The Watchtower Review

The point of the Watchtower posts was to forecast the performance of the Detroit Lions against their weekly opponents. From the start, I’ve used historical performance data of the Lions coordinators against their opposition’s. By controlling for the relative talent of the players, I tried to isolate systemic advantages at the X-and-O level. I then tried to apply those advantages to the teams’ current skill levels, and project a result.

The Watchtower is one of my most popular features; people really dig it. It’s fun to write, especially researching every coordinator’s coaching tree, and picking the picture. However, after three years, I’m no longer satisfied with The Watchtower an alternative “game preview,” or as a predictive tool.

Watchtower Problem #1: heavy reliance on per-game team averages.

When I use average yards per attempt and average yards per carry, it gives a pretty accurate picture of those players’ performance levels. Whether a quarterback has 25 or 50 attempts, or 200 or 400 yards, dividing one by the other tells you at what rate the quarterback is generating offense, every time. But dividing “points scored in a season” by “games in a season” doesn’t work. A “game” is not a fixed unit of measure; there’s a wide variance in the number of possessions and plays in a “game.”

In every pass attempt, there is exactly one pass attempt, one bite at the apple. In every game, there’s a wide variance in possessions, time of possession, and plays. Example: when the Lions hosted the Vikings, they scored 34 points. When they hosted the Chargers, they scored 38. On the face of it (and in terms of the “points per game” numbers I’ve been using), the offense was very effective in both games.

However, in that Minnesota game the offense netted just 280 yards and 20 points from ten possessions. Against San Diego, the offense netted 440 yards and 31 points from eight possessions. This is a massive difference in effectiveness and it’s almost completely uncaptured by the current Watchtower methodology.

Dropping the "per game" team averages would allow me "tell the story" more effectively; I thought there was a very high chance that the first Packers game would be shockingly conservative—and the rematch a track meet. But using season average against season average, there’s no way to project either of those outcomes.

Finally, that "track meet" effect means something: there is a tendency for points to follow points, and that speaks to a very real offense/defense interaction effect that isn’t accounted for, either in traditional analysis or in The Watchtower. When one offense puts the pedal to the metal, the other one follows—and both defenses, apparently, just let it happen. Why? What’s going on here?

Watchtower Problem #2: No real accounting for turnovers or special teams.

This is one that’s bothered several readers from the get-go. The Watchtower is a study of offense-defense interaction: what happens when offensive scheme A meets defensive scheme B. But special teams and turnovers play a huge role in the final score.

In the Thanksgiving Day game, when the Lions and Packers played to a stalemate for most of the first half, a tipped pass fell into enemy hands and the Packers’ offense got to start deep in the heart of Lions territory. That was the game-changing play both teams desperately needed. Despite incredible down-to-down play by the defense, the offense was really the unit that put the Packers in position to score.

On special teams, the Lions’ coverage units struggled mightily throughout the first two thirds of the season, and it regularly hung the defense out to dry. Moreover, the iffy upfield blocking for Stefan Logan (and his own iffy fair catch decisions on kickoffs) failed to make the field shorter for the offense.

Watchtower Problem #3: The Human Element.

I project ranges for points, passing effectiveness, and running effectiveness for each side—then basically use the “Mitigating/Aggravating Factors” and “Conclusions” section to winnow those down to the final score I deem “most likely,” usually via talking-out-loud thought experiment.

There are several layers of my own bias involved here—and even though I work hard to follow where the data leads me, a little bias on top of a little bias on top of a little bias makes a big difference. I can definitely lead the statistical horse to water if I want to—and sometimes I do even when I’m trying not to.

What I’d love to be able to do is project a range of possible outcomes and their probabilities, so when I say “The most likely outcome is . . .” my a hand won’t be moving the data’s mouth.

Thoughts?

Read more...

Detroit Lions at New Orleans Saints Retrospective

I have a thing about justice.

I want things to be right. I want things to be fair. I want the playing field to be level. I want people to get what they deserve. When I hear about a story or a news event without a clear right or wrong, I consume all the information about it I can; I have to know what I think is right.

Yesterday I sat down with my third cup of coffee (more or less (more)) and put my hands on the keyboard. My thoughts weren’t ready yet. Perhaps it was the short turnaround from Sunday Night Football to Monday morning, but the events of the Saints game had yet to percolate through my mind. With nothing but the title typed, I stopped “writing.”

Last night I didn’t go near the computer. I didn’t write a word. I let my thoughts brew slowly. I let frustration and disappointment dilute in the liquid heat of my id while I played some FIFA and ate pancakes. Eventually, I found the words to “Justice” on my lips. This morning, I went to the coffee pot and found that overnight my post had brewed.

The way fans talk about football with each other is a shadow of the way players talk smack on the field. As I said last week in response to a post on Cheesehead TV, it’s part of the illusion of fandom. We brand ourselves with their brand, we wrap our identity up in theirs, we take their successes and failures to heart, and others project their successes and failures onto us.

Complaining about the refs is for losers. It’s what happens when loser fans watch their loser teams lose. Winners know the better team usually wins. Why do you think Vegas sets a betting line? In the NFL, any fan can pick straight winners far more often than not. When upsets happen, it’s because the “worse” team showed up and played better on that day.

The Saints are better than the Lions; there’s no doubt about it. They’re undefeated at home. They went up to Lambeau in Week 1 and were a yard and/or bad playcall from taking the Packers to overtime. By SRS, they’re the second-best team in the NFC, and fourth-best team in the NFL. As the Watchtower showed, they’re scoring a little more, and allowing a little less, than the Lions.

The game’s stats reflect that exactly. The Lions gained 408 yards passing on 44 attempts; that’s an awesome 9.27 YpA. The Saints, though, threw for 342 on 36 attempts, a slightly-better 9.50. The Lions did okay on the ground, getting 87 yards on 22 carries (3.95 YpC). The Saints were a little better; 100 on 23 (4.35 YpC). The Lions were sacked 3 times for The Lions turned it over once, but only after the game was all but in the books. As I said in the Watchtower:

The Saints are like the mini-Packers, and the Lions are like the mini-Saints. These two teams hold up a mirror to one another, and the Saints are a little bit better in every phase of the game—except the Lions play much, much better pass defense. I could see this going either way.

So what turned my projected 30-28 Lions win into a 31-17 Lions loss? Well, the Lions had one field goal blocked, and missed another—getting zero points from two good drives. But that only makes it 31-23. Where’d that other touchdown go?

Oh, right: the Lions were flagged 11 times for -107 yards, and the Saints were flagged 3 times for –30 yards. The penalty differential was exactly one scoring drive.

It isn’t that the Lions didn’t commit these infractions; they did. But as with last week, the problem is consistency. Offensive and defensive holding, and offensive and defensive pass interference, are subjective calls. Much like charging and blocking in basketball, there are technical definitions but enforcement is done by “feel.”

When Nate Burleson is flagged three times for offensive pass interference—as Pride of Detroit user “NobodySpecial” pointed out, that’s as much as any offensive player in the NFL had been tagged with OPI all season long—the game “feels” rigged.
I mean, here’s the SaintsNation Saints blog:

In my entire life of watching football I've never seen an offensive player get flagged 3 times for the same penalty in a game, especially something rare like offensive pass intereference. That goes back to the poor discipline, but it's also just weird.

I don't believe the referees were being controlled by Vegas, mobsters, Roger Goodell, or any other outside force determined to make sure the Lions lose. But it’s impossible to claim with a straight face that the Lions are being officiated the same way other teams are. For instance: pushing a hand into another player’s face is a penalty, except when it’s a ball carrier stiff-arming a defender . . . unless that ball carrier is a Detroit Lion, then it’s a penalty again.

My personal belief is that the league and/or officials are trying to send a message to the Lions. Now that they’re a “dirty team,” the Lions not only have to play as clean as everyone else, they have to play cleaner. They’re going to get flagged for things no other team gets flagged for. Rough stuff from the other side is going to go unpunished. The league is sending a message to the Lions, and it’s up to them to listen.

For once, the one-way street of fandom cuts both ways: on Sunday, the Lions were just as furious as fans were. They were feeling just as confused, just as upset, just as impotent as everyone hollering at their TVs. They were playing the game just as well and just as clean as the Saints, but the refs were taking the results away from them. It’s not fair. It’s not right. It’s unjust.

Civil disobedience isn’t a choice here; the Lions cannot #OccupyTheNFL. There are two possible responses: have a tantrum, or grow up. Remember when Gosder Cherilus was our resident hothead? He of all people should be the first to jump in and make a stupid play after the whistle—but he got it. He gets it. He lets it roll off his back. The Lions are going to have to learn to do the same.

The great thing is, the Lions are talented enough to do it. They can win playing clean. They can win playing cleaner. They can beat the Vikings, Raiders, and Chargers with one hand tied behind their back—and that’s good, because they’ll have to.

Read more...

Watchtower Review: Lions vs. Packers

 

It’s nearly a week after the fact.

It doesn't feel any better.

I couldn't complete the ritual healing of the Fireside Chat on Sunday night, despite my best efforts. My struggles with my DSL connection provided an apt metaphor for the Lions’ struggles against the refs, injuries, and their own failures. The talent was there, the effort was there, but they shot themselves in the foot and circumstances conspired to prevent them from recovering.

