The Watchtower Review

>> 1.10.2012

the Lowenbrau Lion, by Adrian Valenzuela

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?

9 comments:

formyministy,  January 11, 2012 at 11:54 AM  

ok, thoughts are as follows: i love the watchtower. i look forward to it and get rather annoyed when it is delayed and/or missed altogether (although i do understand you have a life outside of tliw). the bias is what it is. it's the reason i suck at fantasy football and picking games. i let what i want to happen get in the way of what will happen. but if you figure out a way to completely remove bias from the equation, i truly believe that you will be the FIRST one to do so. all the talking heads at espn, the bloggers, walterfootball, fox, cnnsi, etc...none of them have removed bias from their equation. you are already head and shoulders above any other source that i'm aware of in the amount of hard data that you use. (i thought i might destroy the radio in my truck when colin cowhead was talking about the lions before the saints playoff game, and basing all of his predictions on the week 17 green bay game!)
by all means, tweak it, play with it, and do what you feel makes it as accurate as possible. just don't stop with the watchtowers!
ps already looking forward to the old mother hubbard offseason features as well!

Jeremy Reisman,  January 11, 2012 at 12:23 PM  

I have to say, I run into almost all of the same problems and criticism with my previews, and I, too, am looking to figure out how to right for these issues.

It's really tough to find an appropriate statistic for special teams. Punt average is just a god-awful stat, as the distance of the punt is almost completely dependent on position on the field; same goes with punt returns. Kickoff stats are more dependable, but their significance in the game has gone down since moving the kickoff up five yards. I used to have special teams in my preview, but got rid of it when I found the stats misleading and ineffective in predicting anything.

Same goes with turnovers. Sure the stats are there, but they are hardly predictive. Outside of a few teams who are very careful with the ball (Packers), it's incredibly hard to predict who will win the turnover battle. This is where I think your "possible outcomes" idea would fit the best. Create multiple situations and examine what might happen if the Lions win/lose/draw the turnover battle, because trying to predict who will turn the ball over most will become a fruitless endeavor.

Anonymous,  January 11, 2012 at 1:31 PM  

i just typed a four paragraph response that was (sort of) well-written, but it vanished instead of posting. don't feel like re-doing it, but suffice it to say, love the watchtowers! please keep doing them, and i can't wait for the old mother hubbard features.

Robert theaker,  January 11, 2012 at 3:01 PM  

I aggree with you. Have you considered weighting the turnover buy a value such as probablity of opponent score by turnover field position and offensive strength.

DetFan1979 January 12, 2012 at 11:08 AM  

Ty - I think you are over-thinking. I don't do predictions just because of the factors you are stating. The best of the best at picking games and predicting them does no better than picking "home team wins" for every game. In essence, "any given Sunday" truly is an NFL fact. Because it is a human game, and the human factors are so prevalent even with all factors taken into account, you still get accuscore predicting a Saints shutout of the Lions.

Really? Really? So I would say to stick with your format -- or whichever format you prefer -- but omit the final score/prediction. In a way, it would make the article even better.

Why? because readers can then take your info, and their own bias to predict a final score based on their interpretation of "the human factor". You can even ask for it in the comments

Matt,  January 12, 2012 at 2:19 PM  

The first thought I had was on the last issue you brought up (bias). Easy fix: lay out what the statistical analysis points to as the final score, then do the personal analysis, taking care to clearly distinguish one from the other.

I also have to disagree with Dennis that a final score prediction can be omitted. Putting all that effort into analysis begs the question: so, given all that, what's gonna' happen? It is also SOP in analytical sportswriting. If you're not trying to predict a winner, then what's all the work for? If you're right, you're a guru. If you're wrong, you might get lightly flamed, but everyone's wrong about this stuff most of the time. So who cares? It's definitely the analysis that brings folks in (and back), but the final score prediction is obligatory, IMO.

Anonymous,  January 12, 2012 at 2:26 PM  

Before we discuss this any further, where is that tower in the picture and how would one go about stealing the Lion on top of it?

DetFan1979 January 13, 2012 at 2:38 PM  

Ty -

I think you should just drop the game prediction part, include the stats analysis like you do already -- and then add a reader component -- give us your take, and then ask your readers to comment on how they think the "human factors" will push the stats for the game in one direction or another. would add some fun interactivity, and see who is most in tune with the Lions' human element each week.

Anonymous,  January 15, 2012 at 8:38 PM  

Based on the nefarious referee calls of last season, skip the OC vs DC comparisons and focus on outcomes based on who is refereeing the game.

Post a Comment


  © Blogger template Simple n' Sweet by Ourblogtemplates.com 2009

Find us on Google+

Back to TOP