of field route run to, receiver, defender, type of defensive coverage, and result. The player performance metrics include how open the receiver was, whether the defender had good or tight coverage, how accurate the pass was, whether the pass was dropped, whether the defense was playing prevent, and how far off the defender was at snap, among others.
4. What types of analysis are covered in the book? What positions are reviewed?
Joyner: The book contains an in-depth review of the passing game, so it includes analysis of WRs, TEs, QBs, FSs, SSs, and CBs.
5. How long does it take you to break down a game? You claim to have broken down almost the entire '04 season. How many hours per week are you watching tape?
Joyner: It takes me approximately 2-3 hours just to break down the tape, but that doesn't include the other parts of the analysis like data mining and writing up the analysis. I did break down every game that I had from the 2004 (about 95% of the games), but most of the actual breakdowns were done during the 2004 season. I spend about 40-50 hours watching tape and then some additional time actually analyzing the breakdowns.
6. Do you have "formal" training? Are you a former player? A scout? A coach?
Joyner: I don't have any "formal" training per se. I have a very large football book collection, including a large number of scouting and player-analysis-based books, so I do have a large amount of self-study training. The analogy I use for this is one I found in the telecom world. The best reference book for technical terms in telecom is Newton's Telecom Dictionary. Any technical telecom person worth his salt will have a copy of this book, as it simply defines the terms used in telecom better than any other book. Harry Newton writes the book and he has absolutely no formal training in telecom. He has a Ph.D in another field, so he's very smart, but he isn't an engineer by trade. He just loves telecom and studies the hell out of it, and from this was able to create the quintessential telecom guide. I like to think of my entree' into the football world as being similar to Dr. Newton's in telecom.
7. What can your book offer that the preseason annuals can't provide?
Joyner: The best way to put is that my book goes into much greater depth than the preseason annuals. It is 140,000 words and 467 pages, and believe me when I say that the content isn't fluff. I have a NO-HYPE editorial policy, and you'll find the analysis in the book more direct and without hype-based bias. From the feedback I've received on the book, it is this lack of hype that is very popular.
8. Do you have a QB ratings system and now does it differ from the NFL's?
Joyner: I want to create some rating systems, but the gist of my analysis this year was to create and track different metrics for every position. For QBs, I offer ratings in a large number of categories, including the percentage of time a QB makes a bad decision, how accurate a QBs passes were (not just his completion percentage, but how many times he threw a pass that a receiver could catch), and also have a detailed analysis of how successful each QB was at throwing at each depth level as defined by NFL passing trees. Two good examples of this are Jake Delhomme and Daunte Culpepper. Delhomme actually threw more deep passes than any other QB in the league, and Culpepper was probably the best QB in the league on short passes. I think that Culpepper's success at short passes led the Vikings to rethink their offensive philosophy (Culpepper was almost phenomenal in his short pass success) and that's what led to their allowing Randy Moss to go.
9. In your system, assess Drew Bledsoe's strengths and weaknesses.
Joyner: I'll do you one better than that, if you don't mind sparing the space. I'll actually give you part of my offensive analysis for the Cowboys. It includes a comparison of Vinny vs. Drew.
One of the most intriguing questions this offseason is what will Drew Bledsoe be able to bring to the Cowboys that Vinny Testaverde didn't. Take a look at their stat comparisons.
I'll start by comparing the totals. Their attempts and completion percentages were very close. Vinny had a higher yards per attempt, which is kind of surprising considering Bledsoe was much better on deep passes and had Lee Evans as a vertical threat. Vinny had almost no one as a vertical threat but Vinny's yards per attempt were higher on short passes and much higher on medium passes. Their percentages of receivers open by X steps were very similar, as was their throwing into tight/good coverage, but Vinny was more accurate on his passes.
I think the biggest difference is their decision making. While Drew wasn't great at avoiding bad decisions (tied for 17th in bad decision percentage), he did make fewer of them than Vinny (26th in bad decision percentage) and the bad decisions he made weren't as bad. Vinny handled the blitz very poorly and forced passes into coverage. Bledsoe won't do that in large part because he'll hold onto the ball for so damn long.
The Bills only allowed 2 more sacks than Dallas did, and only had 4.5 more coverage sacks (8.5 to 4). Those numbers don't look like much of a discrepancy until you realize that Vinny threw many more deep passes than Drew (93 to 54). It very well might be that Buffalo didn't throw the ball deep because of Bledsoe's issues with the pass rush (see Buffalo comment). This could still be an issue for Drew and Dallas will have to pass block better than they did last year or their vertical game could be crippled because of this.