Though I did file my weekly “The Absolutely Worst of the NFL” column for B/R, I didn’t post here yesterday. I’m up to my eyeballs in film, studying what went wrong and how the Lions can fix it. I’m also studying Ndamukong Suh’s claims that he was mugged, and also the bizarre inconsistency in officiating. In the meantime, let’s have a look at that Watchtower.

When Scott Linehan offenses meet Dom Capers defenses, something very interesting happens. It’s one of the strongest statistical trends that has ever cropped up in The Watchtower. First, the Linehan offenses tend to rush much better than expected. Second, they tend to outperform scoring expectations as a result. Third, and most interestingly, quarterback runs are wildly more successful than usual.

Might the Lions draw up some surprise quarterback draws for Stafford? Or, might he scramble for some yardage? It’s something to keep an eye on.

Well, at least that part held up. Stafford scrambled for a career-high 31 yards, including a career long of 21, on a tied-for-career-high four rushing attempts. That and six dollars will get me a latté these days.

It’s hard to project Kevin Smith to repeat his NFL Offensive Player of the Week performance, or anywhere close, on Thursday. But his dramatic welcome-back party plays right into the Lions’ hands. With a back that can take advantage of the Lions’ systemic advantage, the Lions should exceed nominal expectations.

Therefore, I project the Lions to score 33-35 points, passing for 7.0-7.50 YpA and rushing for 4.75-5.0 YpC. I have very high confidence in this projection.

Or, you know, Kevin Smith could go out in the first quarter with a high ankle sprain . . .

Actually, that didn't have much effect on the game. Between Smith, Maurice Morris, Keiland Williams and Nate Burleson, the Lions racked up 136 rushing yards on just 21 attempts, a per-carry average of 6.48 yards. This wasn’t a Javhid-Best-against-the-Bears YpC average inflated by one length-of-the-field scamper, either. The Lions were legitimately running it down the Packers’ throats.

Once again, we see the schematic advantage. When Scott Linehan offenses face Dom Capers defenses, they run wild. It happens every single time.

Passing, not so much. The Lions gained 276 yards on 45 pass attempts, a 6.13 YpA average. Maurice Morris led all Lions with nine receptions. Burleson was second, with five. Calvin Johnson, Brandon Pettigrew, and Keiland Williams all had four. Kevin Smith caught three, Tony Scheffler two, and Titus Young one. Seeing a pattern here?

The Lions played it very, very very close to the vest. Against the Panthers, Stafford followed the mantra of “settle down and execute the offense” I’d been chanting for the previous month—and it served him very well. Against the Packers, his fear of turnovers (or the coaches’ fear of turnovers) turned “settling down” into “going into a shell.”

Against the Packers, the Lions more-or-less abandoned the downfield pass, settling for slants and curls and ankle-biting. If this had actually resulted in a turnover-free performance, the Lions would have won this game. Instead, a batted ball, a picked pocket, and a combination underthrow/mad linebacker hops hurt the Lions like they took bomb-it-downfield risk, with no bomb-it-downfield reward.

The defense gets sacks, gets turnovers, stiffens up on third down, and gets stops. This is doubly true when the offense isn’t going three and out, or turning it over right back, and the coverage units aren’t allowing scores. On Thanksgiving, the Lions’ defense will need the Lions offense to help them get it done.

Without any systemic advantage, I would expect the Packers to slightly outperfom their season average against the 19th-ranked defense. Taking the strong systemic advantage into account, I project the Packers offense to score 27-30 points, passing for 7.00-7.50 YpA and rushing for 4.25-4.50 YpC. I have medium-high confidence in this projection.

This is eerily similar to the very first Packers Watchtower, where my expectations of the Packers’ offensive output was right on the nose, while Daunte Culpepper and Drew Stanton combined for three interceptions and zero points. In both cases, if the offense hadn’t thrown three picks, the defense would have dramatically outperformed expectations.

The Lions defense did a magnificent job against the Packers, holding them to just 86 yards in the first half. Gradually, all four active Lions cornerbacks, plus safety Louis Delmas, went out with injuries. At one point, At one point, #5 safety (and special teams ace) John Wendling was in at cornerback, and WR Rashied Davis was playing safety. With all Lions secondary hands on deck, Aaron Rodgers started moved the ball well with quick passes to the outside—and his targets started to make hay after the catch.

With the secondary in tatters, Rodgers and the Packers passed for 9.30 YpA. Though they rushed for only 2.94 YpC, the Packers scored 27 offensive points, right on the button with my projections. Still, the defense has to be commended for this shorthanded effort in a game where their offense (and one of their own) hung them out to dry.

Read more...

Watchtower Review: Detroit Lions at Chicago Bears

I’ve been procrastinating. I didn’t want to do this. I really, really didn’t. But it’s time to review the Bears Watchtower:

I am extremely, extremely nervous about this game. It was a low-scoring slugfest before, despite only one turnover and four sacks between the two teams—mostly thanks to an absolutely ridiculous 26 total penalties that wiped 198 yards of combined offense off the field.

If the hankies aren’t flying this time like they were a few weeks ago, which offense benefits more? Can two teams currently ranked 1st and 7th in the NFL in turnover margin play another game without turnovers playing a significant role? I’m also haunted by one of my strongest Watchtower rules of thumb: rarely do two teams meet twice in the same season and play to the same result.

Partly because I’m so twitchy about it, I’m going to take the high end of my Bears projection and the low end of my Lions projection; I’ll say the most likely outcome of the game is a 23-21 Lions win.

Ouch. Right. Well.

The story here wasn’t the failure of the Lions defense. They allowed just 16 offensive points, 3.11 yards per carry, and 6.15 yards per attempt. In fact, that’s even better than I projected:

I project the Bears to score 17-21 points, passing for 6.50-7.00 YpA and rushing for 4.75-5.25 YpC. I have medium-high confidence in this projection.

Of course all that doesn’t matter. What matters is the dumpster fire that was the offense, and its six turnovers. The Lions scored just 13 offensive points, and passed for only 5.22 yards per attempt. They did rush for 4.22 YpC, but that doesn’t even sort of matter in the grand scheme of things.

Before this season, when Linehan has faced Lovie Smith, Linehan’s offenses tended to move the ball well—but score fewer points than expected, due to lots of turnovers. It was a really strong statistical trend—but on Monday Night Football, the Lions didn’t turn it over, so I wondered if the offense had evolved to the point that was no longer true.

Nope.

Going forward, there isn’t much to take away. The two lost fumbles that ended the first two drives were total flukes—and the rest of the game, Stafford was pressing like crazy trying to get them back. The Lions need to learn from this debacle, but I’m not sure there’s much we can learn from breaking it down.

Read more...

Watchtower Review: Lions at Broncos

The accuracy of last week's Watchtower was a wildly pleasant surprise--both in terms of the confidence I had in the projection, and in the result itself.

I project the Broncos to score 15-17 points, pass for 5.75-6.25 YpA, and rush for 4.75-5.50 YpC. I have medium-to-low confidence in this projection.

Turns out this gave the Broncos too much credit. The Lions defense hassled Tim Tebow and the offense all day, excepting only the first and last drives (which netted the Broncos their 10 points). The Broncos ran for a ridiculous 195 yards on 30 carries, a stonking 6.4 YpC.

In a lesson I'm trying to drive home with a sledgehammer, THAT DOESN'T MATTER.

Points come out of the passing game, and the Broncos neither passed nor scored. They gained 172 aerial yards on 39 attempts, a miserable 4.41 YpA. They were sacked 7 times for –55 yards, fumbled five times (lost two, including one for a touchdown) and interception (also returned for a touchdown).

The Lions should score 30-34 points, averaging 8.50-9.0 YpA and 4.00-4.25 YpC, and so I will project them to, but at this point I have medium-to-low confidence in this projection, despite having plenty of data.

Bingo. The Lions scored 31 offensive points, passed for 8.50 YpA, and rushed for 4.04 YpC. This officially qualifies as spookily accurate. I thought the Lions could win even bigger or make it much closer, though, and said so:

In fact, what the Lions should score is almost assuredly what they will not; they’re more likely to do much better or worse than to meet expectations square on the nose.

Whoops. The Lions spooked me so bad with the Falcons game that I just couldn’t trust the numbers:

Given that the Lions offensive starters have more talent than the Broncos 53-man roster put together, I’m not sure why I feel so bad about signing my name to that projection. Maybe it’s because I said almost the same score last week and was brutally disappointed. However, Denver is not Atlanta—not even close. And I have to believe the Lions defense will feast on Tebow and the Broncos offense, giving Stafford many bites at the apple. Worst-case scenario, the D should have some chances to take matters into their own hands, as they did against Minnesota and Dallas.

. . . and indeed they did, boosting the Lions up to a 45-10 clobbering.

As I said in the Fireside Chat, this win, in this fashion, with these stats, proves the numbers right; the Lions are who we thought they were. They are one of the best offenses in the NFL, and one of the most opportunistic defenses in the NFL. When both units are “on” they play football at a very, very, very high level.

Read more...

The Fallen Watchtower Review: Lions vs. falcons

I did get something sort-of resembling a Watchtower up before the Falcons game on Sunday, if only just. Without any of the historical breakdown or schematic analysis, I just looked at season ranks and averages. It was done entirely on my iPhone in stolen moments. It looks like it:

The Falcons have the NFL's 17th-ranked offense, averaging 22.5 points per game. The passing attack has garnered a respectable-but-not-great 6.22 YpC, while the rushing game is netting a very solid 4.37 YpC.

