Ask Your Numbers Guy Football Expert: James Todd of RotoViz

"So, NFL intangibles believers….Nawrocki is carrying your banner. Go forth and follow!"

I can’t call the process of picking new interview subjects a scientific one. Some weeks, it’s a matter of finding a person I find interesting. I’ll leave the detailed statistical study on the subject to the crazy, but in a good way, staff at RotoViz, and I hooked another one in James Todd. He can tell you where to put those intangibles.

What’s the Twitter handle “spidr2ybanana” all about?
It’s a re-purposed twitter account, so I wanted something footbally, and a little ironic, referencing back to Jon Gruden’s always enthusiastic MNF analysis.

You’re with RotoViz, where the stats flow like wine. How does one break through the intimidation of so many numbers, relax, and become a better fantasy player through science? When I first started playing fantasy football in the mid/late 90s, there wasn’t as much diversity of analysis. I followed the same advice as everyone else, and didn’t think too much of it. Whatever edge I had was related to just having more information generally than the guys I played against. Once I got invited to my first “competitive” league (with coworkers, for a small pot of $ and break room prestige), things changed. I followed the same advice I always had, and thought I’d do very well, since many of my leaguemates seemed to be making “weird” draft/trade decisions. But I got my rear end handed to me the first couple seasons, so I set about to find out “why”. I started by going back and looking at the “success” of the guys whose advice I’d followed. I thought that would be a good approach- figure out who gave the best advice and stop following those who gave poor advice. (Side note: this is basically what Fantasy Pros does- really good site). [Zach note: There’s a good reason why don’t contribute rankings to that site.] At the time though, that was tedious work. There were so many opinions on teams/players, and tracking them all for later judgment was a pain.

One of my leaguemates turned me on to Football Outsiders and I was hooked. From there it was Pro Football Focus, and then on to other, more “hard core” analytic sites and writers, each with less of a focus on making projections based on “watching” a player or team. To me, several things were immediately appealing about this approach:

Accessibility. Almost anyone can find and interpret statistical information – things like percentages, rates and ratios, probabilities, are easily grasped.

Forward progress. in my opinion, the evolution of stats-based analysis is always “falling forward”. That is, what changes and “conflicts” exist are often a product of a given model being refined, leading to a new and improved model that wasn’t possible earlier. Conversely, film-based scouting often seems to be focused on increasing it’s esoterism. In “digging deeper”, the focus of film scouting increasingly seems divorced from what ultimately matters: winning games, scoring (or preventing) points, winning matchups.

Replicability. The language of math/stats is much more standardized, so it’s easier to not only compare analyst A to B, but for analyst A to test the work of analyst B, and for analysts A & B to teach analysts C & D.

Sample size. This is frequently a criticism of stats-based analysis, but I think it’s equally, if not more, an issue with film-based analysis. Can an analyst relying on watching games/players really watch a significant sample size of plays? Is watching three games of a college prospect enough? How do you know if the competition in those three games was representative? To double your sample size, you have to double your time commitment. But using stats allows you to analyze every play/game, and to slice/dice by level of competition, etc. Doubling a sample size doesn’t exponentially increase your time commitment. I just think math-based analysis has an easier path to solving the sample size issue than film-based analysis does.

Time. Assume that film and stats-based analysis produced equal results. Film analysis takes some (probably large) amount of time more than stats analysis. Even if film analysis produces results that are X percent better than stats analysis, there’s still the time trade off. In other words, film analysis would have to demonstrate that it outproduces stats based analysis by a rather large margin in order to justify the increased time expenditure.

So, short answer, less time commitment, and more standardized/verifiable results. [Zach note: You provide the numbers and I’ll provide the beverages.]

Did you take Jim Schwartz off your Christmas card list? I feel like I need at least a thousand words about what it’s like to be a long-term Lions fan. Ugh. [Zach note: That will suffice.]

Suppose I should be just as mad at the Ford family, which continues to mismanage the franchise. Schwartz just strikes me as, personally, a pompous ass. It’s not even the losing- although the way in which the Lions collapsed under his leadership is pretty epic- we’re used to that around here. It’s the smirking. The headset throwing. The handshake debacle. The condescending tone with media/fans. Consider the end of Gary Kubiak’s time in Houston: he thanked the owner, players, and fans, took out a full page newspaper ad, and was given many a fine send off in return. Schwartz? Yelled at fans in the final games, blamed them for being bad fans in his pressers, and owned none of the Lions’ shortcomings. No give-backs, Buffalo. Being a Lions fan is like getting a new tattoo and then figuring out it’s mis-spelled. It’s hideous, and you’re stuck with it forever. And probably dumb for getting it in the first place. [Zach note: One thing Jim Caldwell isn’t going to do is throw a headset.]