A synopsis of their pros and cons:
Although Testaverde has more pros and less cons, Bledsoe's pros are very big. Vinny really started struggling more as the season progressed, and it did look like he was running out of gas. At this point in his career, Vinny is probably much better suited to being a backup. The fewer bad decisions are a huge plus for Bledsoe. Since both QBs have trouble with defensive pressure, the Cowboys offensive line would be a big factor for either of them anyway. All in all this is a good tradeoff for Dallas.
10. Do you think Bledsoe will improve in the Dallas offense, decline or stay about the same?
Joyner: Parcells knows what Drew can do and won't ask him to do things he's not capable of. I think the Bills staff was asking Drew to do things he either didn't like to do or wasn't very good at (which I cover in greater detail in the book).
11. Anthony Henry got a big contract from Dallas, but he was the least known of the major CB free agents. How did he play in '04? Where does he fall in the CB hierarchy?
Joyner: It was really hard to grade Henry in Cleveland because the Browns always had their CBs playing 7 yards off. They didn't want their CBs giving up big plays and let them give up other types of plays because of it. Henry's stats were only OK because of this, but I don't know that it's an accurate representation of his talent level. It looks like he has more talent than the other Cowboy options, so I think in all it's a good pickup, but I don't know how well his skill set meshes with a tighter press coverage scheme the Cowboys will likely ask him to play.
12. Terrence Newman regressed last year. Can you give a brief rundown of his game?
Joyner: Again, I'll give your readers a freebie.
His stats for 2004 were excellent. His completion percentage was 8th best, his yards per attempt were tied for 8th best, and he had the 14th best tight/good coverage percentage.
He had terrific short passing stats, with the 7th lowest completion percentage and 17th highest tight/good coverage percentage. His medium stats weren't as good as the short, but they were good. He ranked tied for 15th lowest completion percentage and tied for 26th in tight/good coverage percentage at the medium level. His deep stats were very solid as well.
If you didn't know any better, you'd think Newman had a pretty good season. What he actually had were some really good games paired up with some terrible, awful games. Take a look:
Terence Newman's worst games
Wk. Opp. Att. C. Yds. TD
1 Minn. 6, 5, 74, 2
3 Wash. 14, 9, 75, 1
6 Pitt. 12, 9, 104, 1
7 G.B. 5, 3, 92, 1
13 Sea. 14, 12, 147, 1
Totals 51, 38, 492, 6
Terence Newman's best games:
Wk. Opp. Att. C. Yds. TD
2 Cle. 10, 1, 12, 0
9 Cinn. 5, 1, 2, 0
12 Chi. 4, 0, 0, 0
15 Phi. 5, 2, 13, 0
17 NYG 6, 2, 23, 0
Totals 30, 6, 50, 0
Newman was perfectly capable of losing a game all by himself, and perfectly capable of shutting an opponent down. His shutdown games came against teams with weaker receivers, but he could still shut those receivers down.
Newman had many games where he was targeted and beaten. Parcells at one point said that Newman was losing his confidence, and the Seattle game was the nadir of that time. After the Seattle game Newman started playing the receivers much tighter than he had been up to that point. It's almost like Parcells told him, "Look, you play best when you play tight. When you play soft you get beat. Go back to playing tight and stop worrying about if you're going to have help over the top. If you get beat deep because of that, you'll never hear me say a word." Newman's confidence returned and he started playing like a top-level cover corner and had two of his best games in weeks 15 & 17.
They always say that the best CBs are the ones who have bad memories and don't remember being beaten. Newman had his first struggles last year but he seemed to forget them after a time. I was worried about his future, but I'm not anymore. This guy is good as long as he plays tight and doesn't worry.
13. How will Henry's addition, in your opinion, affect Newman in '05?
Joyner: It all depends. Over the past two seasons it seems that the Cowboys' opponents have decided they wanted to pick on a particular CB. Sometimes it was Mario Edwards, sometimes Newman, or sometimes it was the alternating CBs opposite Newman. If Henry plays solid and Newman continues his end of 2004 performance, I think that the Cowboys opponents will stop trying to target one of them and just go their best receiver.
Again, my apologies for keeping you waiting. I'll do better on the next set of questions!