The Lions defense is allowing a stingy 19.0 points per game, 7th-fewest in the NFL. Opponents are passing for a miniscule 5.55 YpA, while they're running all over the Lions at a 5.21 YpC clip.

The Falcons power running game directly attacks the Lions defensive Achilles heel. Then again, so did the 49ers, and their offensive output exactly matched my projections. Therefore, I project the Falcons to score 17-20 points, throw for 5.5-6.0 YpA, and run for 5.0-5.25 YpC. I have medium confidence in this projection.

The Falcons scored 23 offensive points, just above what I'd projected. Their passing game slightly outperformed expectations as well, gaining 6.23 YpA--matching their season average to 1/100th of a yard. The running game actually performed BELOW their season average, well below my projections: 4.16 YpC.

This is an example of what I call the "Whack-a-Mole" effect. When a defense has a glaring weakness in one area and they’re going up against an offense with a corresponding strength, often the D will “sell out” the strength of their D to shore up the weakness. The net result is usually the same in terms of team scoring and overall per-play effectiveness.

This is best known to fantasy football fiends who are drooling over starting their stud running back against the worst run defense in the league. Somehow that never quite results in 200 yards and 3 TDs, does it? According to Mike Mady at Scout.com, the Lions were emphasizing the back seven stopping the run against the Falcons; could it be that pass coverage suffered as a result? I’ll go back and look at this.

The end result of the defense's effort would have looked much better if the Lions’ offense hadn’t repeatedly hung the defense out to dry:

The Lions offense is still the 4th-best in the NFL, racking up 29.7 points per game. The passing attack is slightly slowed after last week's performance, but still high-flying at 6.99 YpA. The ground game is chugging along at 3.92 YpC.

The Falcons are allowing a ridiculous 7.97 YpA, surely a big part of why they're allowing a 22nd-best 24.5 points per game. The run defense is much better, allowing 3.84 YpC.

The Falcons defense is particularly ill-equipped to stop the Lions pass attack. I project the Lions to score 34-37 points, passing for 9.50-10.0 YpA and rushing for 3.50-3.75 YpC. I have medium confidence in this projection.

The brutal reality: 16 points, 5.72 YpA, and—surprisingly—5.2 YpC.

This is the worst underperformance of expectations I’ve seen since Daunte Culpepper got the Lions completely shut out by the Packers (I’d projected 24-28 points). Let’s be clear, here: the Falcons allowed 8 YpA through six games and faced a team gaining 7 YpA and held them to less than 6 YpA. This does not compute.

This isn't Whack-a-Mole effect, either; if it were the Lions would have rolled up 200 yards on the ground and still scored 34-37 points. This has nothing to do with the Falcons defense and everything to do with the Lions offense playing like a shadow of itself.

Next up: numbers and film.

Read more...

Watchtower Review: Lions vs. 49ers

Reviewing the first “wrong” Watchtower of the year is never fun, especially when it’s also a loss. From the defensive side:

Defensively . . . you may want to sit down for this. The Detroit Lions have the fourth-best scoring defense in football. They’re allowing a miniscule 17.7 points per game, and a meager 5.91 YpA. Their running defense has been less than stout; opposing running backs have been rolling up the Lions for 4.78 YpC.

Without any systemic advantages either way, we must project the 49ers offense to to meet expectations against the Lions defense. I project the 49ers to score 21-24 points, passing for 6.50-7.00 YpA and rushing for 4.25-4.50 YpC. I have medium-to-low confidence in this projection.

Mitigating/Aggravating Factors

Of course, it’s pretty much all up in the air, schematically speaking. There are many different flavors of the “West Coast Offense,” and the Callahan/Gruden branch is one we don’t have a lot of prior history with. The Lions are definitely soft against the run, and the 49ers are calling runs 54.2% of the time. The Lions showed against Chicago that merely grinding it out against them won’t work if the offense can score. For the Lions’ defense to exceed expectations, the offense may need to force Alex Smith to play catchup.

The scoring projection was spot on: the 49ers scored 23 offensive points. The run/pass effectiveness was all over the map, though: they passed for a miserable 3.91 YpA, and ran for a stonking 7.00 YpC. I hinted at the ultimate problem, though, in the Mitigating/Aggravating Factors: “The Lions showed against Chicago that merely grinding it out against them won’t work if the offense can score. For the Lions’ defense to exceed expectations, the offense may need to force Alex Smith to play catchup.”

Yup, that was the problem. The Lions defense precisely met expectations, but needed to exceed them to win the game. The only way they were going to exceed them was if the 49ers had to do a lot of passing . . . meaning, the Lions offense needed to force the 49ers offense to keep pace.

Given these unit’s equal strength (they’re both outstanding), if there really is no advantage for Scott Linehan’s offenses against Vic Fangio’s defenses, the Lions should exactly meet expectations. Since I’ve been trying to weigh recent performance more heavily than historical, I’ll roll with that as my official projection. Factoring in the 49ers penchant for allowing more yardage than points, I project the Lions to score 23-26 points, throw for 7.5-8.0 YpA, and run for 3.75-4.0 YpC. I have medium confidence in this projection.

Mitigating/Aggravating Factors

Of course, if there really is a systemic advantage for Lions offenses against Capers/Fangio 3-4 defenses, then the Lions should add another touchdown or so. As with Monday Night Football, I expect the crowd to give the Lions a lift—though not to the same extreme. However, if the Lions turn the ball over two or more times, they may not make it into the mid 20s.

Well, they didn’t make it into the 20s, but it wasn’t due to lack of turnovers. It was due to a total lack of passing effectiveness. The Lions averaged 5.86 YpA, a ridiculous 2.07 fewer yards per attempt than what they’d averaged to day. The Lions rushed for 3.67 YpC, which is not fantastic—but it’s only slightly below the Lions season average coming into the game (3.96), and above the 49ers season average allowed (3.57).

Yesterday’s Matthew Stafford piece got into the whys and wherefores behind the failure of the passing game, and I don’t want to repeat myself. Long story short, the 49ers pass defense was too good for the Lions dominate with pure talent. The 49ers defense was too good for Matthew Stafford and Calvin Johnson to just pitch and catch over. The Lions decided to put the pedal to the metal rather than downshift into a dink-and-dunk gear, and it didn’t work.

Conclusion

This game gives me the heebie-jeebies. Both teams are coming off enormous wins. Both teams are “due” to drop one against an extremely tough opponent. The data shows the Lions to hold about a field goal advantage in this one. Both teams take care of the ball extremely well, and both teams force a lot of turnovers. I feel like this will be the difference in what could be a very dramatic, very physical, low-scoring slugfest. The most likely outcome of this game is an extremely hard-fought 23-21 Lions win.

Heebie jeebies, yes. Lions win, no.

Read more...

Watchtower Review: Bears vs. Lions

Reviewing the Watchtower after wins is always fun; reviewing The Watchtower after THIS win is going to be especially delicious. Let’s look at what we came up with for the Bears offense:

I project the Bears to score 20-23 points, passing for 5.75–6.25 YpA, and rushing for 4.85–5.15 YpC. I have high confidence in this projection.

If there’s anything we learned from the dirty cheating Vikings, it’s that crowd noise can give a huge advantage to a pass rush. The Bears have struggled to protect Cutler this year, allowing 15 sacks—more than any other team except the Rams. As I wrote for Bleacher Report, the Lions have played much, much better at home than on the road in 2011, and if that trend continues the Bears won’t score anywhere near this many. Unless Forte rolls for 200 yards again, I don’t see a situation where the Bears score significantly more than 23 points on the Lions.

What the Bears actually did was score 13 points, pass for 6.55 YpA, and run for 4.88 YpC. Jay Cutler—who, as I said in Three Cups Deep, was phenomenal—was a little be better at moving the ball than I’d projected, but the Lions held Forte’s effectiveness to the bottom end of my expectations.

The crowd noise and pass rush did the trick, preventing the Bears from converting 371 yards of offense into any more than 13 points. The “Mitigating/Aggravating Factors” paragraph explained that I saw 23 as a hard cap, but if everything broke the Lions’ way, 20 was far above the floor. Commenter Imperical Evidence was bold enough to predict the Bears wouldn’t top 14, and I doff my cap to him.

Besides the ridiculous amount of penalties on both sides, which slowed the pace of the game to a crawl, the Bears managed to string together a lot of long drives. The time of possession favored Chicago by a nearly two-to-one margin. Forte and Cutler kept magically converting third downs—but the defensive backs kept all the Bears well in front of them. Fleeing the relentless front four, into the smothering horde of the back seven, Chicago’s offense was trapped. Slate’s Tom Scocca said it best:

Cutler had crossed over into the same waking nightmare state that Donovan McNabb was in during Super Bowl XXXIX, as the Eagles offense dawdled away any hopes of a comeback. Cutler's intentions no longer had anything to do with winning, or even scoring—his world had shrunk down to the space between finding the next dumpoff receiver and getting clobbered by the Lions again. Checkdown, crawl to his feet, checkdown once more. That was the only thing he could do, so he did it.

The Lions’ offense fared a little better against Chicago’s defense.

If we leave schematic interplay out of it, we’d expect the Lions to score quite a bit more than their season average on the year, throw a little better than usual, and have a nice game running the ball. However, if we apply what might be the most tried-and-true schematic effect I’ve ever identified, the scoring should be less high because of turnovers, but the passing and running games should have banner days.

Therefore, I project the Lions to score 33-36 points, while passing for 8.50—9.00 YpA, and rushing for 3.50—4.00 YpC. I have high confidence in this projection.

The Lions only scored 24 points, though they did pass for 8.43 YpA. Of course, Jahvid Best’s ridiculous night ballooned the rushing average to 9.05 YpC. On a night the Lions were moving the ball quite well, they beat themselves when trying to convert those yards into points.

Matthew Stafford killed the Lions chances to reclaim the lead before halftime with an interception that was entirely his fault. In the fourth quarter, a Willie Young sack ultimately forced the Bears to punt from their own 4. Stefan Logan got the ball to the Bears’ 44, but Keiland Williams got called for a block in the back on the return. On the ensuing 3rd-and-4, a Dominic Raiola penalty negated a brilliant Jahvid Best 14-yard reception, and the Lions had to punt away a great scoring opportunity.

The Lions also took their foot off the gas late in the fourth. Throughout the game, the Bears had shown a “cover one” look with safety Brandon Meriweather playing deep centerfield, with press man coverage on every Lions target. It looked like this:

Meriweather is off-frame, playing that deep deep centerfield. The Bears showed this look several times throughout the game, and the Lions consistently fed it to Megatron. It’s true: if Johnson is single covered, it’s automatically going to him. In fact, Stafford took his only sack of the game on this play waiting for Johnson to get open. He stared Johnson down, reared back to throw, then felt heat and pulled it back down. Israel Idonije had pushed Cherilus back with a bullrush, and Stafford didn’t feel Johnson was open enough to let fly.

Once inside four minutes, though, the Lions went from hitting Johnson every time the Bears showed this look to running every time the Bears showed this look. They were indisputably trying to put it in the cooler. Now, they chewed two full minutes off the clock and made it a two-possession game at the end, but the opportunity for more points was there.

On the whole, the defense did an amazing job of containing the Bears offense on a day when Cutler did no wrong. The offense missed some opportunities, but made the plays the opponent couldn’t make. In the end, it was clear who the better team was—and as the stats projected, the better team won.

Next week might be a different story.

Read more...

Watchtower Review: Lions at Cowboys

Last week I correctly predicted a lot of things about the NFL. I correctly picked the winner of 14 of last week’s 16 NFL games, including the exact score of the SF – Philly game. In the Watchtower, I also correctly predicted the winner . . . but that’s about all I got right.

The Cowboys have no systemic advantage over the Lions, so I project them to meet expectations, scoring 17-20 points. If last year’s pattern holds, they will throw for 6.75-7.50 YpA, well below season averages, and run for 3.75-4.0 YpC, well above season averages. I have medium confidence in this projection.

Of course, the Cowboys scored 30 points. Here’s where it gets interesting, though: They threw for 7.04 YpA and ran for 4.19 YpC—almost exactly in line with my expectations. The Cowboys moved the ball against the Lions defense just as well as I thought, but scored way more points. Why?

Willie Young got one sack on the last drive--that's it. The Lions committed 4 defensive penalties, 3 by Corey Williams and 1 by Ndamukong Suh. One gave the Cowboys a free first down on 2nd-and-five, and the Dez Bryant TD came three downs later. One made a 3rd-and-10 into a 3rd-and-5, which was converted—ultimately leading to a Cowboys field goal. The third and fourth were on the Cowboys’ final scoring drive, allowing the field goal that made it 30-17. That last one was Suh’s disputed roughing-the-passer penalty on a failed 3rd-and-3 that would have stopped that drive.

As I’ve said before, the Lions are built to jump to an early lead, then turn up the defensive heat as teams play catch-up against tem. The offense hung the defense out to dry time after time after time—and only because the back seven took the game into their own hands was the victory delivered.

The Lions shouldn’t be winning games because the defense is dragging the offense back into the game kicking and screaming; the offense should be dominating from the jump. This has to get better.

We’re left to conclude that Scott Linehan’s balanced offense has a systemic passing advantage against Rob Ryan’s hyperaggressive 3-4 defense. Linehan’s offenses pass much better than expectations, and score much better than expectations, when facing a Rob Ryan defense. Therefore, I project the Lions offense to outpeform expectations, especially through the air, scoring 33-35 points. I have medium-to-high confidence in this projection.

Whoops. The Lions’ offense scored just 20 points. I apparently forgot to project per-play running and passing effectiveness, but believe me when I say 5.58 YpA is way below expectations. The Lions had been averaging 8.45 YpA coming into the game, and I was projecting a systemic pass advantage. The Cowboys had been allowing 6.48 YpA coming into the game, which is a little on the stingy, but the Lions’ passing attack should have been much more effective than it was, even without an advantage.

The run game bounced back after last week’s awful performance, averaging a tepid-but-acceptable 3.71 YpC. The Cowboys had been excellent against the run, allowing just 2.97 YpC, so that’s actually an excellent performance. Between the offensive line holding the #1 pass rush in the NFL to no sacks, and the running game producing as-usual against a stout run D, the offensive line really redeemed themselves against Dallas.

I believe this game is critical to the outcome of the Lions’ season. The Lions will be a very tough out at home, but have a vicious last three games of their schedule. To get a road win on this big of a stage would be huge, and it makes me gunshy that the data leads my proverbial horse to water. Yet, what have I to do but drink? The most likely outcome of the game is a 34-20 Lions win.

Got the score wrong, but got the conclusion right: no matter how you slice it, this Lions win has enormous implications for the rest of the season.

Read more...

Watchtower Review: Lions at Vikings

From this week’s slightly-truncated Watchtower:

I can't find any game where Bill Musgrave faced off against a Schwartzingham defense as a playcaller, at least not while running anything like this offense.

With no schematic advantages either way, we must project the Vikings’ offense to meet expectations. Though I don’t typically use data from the current season until after the third game, at this point I’m left with no choice. The Vikings are currently scoring 18.5 points per game, after averaging 17.6 points per game in 2010. The Lions are the second-best scoring defense in the league, allowing just 11.5 points per game. I don’t expect the Lions to finish the season allowing so few points per game, but I do expect the Vikings to score fewer than their season average today.

Therefore, I project the Vikings to score between 10 and 13 points. I have extremely low confidence in this projection.

Obviously, this isn’t what happened.

The Lions are scoring 37.5 points per game, but I don’t expect that to last all season. I’ll project the Lions to outperform what I think will be their season averages on the year, scoring 34-37 points. I have extremely low confidence in this projection.

Obviously, this isn’t what happened either.

The objective of the Watchtower is to tell the story of the game from what we see in the data; to look at the matchup of the two teams and see how their offenses and defenses will play against each other. The story the data told me was this:

I project a 34-13 Lions victory. My gut feeling on the relative strengths of these two teams agrees with that projection.

The key phrase in there is “the relative strengths of these two teams.” If the Lions and Vikings played ten times on a neutral field with no one watching, I suspect the average outcome would be something like 34-13. I truly believe the Lions are one of the best teams in the NFL this season, and the Vikings are going to struggle to something like a 5-win season.

However, the two teams weren’t playing on a neutral field, which I considered a mitigating factor:

One thing I'd keep an eye on: the Vikings have started hot in each of their first two games, only to fall apart. They’re pretty much playing for their season, at home, in a place where the Lions almost never win. If they jump out to an early lead, momentum may carry them beyond expectations . . .

I feel like this Vikings team is backed against the wall at 0-2, and—though they’re saying all the right things—I have to believe the Lions are believing their own press at this point, too. I worry about the Vikes getting the jump on the Lions and popping the hype balloon.

I avoided a Michigan State-Notre Dame comparison because I don’t want to associate my Lions stuff with various local rivalries. There’s a reason I keep my college stuff over at A Beautiful Day For Football. But the comparisons were just too obvious. A 2-0 team feeling its oats after a complete and total blowout win, travelling to a notorious stadium to play a rival fighting for its life at 0-2.

The Vikings moved the ball just as well as they had through the first two weeks, gaining 5.86 YpA (5.93 avg. through Week 2), and 5.74 YpC (5.85). Since the Lions have a better-than-average defense (through Week 3, 3rd-best defense in football), this means the Vikings performed better offensively against the Lions than they did in the first two weeks.

Defensively, the Vikings stopped the Lions’ run game cold, allowing 1.05 YpC (3.31 avg. through Week 2). The passing offense moved the ball at an 8.22 YpA clip, down slightly (8.59 avg.)—but of course, Stafford was sacked five times for –40 yards, and the Lions were assessed 7 penalties for –65 yards. This was how the Vikings held the Lions’ 33.7 PpG offense (through Week 3, ranked 4th) to 26 points.

Really though, the story of the game was the one I worried about: the Vikings playing their brains out, the crowd noise and the pass rush handing the Vikings the momentum, and the Lions finding themselves having to scrap their way out of a hole, as opposed to putting it in the cooler.

This is why I was so grumpy about the probably-enhanced crowd noise: it made an enormous difference in the way the game played out. The lesson to take away from this game, and this Watchtower, is that the Lions are clearly much better than the Vikings. The Lions took Minnesota’s best shot—and staggered—but leaned on the ropes long enough for their talent and skill to win the day.

Read more...

Watchtower Review: Lions vs. Chiefs

From the Lions vs. Chiefs Watchtower:

Last season, the Lions had the 19th-best defense in the NFL, allowing 23.1 points per game. The Chiefs scored 22.9 PpG, 14th-best in the league. I’m going to project the Chiefs to score 23-27 points. I have hellaciously low confidence in this projection.

Heh.

When a Romeo Crennel defense has average-or-better skill level, it disproportionately disrupts a Linehan offense’s scoring—primarily, by depressing per-play pass effectiveness. Run effectiveness seems to go up in response, possibly due to Crennel ‘ceding’ the run in order to stop the pass. When Crennel’s available talent is lesser, though, the situation reverses itself: Crennel’s defense becomes extremely susceptible to the pass, and therefore allows points in bunches.

Given that the Chiefs have lost their best pass defender, Eric Berry, to an ACL tear, and given the Lions’ explosive passing offense, I believe Detroit able to at least compensate for this schematic advantage—and possibly, overcome it and flip the tables on Crennel.

I project the Lions to score 31-34 points. I have extremely low confidence in this projection.

Heh.

Knowing this is more of a shot in the dark than an actual projection: I project a 34-23 Lions win.

MWAH HA HA HA HA HA.

When looking at the history and resumé of Chiefs’ OC Bill Muir, I noticed eerie parallels between him and former Lions OC Jim Colletto. Both Muir and Colletto were previous colleagues of the OCs they replaced, Charlie Weis and Mike Martz. Both Weis and Martz are known as pass-first schematic innovators who occasionally overestimate their own superiority. Both Muir and Colletto were charged with maintaining the existing framework of the offense, while streamlining out the “wizardry” and reinforcing the MANBALL.

So far, Muir’s tenure has been every bit the disaster Colletto’s was.

Early on, the Chiefs used tosses, sweeps, and reverses to exploit the Lions’ defensive line. As I’ve said on here before, and as Pro Football Focus has repeatedly identified, the Lions’ linemen often abandon gap responsibility in the name of penetration. Notably, I saw Kyle Vanden Bosch drawn well out of position on the reverses—leaving nobody home to make the tackle. This was a clever bit of scheming by the Chiefs, and one I expect to see deployed by other teams this year.

After halftime, the Lions did a much better job of staying home—and of course, the Lions’ offense made stopping the run a moot point anyway. I believe this is a design principle of the defense. As someone on the Fireside Chat noted, Schwartz has said “We’ll stop the run on the way to the quarterback,” and that’s exactly what we saw in effect: the Lions did not stop the run well, but the Lions’ offense made stopping the run irrelevant.

Instead, the Lions focused on pass rush and pass coverage, and the result was incredible. Two sacks, one QB hit, nine pressures, a batted pass, four forced fumbles (three recovered), and three interceptions. As I said in my latest Bleacher Report article, 5 Ways NFL Experts Were Wrong About the Detroit Lions, the Lions’ back seven currently rank number one overall in Pro Football Focus’s pass coverage grades.

On the offensive side, I’d found Crennel’s defense did a disproportionately good job against Linehan’s offense in years when Crennel’s defense was ranked in the top two thirds of the league. Crennel’s 17th-ranked Patriots held Linehan’s 8th-ranked Vikings to just 17 points—and Crennel’s 11th-ranked Browns completely shut out Linehan’s 16th-ranked Dolphins. In both games, passing effectiveness was static, or down, while running effectiveness greatly increased. It seemed like Crennel was allowing the run in order to stop the pass.

However, the year Crennel’s D was ranked 23rd and Linehan's Rams were the 28th-best (read: 3rd-worst) offense in the NFL, the Rams went wild. Those timid Rams hung 27 points on the Browns, 65% above their season average. They averaged 8.36 YpA, 48% better than their season average—though their run game managed only 3.29 YpC (-13%). I concluded that when Crennel’s defensive talent is well below average, Linehan’s offenses explode through the air (despite running less effectively).

Well.

The Lions mustered only 2.97 YpC, but blew up for 8.05 YpA—and, of course, scored 48 offensive points. It must be said: the offense was assisted by a short field several times, as the defense forced an astounding six turnovers—but the Lions were excellent in the red zone, converting many of the chances they got. It looks as though my analysis of what happens when Linehan meets Crennel was spot on.

In conclusion, this game was super-awesome, and the Lions haven’t even played their best football yet. It’s all part of finding out exactly how high expectations for the Lions should be—and in this case, how low expectations for the Chiefs should be.

Read more...

Okay so let's talk about this Watchtower thing

Recently, Commenter Matt—longtime friend of this blog—challenged several of the principles this Watchtower thing is built on. Many of you weren't around for the Watchtower's genesis, and I've been doing these things for three years now. Now's a good time to air this stuff out, explain my methodology, and make sure I haven't made a wrong turn somewhere in my execution.

The original mission of the Watchtower was to find out what the new-look Lions were going to look like. To, like the forest ranger in the old Lassie episodes, use a combination of familiar landmarks and mathematics to pinpoint a direction and distance to the truth.

With so much of the new-look Lions roster being turned over from the year before, we didn't know what to expect when they took the field that first week. Since Linehan and Cunningham have long track records with distinct schematic styles--and, because I've always thought the schematic interplay between offense and defense is overlooked--I  decided to analyze historical matchups between the Lions' coordinators and their first opponents'.

I found their season averages of points scored and points allowed, saw how each team performed relative to their averages when facing the other coordinator, and concluded the following:

As all three metrics of output--per-play passing, per-play rushing, and points scored—are way above their averages for the season, I’m going to say that given superior talent and execution, and/or excellent pass protection, Scott Linehan’s balanced offense disproportionately gives Gregg Williams’ attacking 3-4 defense fits.

Given equal talent, Cunningham’s hyperagressive 4-3 is extremely effective against Payton’s pass-heavy offense, but only if that aggression leads to mistakes and turnovers—otherwise, the holes in the defense will be exploited. Effective quarterback play may neutralize the defensive advantage.

Therefore, the most probable outcome of this game is a shootout that the Lions lose.

I didn't project a score, and didn't project passing or rushing effectiveness. I just tried to retell the story of the upcoming game, as clearly as I could read it in the data. Over time, I've added more data, made more projections, and reviewed the articles and methodology for their accuracy and completeness. I emphasized method's qualities as predictive tool because it showed great promise as such--and, I came to discover, that's what people want: a prediction.

However, it is still a study of offense vs. defense, and analysis based purely on per-play effectiveness and scoring output has limitations. Sacks, turnovers, penalties, injuries, special teams play, weather conditions, crowd noise, and clock management all dramatically affect the relationship between "how good is the team doing" and "how many points the team has scored or allowed". Look no further than the Lions’ 2009 trip to Chicago for proof of that: the Lions outgained the Bears 440-292, but lost 24-48.

When reviewing historical games, I always check for these factors to see if they're a more likely cause for unexpected outcomes than "Ol' Gun's just got Coach So-and-so's number," and I not only note that but build it into the prediction. Frequently, Cunningham's defenses have held offenses to lower totals than you'd expect based on their relative talent levels, frequently because Gun's Ds are designed to get sacks and force mistakes.

Matt said:

I totally disagree that fumbles and fumble recoveries are "random." Unpredictable, sure, but not random.

Here's some reading material on that:

Football Outsiders Basics: "Fumble recovery is a major reason why the general public overestimates or underestimates certain teams. Fumbles are huge, turning-point plays that dramatically impact wins and losses in the past, while fumble recovery percentage says absolutely nothing about a team's chances of winning games in the future. With this in mind, Football Outsiders stats treat all fumbles as equal, penalizing them based on the likelihood of each type of fumble (run, pass, sack, etc.) being recovered by the defense. Other plays that qualify as 'non-predictive events' include blocked kicks and touchdowns during turnover returns. These plays are not 'lucky,' per se, but they have no value whatsoever for predicting future performance."

Advanced NFL Stats: “The ratio is not significant for wins, points scored, or points allowed while overall fumbles and fumble rates are significant. This result indicates recoveries are indeed random. It also suggests that fumbles, and not fumbles lost is a better stat for estimating a team's future likelihood of fumbing.”

CougarBoard.com: “Look at LSU. They were 70th and 102nd in 2005 and 2006, then suddenly finished 1st in back-to-back years in 2007 and 2008. Oklahoma went from 118th to 1st in the space of two years. BYU went from 112th to 2nd in two years and two years later were back at 95th. There is no pattern here. If you take a histogram of average fumble lost rankings by team, you get a tight, centralized cluster which is the hallmarks of a random distribution.”

As the New York Times explained, Jim Schwartz once told Bill Belichick the same thing: "“Fumbles are a random occurrence,” Schwartz said he told Belichick. “Being able to get interceptions or not throw interceptions has a high correlation with good teams. But over the course of a year, good teams don’t fumble any more or less than bad teams. Bill didn’t agree. He said, ‘No, good teams don’t fumble the ball.’ But actually, they fumble just as often as bad teams.”'

Long story short, Matt is right that fumbles themselves aren't completely random. Daunte Culpepper fumbled the ball 80 times in 73 games as a Minnesota Viking; that's definitely his fault and not pure misfortune. But in general, using a team's current, or recent, fumble rate and projecting it forward to the next game simply won't be accurate.

When it comes to recoveries, there's no doubt--once that ball hits the turf, it's statistically random. Just look at the ridiculous tip drills that occurred last Sunday. There were several near-heroic individual efforts on the part of the Buccaneers to force that fumble, keep it in play, and come up with it; that it ended up where Rob Sims could make a play on it was pure chance. The Bucs could have put their hands in their pockets and gotten the same result!

The same thing goes for interception and fumble returns; whether they can be returned for scores depends more more on the spot on the field they occur than on team defensive ability. In my opinion, Matt is right that the Will Heller tip-pick was caused by a combination of Stafford throwing it too high and too outside, Heller batting it up instead of down, and Talib having excellent awareness and ball skills.

However, Talib picking that pass off with nothing but green turf and white stripes between him and the end zone was purely a matter of location. Chris Houston's interception over Arrelious Benn's head was no less outstanding a play, but he was in no position to return it. If the same players had made each other's plays, the result would have been the same.

The upshot of this: even though fumbles, fumble recoveries, special teams play, return touchdowns, and the like are either completely random or completely unpredictable, Matt’s right that without accurately accounting for them I’ll never be able to accurately predict the scores of NFL games. In fact, Advanced NFL Stats concluded that 42% of all NFL wins-and-loss records are determined by random variance, not relative performance. There’s a very hard cap on how close my projections can get . . . that’s why I remove, or adjust for, as much of these non-predictive factors as possible and then see how close I got.

Maybe I’ve taken my eye off the ball a bit. Maybe I’ve been too focused on the score projections, and the accuracy thereof. I really only do it because, as Brian Cook from MGoBlog says, “the strictures and conventions of sportswriting compel me.” What do you folks think? Is The Watchtower a worthwhile read even if I don’t take a stab at the score? Should I be reviewing every Watchtower I write?

Read more...

Watchtower Review: Lions at Buccaneers

From the season’s first Watchtower:

While highlighting the extreme paucity of data, I’m going to swallow hard, wipe the sweat from my forehead, tug at my collar and project the most likely outcome to be another narrow Lions victory: 24-21, albeit in regulation this time.

Now, the final score was 27-20, but the Watchtower only accounts for offense-defense interaction. For the purposes of Watchtower evaluation the “real” score was 27-13. Before you folks cry foul, a quick reminder: the goal of The Watchtower is to spot systemic advantages that the Lions’ offense and defense have over their opponents’ schemes. Counting special teams and return scores throw off our evaluation of those effects.

Detroit Lions defense vs. Tampa Bay Buccaneers offense:

I can’t identify a systemic advantage from one data point. Disproportionately disrupting scoring by stopping drives with sacks is the design goal of the defense. I’m kind of stunned it worked with Blount being so devastating; I expect the Bucs to give Blount more than 15 carries this Sunday. We have no current performance data to go on, so I’ll have to recycle last year’s. I project the Bucs’ offense will meet expectations, scoring 20-23 points. I have extremely low confidence in this projection.

Right off the bat, we see the enormous difference: instead of LeGarrette Blount destroying the Lions at a 7.33 YpC clip, he was held to just 15 yards on 5 carries; a 3.0 YpC. As a team, the Bucs didn’t fare much better: 16 carries for 56 yards, 3.50 YpC.

The Lions’ pass rush didn’t hit home quite as often as last year’s contest, or for as many negative yards (two for -7 vs. three for -25). But, there was just enough pressure to rattle Freeman a bit.

Freeman’s completion percentage was nearly identical, 65.1% to last year’s 65.6%, but his yards-per-attempt was down dramatically. Freeman netted only 5.78 YpA this year—well off last year’s average of 6.80, and well well off the 7.64 he managed against the Lions last season. Between the there-enough-to-be-felt pressure and the excellent downfield coverage, the Bucs’ passing offense had to settle for underneath stuff—again, depressing scoring.

The Lions also intercepted Freeman once, and forced four fumbles, recovering one. Last season, the Bucs didn’t give away the ball at all, so two turnovers this time doubtlessly depressed scoring. We don’t have season averages to work with, but I ascribe this better-than-expectations performance to the Lions’ defense skill level being vastly improved. Over the course of 2011, I expect the Buccaneers to ouutperform their 2010 scoring average—meaning this Lions defense performed like an above-average unit on Sunday.

Detroit Lions offense vs. Tampa Bay Buccaneers defense:

The story is basically the same in reverse: the Lions more-or-less met expectations last time. Without any established trends for this year, I can only project the Lions’ offense to meet expectations, scoring 21-24 points. I have extremely low confidence in this projection.

Of course, the Lions offense did outperform my “wildazzed guess,” as one MLive wag (downbeat) called it. As with the Bucs’, the Lions rushing attack wasn’t nearly as effective this time around. 35 carries for 126 yards netted only 3.6 YpC—nearly identical potency. But twice as many carries is a strong indicator that the Lions were A) moving the chains and B) winning.

There’s an old saying amongst young statisticians: “You’re not winning because you’re running, you’re running because you’re winning.” The Lions weren’t pounding the rock any better than the Bucs were. They were just effective enough to control the ball and deny the Bucs possession. Tampa Bay couldn’t do the same thing because they were playing catch-up. Once they abandoned the run, the Lions could focus on pass rush and coverage—which, I’ll say it again, was excellent.

Through the air, Matthew Stafford’s execution was visibly worse than it was in the preseason—which, since that was quite nearly perfect, was to be expected. He completed 24-of-33 passes—that’s an astounding 72.7%. He wasn’t just dinking and dunking either: he netted 9.24 YpA; equally astounding. He threw three touchdown passes, not including a perfectly-placed TD pass Brandon Pettigrew dropped, and a wide-open Calvin Johnson He threw one interception, counting one tipped catch that came down in Aqib Talib’s hands, and not counting another one Jahvid Best basically shovel-passed to a stone-handed defender.

Even though the story of the game was correctly told as, “Stafford had the butterflies early and then settled down,” Stafford’s “jittery” performance was still outstanding by any measure. The ongoing performance of the Buccaneers’ defense will tell us a lot about how good the offense was this day. If the Bucs’ D turns out to be as good as many expect, this was an excellent statistical showing by an offense likely to average 30+ points per game the rest of the way out. If the Bucs’ defense is more hype than substance, then this was a pretty okay showing—but not near what we think the offense will be capable of.

Conclusion

I projected a close Lions win in regulation, 24-21. I can’t call it “right” or “wrong” because there was neither a long track record of coordinator matchups, nor any in-season data with which to establish expectations. This 27-13 performance showed us a lot of good things, though: the defense stopped the run and covered downfield better than we expected, and the running game was just good enough to control the game after Stafford built a lead.

There were a lot of mistakes, a lot of missed opportunities, and a lot of room to improve. But overall, this Lions team won on the road against a notional playoff contender—and looked better doing it than we could have hoped for.

Read more...

Watchtower Review: Lions vs. Vikings

The data said this Lions-Vikings game would be higher-scoring than it was:

I'll be honest: my gut is calling this one a Lions romp, something like 35-10. However, the data’s telling me to be far more cautious, so I will be—to a point. According to my projections, the most likely outcome of the game is a 23-17 Lions victory. And, if I’m right, the worst decade in Lions history—arguably, in NFL history—will truly be a thing of the past.

By The Watchtower's offensive-scoring-only reckoning, this game's score was 20-6, so I was in the ballpark on the Lions’ score, but way, way off on the Vikings’.  Most of this was due to the phenomenal job the Lions did of bottling up Adrian Peterson: 14 carries needed him just 31 yards—that’s a vanishingly low 2.12 YpC average.  Contrast this to Week 3, where All Day picked up 160 yards on 23 carries.  It’s true that 80 yards came on one play, but even 22-for-80 is slightly subpar rate of 3.63 YpC.  There’s no other way to put it: the Lions defense did a masterful job of clamping down on Adrian Peterson.

They dared the Almighty Joe Webb to beat them, and—even though he’s like just as fast as Mike Vick or whatever—Webb was completely ineffective.  In 35 dropbacks, Webb was sacked three times, threw a pick, and completed only 20 passes; his average YpA was a wretched 4.53.  I loved the way the Lions switched schemes in the second half: they moved to a 3-3-5 nickel, often dropping eight, getting pressure anyway, and leaving Webb with no options.

Offensively, the Lions fell short of expectations—but a goal-line fumble from Tony Scheffler cost the Lions a touchdown they’d driven 75 yards to score.  Of course, him breaking the plane with possession would have resulted in my projection being short by four instead of long by three—but the point is, the numbers were fairly accurate here.  Other than the admittedly brutal pick-six, Shaun Hill displayed his burgeoning mastery of the offense: he went 28-of-39 (71.8%) for 258 yards (6.62 YpA), while distributing to eight different Lions—none of whom was Calvin Johnson.

On a meta note, I’m going to be doing some heavy modifying of the Watchtower formula over the course of the summer.  After a season-long Watchtower Review that looks at the process entire for this season, I’ll move on to exploring new stats, new approaches, and trying to incorporate special teams/turnover scores in one way or another . . .

Technorati Tags: nfl,detroit lions,minnesota vikings

Read more...

Watchtower Review: Lions vs. Packers

Let’s start with all the totally wrong things I said in last week’s Watchtower:

It seems unlikely that the Lions’ defense does that well against an offense that good twice in a row, though, so I’m going to project the Packers to slightly underperform scoring expectations: gaining 23-26 points. I expect the Lions to be more successful defending the pass with zone coverage, allowing 8.0-8.5 YpA, at the expense of fewer interceptions. The Lions will likely cede the run to the Packers again, allowing 4.25-4.5 YpC. I have high confidence in this projection.

3 points, 3.30 YpC, 6.02 YpA.  Tremendous performance by the defense.

As I said, in Week 4, the Lions managed to hold the Packers’ offense to as few points as can be expected, given how effective they were on a per-play basis. If the same offense shows up this week, I don’t see the Lions being so fortunate.

It can be argued that the same offense didn't show up—but the defense far outstripped their Week 4 effort, even with several key starters lost.  I’m still stunned at how well this defense performed on Sunday.

I don’t expect the Lions to put up 454 yards against the Packers this time around, but I do project them to be more productive with their yards: scoring 17-23 points, gaining 4.50-5.00 YpC, and throwing for 5.75-6.00 YpA.

Yeah not so much.  The yards projections were fairly close—4.63  YpC and 5.32 YpA—but they spun that effectiveness into a single measly touchdown.  Two interceptions and a missed field goal didn’t help.

The Lions gave the Packers their best shot in Week 4, and lost the game mostly by shooting themselves in the foot. Between a truly massive amount of penalties (13 for 102 yards), and the now-obligatory Charles Woodson pick-six, the Lions let the Packers off the hook. As I said above, it’s hard to believe that the Lions could play that well against a team as talented as Green Bay twice in once season—and yet, it’s there. It’s there for them, again. They lost by two on the road before, they can win this one at home. They can . . . but they won’t.

They could . . . and they did.

CTRL-C, CTRL-V, folks. This will be an extremely painful 17-24 defeat.

Or, an incredibly improbable 7-3 victory!  I haven’t ever been more wildly wrong in a score projection, and I couldn’t be happier about it.  I will take that over, and over, and over, and over.  Okay, I’ve got an idea: let’s talk about everything I said that was right.

Look for the Lions to drop eight men into coverage on a regular basis, essentially the same game plan we saw against the Patriots on Thanksgiving. The idea is to get enough rush from the front four to slow down the Pack’s offense just long enough for the Lions’ offense to outscore them.

The Lions did blitz the Packers more than they blitzed the Pats—but they certainly got most of their production from the front four.  It also took a looooooooooooooong time, but they slowed down that Pack offense just long enough for the Lions’ offense to outscore them.  Barely.

Greg Jennings will be the key to this game for the Packers’ offense.  If the Lions can stop him with double- and triple-teams, I don’t think Donald Driver and Donald Lee will be enough to score more than the 23-26 I project, if enough to score that much at all.

Greg Jennings: 4 catches, 52 yards, one bobbled pass that was ultimately intercepted.  They didn’t score the 23-26 I projected, either.

Last year, I thought I’d identified a systemic advantage for Scott Linehan against Dom Capers (and most 3-4 defenses), where the running game was disproportionately effective, and scoring was therefore disproportionately higher.

They didn’t score much, but the Lions ran very well against the Packers—the #1 defense in the NFL—and it was the difference in the game.

However, the Packers are allowing 4.49 YpC on the ground, 7th-worst in the NFL—so clearly, my notion that the Packers are generally struggling against the run was true.  Further, the Lions will be starting Drew Stanton, whose running ability is well-known, and whom the Lions have never hesitated to call designed running plays for.  I expect to see at least one rushing touchdown, or 20-plus-yard scramble, from Drew Stanton on Sunday.

Drew ran 4 times for 44 yards.  He didn’t score, and his long was only 17 yards (not 20).  However, on his other three carries he averaged 9 yards per; he was asked to make plays with his feet and he did—along with the six other Lions who toted the rock on Sunday.

First, this presumes that Drew Stanton plays like he did last week: an efficient, effective, conservative backup quarterback.  If Stanton has a regression to his “2009 49ers game” form, this will be an ugly blowout.  However, there’s an X factor here named Jahvid Best. We saw a return to form last week, and if he still has that burst, that bounce, he could do an awful lot of damage against a Green Bay defense that’s lost three of its top five linebackers for the year.

Unfortunately, Drew did regress to 2009 form against the Packers’ defense—at least for the first three quarters—and boy, was it ever ugly.  However, the running of Best, Maurice Morris, Stefan Logan, and Stanton himself bailed the offense out.

Finally, in the first matchup, Brandon Pettigrew made a lot more headlines for the two or three crucial passes he dropped, than the eight he caught for 91 yards. Likely being matched up against A.J. Hawk, as 49ers TE Vernon Davis was last week (4 catches, 126 yards, 1 TD) could result in a field day for Jumbotron.

Okay, this one was pretty much entirely wrong; Pettigrew was held to just 2 catches for 14 yards.  However, it’s worth noting that the game-winning touchdown came on a tight end screen to Will Heller—who was spelling Pettigrew.  Further, Heller ran right through A.J. Hawk on his way to paydirt.

Technorati Tags: nfl,detroit lions,green bay packers,drew stanton,greg jennings

Read more...

Watchtower Review: Lions vs. Bears

Okay, last thing first:

The one thing I’ve learned over the past two years of doing this preview, is that when the same teams play twice in a season, the results are rarely the same. But the data points to a repeat—and the injury problems for the Lions should be offset by the difference between a season-opening road game, and a midseason sellout. This is a statement game in many respects, and turnovers will likely make the difference. Last time, there were five fumbles (three lost), two picks, and 6 sacks for –42 yards. I see a similarly messy game this time around; how those turnovers and sacks are distributed will be the difference in the outcome. I am tempted to call this a draw, but at this point in the season, I’ll go out on a limb one more time—despite having a limb hacked out from underneath me three times already this season. I hesitantly project a 21-20 Lions victory, if for no other reason than the Lions need it much more than the Bears want it.

Right.  At this point, I’m not even going to slap myself on the wrist—the numbers said this game was a push, and that’s pretty much what played out.  The one thing I was most confident, about though, was wrong: the Lions played a nearly clean game and lost.  Stanton was sacked only twice, for a loss of 20 yards.  He threw no interceptions, and the Lions lost no fumbles.  Meanwhile, the Lions sacked the Bears 4 times for a loss of 37 yards, and one of those was a sack-fumble inside the Bears' 10 that the Lions recovered.  Clearly, the distribution of sacks and turnovers wasn't the difference in the game, as I said it would be.

Here’s where I tip my cap to Mike Martz.  The offensive gameplan we saw in Week 1 was Mike Martz football: four wide, spread it out, throw throw throw, lots of yards, lots of picks, a tailback doing more damage through the air than on the ground.  Here’s what I projected based on the numbers the Bears have put up all year:

It looks as though the only statistical trend for these two coaches, when facing off against one another, is that both units will play to their means: the 20.2-ppg Bears scored 19 against the 22.4-ppg Lions. Note, however, that that included about thirty minutes of shutout play in between Matt Forte receiving touchdowns. Given the data at hand, I’m inclined to project a repeat performance: 17-21 points, 7.50-8.00 YpA, and 4.0-4.25 YpC. I have medium confidence in this projection.

Here's what actually happened: the Bears scored 24 points, completed 21 of 26 for 9.00 YpA, and carried 28 times for 4.07 YpC.  That's not Martz football, that's Bears football.  The game plan went like this: run, screen, draw, slant, run, run, draw, screen, slant, run, draw, screen, slant, run, run.  They partially neutralized the Lions’ blitzing, and put Cutler in great situations that minimized his opportunity to make mistakes. 

Chester Taylor, a misfit toy who’s been paid a lot of money to not really do anything this season, had his coming-out party: 9 carries for 33 yards and a score, plus five catches for 31 yards.  Forte, though not as devastating with the long receptions this time, was equally effective: 13 carries for 64 yards and a score, plus two catches for 36 yards.  The Bears ruthlessly attacked the Lions’ outside linebackers, who were rendered helpless by Earl Bennett’s speed up the seam, and overmatched on screen by the Bears’ tight ends.  Again, Mike Martz has found out how to make the Bears’ talent go, and had a perfect gameplan to attack the Lions’ defense.

This isn’t to say that Gunther didn’t do his job, too.  Despite the steady diet of screens and draws, the Lions’ pass rush got to Cutler over and over again; both with well-timed blitzes and great front-four performances.  The Bears’ gameplan was to minimize Cutler’s exposure and they still brought him down four times; that is impressive indeed.  What we’re left with is what we knew all along: the Lions’ back seven cannot cover anybody

Everyone wants to know why the Lions can’t close out games; look at the opponent’s quarterbacks completing pass after pass after pass in the fourth quarter and you’ll see why. Eventually, the front four can no longer hold back the flood; the dam bursts.  This is purely due to lack of talent, and cannot be fixed until the offseason.

Offensively, here’s what I projected:

Given a mild yard-producing advantage for Scott Linehan balanced offenses against Lovie Smith aggressive Tampa 2 defenses, I project the Lions to roughly meet expectations: 17-21 points, 5.50-6.0 YpA, and 3.50-3.75 YpC. I have medium-to-high confidence in this projection.

What the Lions did was move the ball better than they have all year.

They rushed for 4.96 yards per carry.  They completed 16 of 24 passes, for 7.42 yards per attempt.  On a per-play basis, this was the most balanced, most efficient offensive game the Lions played all year—and they failed to commit a single turnover.  Let me tip my hat to Scott Linehan as well; he did a similarly masterful job in tailoring his offense to his quarterback’s strengths.  Unfortunately, it only resulted in 20 points.  Both teams were taking long, slow, controlled trips across the field—the Lions’ opening drive burned four minutes off the clock and only went 26 yards—so the Lions’ offense only had nine possessions.  Part of it is simply missed opportunities; when you get a sack-fumble inside your opponents’ ten-yard-line, you have to score.

What’s funny is, every seasoned Lions observer threw up their hands when the Lions had to settle for a field goal there.  The Lions’ radio play-by-play man, Dan Miller, fought to keep the frustration out of his voice as he talked about the importance of getting a touchdown in that situation.  Sure enough, it was the difference in the game.  This is why I don’t put a lot of stock in the “learning how to win” thing.  The Lions don’t need to go to “put the boot on the throat” school, it’s the sum total of made mistakes and unmade plays throughout the game putting them on the wrong side of the ledger at the end.

This team's talent and execution is right there.  I fully believe the Lions have made the jump into the Great Middle of the NFL; they can play and compete with anyone.  However, their youth, inexperience, and lack of depth push them to the edge of their margin for error.  If anything—anything—fluky goes wrong, they simply can’t make up for it.  This is why the keep coming close, but can’t hang on.  This is why they keep being right there, except for . . .  This is why they’re 2-10, and not 6-6.  

Technorati Tags: nfl,detroit lions,chicago bears,jim schwartz,mike martz,scott linehan,drew stanton,gunther cunningham,lovie smith,rod marinelli

Read more...

Watchtower REview: Lions at Cowboys

I cannot believe I’m about to do this after the past two weeks. I projected the Lions to beat the Jets, and they blew a 10-point 4th-quarter lead and lost in overtime. I projected the Lions to handle the Bills, and they completely failed to show up. If the Lions are the Lions we saw all year up until last Sunday, they’re a better team than the Cowboys and should win. If the Lions are the Lions we saw all decade, forget about it. Interestingly, though, I think the shoes from last week are on the other foot: the Cowboys are fresh off a fantastic performance, feeling their oats, thinking they're way better than the lowly Lions they're about to face--even though both teams have putrid records. After mailing it in last week, the Lions ought to be revved up to prove they are who we thought they were. All that aside, though, the numbers show these two teams to be very evenly matched, but with a definite offensive edge for the Lions.

Hesitatingly, gulpingly, and with an extreme chance of heartbreak, I declare that the most likely outcome of the game is a 27-24 Lions win. Heaven help me.

Heaven help DENIED.

Given no systemic advantage or disadvantage, Jason Garrett’s implementation of the Air Coryell offense should meet expectations against Gunther Cunningham’s aggressive 4-3, scoring 21-24 points, averaging 8.0-to-9.0 YpA, and 3.5 to 3.75 YpC. I have low confidence in this projection.

For the record, the Cowboys scored 28 offensive points—despite being held to just 6.13 YpA through the air.  The Cowboys did rumble for 4.67 YpC, though.

Expectations would hold that the Lions score better than their season average against the Cowboys, and that the Cowboys allow slightly more than usual against the Lions. There, theoretically, is a mild scoring advantage for the Lions, but the defense did play markedly better in its first game under Pasqualoni—so I’ll call those two factors a wash. I project the Lions’ offense to match expectations against the Cowboys, scoring 25-to-30 points, averaging 6-to-7 YpA, and 3.75-to-4 YpC. I have medium-to-high confidence in this projection.

Well . . . the Lions netted 6.15 YpA, and exactly 3.75 YpC—so, I got that bit right.  The offense was working as we’d expect on a per-play basis; it just didn’t translate into any more than 17 points.  Again, at this point, I have nothing to say here.  The Lions just didn’t execute on offense—and their insistence on playing Jahvid Best when he clearly has nothing in the tank isn’t helping.  The defense—until the final two series—did its job, and the offense wasn’t that too bad either.  It’s just the mistakes, the breakdowns, the penalties, the turnovers, and the incredibly unfortunate timing of all of the above that keeps the Lions making a fool of me in these posts.

Technorati Tags: nfl,detroit lions,dallas cowboys

Read more...

Watchtower Review: Lions at Bills

From last week’s Watchtower:

The 26th-ranked Buffalo offense has been averaging just 18.8 points per game, by far the worst Chan Gailey offense we’ve looked at. The Lions, ranked 23rd, are allowing 22.4 points a game. In theory, we should split the difference here—but applying a moderate systemic advantage, I’ll project the Bills to score 13-16 points. I’m not even going to attempt YpA and/or YpC, given how all over the map those numbers are this time, but I have high confidence in this scoring projection.

Good.  Yes.  Well done, data. 

So what’s going to happen when the 7th-ranked Lions offense comes to town? Since most of those points were scored with Shaun Hill at the helm, there shouldn’t be any dropoff from the usual rate of production. Even if there is . . . well, we’re talking a team that averages 25 points a game going up against the worst defense in the NFL. Even if we presume there is no systemic advantage, the Lions should score 30-35 points. Due to the total lack of systemic historical data, I have low confidence in this projection.

Somehow, the tenor of the national conversation surrounding the Bills-Lions game has become “Oh, sweet, the Bills will get a chance to get off the schneid! They pretty good for 0-8; they just almost beat the Bears, you know.” But the numbers just don’t support it. This is one of the best offenses in the NFL going against, by far, the worst—and on the other side, a mediocre offense against a mediocre defense. The Lions have a clear upper hand in this game, and mostly likely will win, 30-14. Hey, twenty-five fell demons of meadow grazing and road defeats! We are apex predators; you lose.

. . . sigh.

Frankly, I've already beaten this to death, so I won’t be long.  The Lions came out with a one-armed quarterback who couldn’t take too many snaps from under center.  Their game plan?  Run it down their throats, minimize risk, and play a low-variance game where their superior talent and execution will bring home the W.

Problems: the out-of-the-shotgun running game wasn’t working, at all (see Tom Kowalski’s play-by-play film breakdown of the Lions’ running game).  Shaun Hill couldn’t throw it downfield, and wasn’t throwing short passes with any accuracy either.  Shaun Hill was physically incapable of opening it up, and Drew Stanton hadn’t practiced all week.  The Bills came to play as if their lives depended on it, and the Lions came to play as if victory was inevitable.  Finally, when all that registered, and the Lions realized they had to flip the switch or lose, Hill’s rustiness, the rain, and some flat-out concentration lapses neutralized the Lions’ downfield attack.  Even then, converted on fourth down multiple times to score the needed touchdown . . . except, of course, Hill’s hand-slip & subsequent airmailing of the game-typing two-point conversion.  Right.

Outside of marking the Lions’ offensive projection way down (way, way, WAY down) for Hill’s rust, there’s really no way I could account for this—I had no reason to believe Jahvid Best would be completely ineffective against a run defense allowing 4.87 YpC, and no reason to believe that the Lions wouldn’t be able to go to the air if the running game wasn’t working.  Poor showing, lads.

Technorati Tags: nfl,detroit lions,buffalo bills

Read more...

watchtower Review: Lions vs. Jets

Well, last week’s Watchtower had an obvious error:

My instincts tell me that a team the Packers shut out at home won’t put up 24-27 against this defense in front of a sold-out Ford Field. Despite all the ways I could be proven wrong, I’m going to go with the data and my gut. The most likely outcome of the game is a 17-13 Lions win.

Durr Sharks?

Durr Sharks.

On the positive side, this is only the second game I’ve been wrong enough on to actually get the winner incorrect.  On the negative side, damn.  I’ll admit, I did not think Sanchez had that just-before-halftime bomb in him.  Not only did it unquestionably deflate the crowd, and reverse the tide of momentum as they went into the lockers, it also accounted for a huge percentage of the Jets’ offense that day.  That pass was the difference between 336 yards passing and 262 yards passing.  It was the difference between 8.62 YpA and 6.89 YpA.  It was the difference between 23 points and 16 points.  It was the difference in the game.

If we assume that the 2004-2005 Chargers provide a representative sample of Brian Schottenheimer’s schematic and playcalling tendencies, the Jets’s offense will either meet—or vastly underperform—expectations against Gunther Cunningham’s aggressive 4-3 defense, depending on which phase of the offense the Lions attack. Given the recent success of the Lions’ pass rush and secondary, I expect the Lions to attack the pass—and therefore, I project the Jets to score 10-13 points. Unfortunately, because we’re working off of a possibly fallacious assumption—the 2004-2005 Chargers’s offense being interchangeable with the 2010 Jets—I have very low confidence in this projection. If there were no systemic advantage or disavantage, the expectation for the Jets’ offense against the Lions’ defense would be 24-27 points.

Okay so.  The assumption has been proven fallacious.  [[ Martyball == Brianball ]] returns false.  Looks like either there’s no systemic advantage or disadvantage for Gunther against Brian Schottenheimer, or that one lapse in coverage threw everything out of whack.  Either way, this half gets a grade of FAIL.

With no systemic advantage or disadvantage, expectations for the Lions’ offense versus the Jets’ defense would be set at 17-21 points. However, if we apply this perceived disadvantage when facing Rex Ryan defenses, I project the Lions will score 15-17 points. I have low-to-medium confidence in this projection.

This is not a big surprise; the only data point for Linehan against Ryan was the 2007 Rams, after they’d lost Steven Jackson and anyone else who had any shot of being a productive offensive football player.  The only reasons I considered a possible advantage in effect were: A) how mediocre the Ravens’ defense was that season, and B) how thoroughly said defense scoured the Earth clean of Gus Frerotte on that day.

Lessons Learned:

Seems like I have to keep learning this one, but: if you have very low confidence in your analysis because of sample size or mitigating factors, do not roll with it anyway because it sounds right.

Technorati Tags: nfl,detroit lions,new york jets,rex ryan,brian schottenheimer,jim schwartz,scott linehan,gunther cunningham

Read more...