What’s going on in this debate between “film” guys and “stats” guys? Yea, that got out of hand a little bit. There are some stylistic differences, obviously, that make tension natural. It can be fun to poke at each other’s foibles, but that gets out of hand sometimes, and I’m making a concerted effort to play nicer. I do think (no, really, I do) film scouting can add value to the process. I appreciate – and consume the work of – guys like Matt Waldman and Greg Peshek. What they have going for them are specific, reproducible templates and processes for scouting, and they connect it to the ultimate goal of winning/scoring.

A big part of the debate is a fundamental difference in Weltanschaung between the two camps, that leads to some Babel-esque misunderstandings. [Zach note: Yeah, I thought that word meant “penis size” as well. It’s world view.] But maybe even deeper than that is a fundamental biological difference- I think “film” scouting, even though many practitioners are advancing the process, is rooted to “tradition”, to conservation, to accepted methods. “Stats” based analysis is more forward-looking, boundary-stretching, and “risky” in the sense that it upsets the status quo, and opens us up to the “unknown”, which makes many people uncomfortable. Some people are just naturally more comfortable in one camp than the other.

Why are you RotoVizers so ageist? You don’t trust breakout wideouts who are older than 21, seems like. No AARP memberships here! It’s pretty straightforward, actually. If all else were equal, would you rather have your team sign a 31-year old free agent, or a 27-year old? The younger guy, right? So why doesn’t the same logic apply to incoming rookies? One argument I hear a lot is that older rookies have had more “seasoning” in college, and have better mastered the game. To me, that’s like looking at two identical make used cars, and buying the one with the higher mileage. It’s not that the older play can’t be a good- even superior- player, it’s just not as likely. Another way to look at it is like this: how old is the average college football player? Roughly 20? I’d rather have the 18 year old who dominated against 20 year olds, than the 23 year old who dominated against 20 year olds. The older rookie has roughly 3 years of physical maturity and experience on his side — he ought to be dominating! So if he is dominating, you can’t assume it’s because he has superior talent. Flip it around: for an 18 or 19 year old to dominate against guys with 2 or more years of age and experience speaks to the likelihood that he has a very high level of talent. [Zach note: There’s no denying it. James likes ‘em young.]

If someone tells you that they like to fantasy draft from the gut, do you punch them immediately, or invite them to join one of your leagues? Definitely invite! But not so I can dominate them, but hopefully learn something from them. Instinct-drafters often do unexpected things that, while I might not agree with, challenge me to think about ADP differently. Don’t get me wrong, I think it’s appropriate and desirable to use your “instinct” when drafting. But hopefully your instinct is refined by a good understanding of positional value, player outlook, etc.

Since the last RotoViz guy I interviewed won a quarter million last year, how did you do in fantasy in 2013? Not nearly that well. Still waiting on approval for my loan application to the First National Bank of Siegele to be approved. I did well in the MFL10 format, cashing in about 20% of my leagues. This was the first year I played DFS, and came in roughly even. I think I could have broken even, had I started with a better bankroll management strategy. I’m giving it another try this year. I’m a relative newbie at playing for $, but that’s part of the appeal at RotoViz- you’ve got money players big and small, so we’ve got all the angles covered. [Zach note: The DFS landscape’s going to change a lot next year since it seems like so many people were playing it for the first time in 2013.]

How is Detroit living? Did the RoboCop reboot never happen? I am robocop. Get off my lawn. It’s interesting, that’s for sure. We like to say that if you can do it in Detroit, you can do it anywhere. To me, it’s just a fascinating region. There is a ton of history- social, industrial, political, a lot of culture, and some real-life experiments in revitalization and reinvention going on. Also I’m really sick of shoveling snow.

The spouse question: Does your wife have any issues with your football obsession? We’ve set it up so that football and scrapbooking/karaoke have a 1:1 relationship. So it’s all good.

Football or sex: which do you think of more? Fifty-fifty. The only problem is when I think about one while engaged in the other. [Zach note: You say problem. I say multi-tasking.]

Where are you on the usual football Sunday? I assume there’s a man cave. A man cave is one of my life aspirations. As it is, I have four young kids at home, so Sunday afternoons often find me pushing swings or going for bike rides. I punish myself by watching Lions games whenever I can. I always watch the Sunday and Monday night games live though, and catch up on the afternoon games through Game Rewind.

Tell me about a hobby/interest outside of football. I don’t understand the question? Just kidding. I enjoy sci-fi and fantasy literature. Yes, nerdy. There’s also a great network of metroparks on this side of the state, so we get out for bike rides/canoeing whenever we can.

Tell me about any plugs that you’d like. Podcasts, sites that you write for, expert leagues, anything. The only place I’m writing is RotoViz. The site gave me an opportunity, and a lot of coaching and mentoring, for which I’m really grateful. When I say “the site” I mean “RotoViz Staff” who is a rather mysterious entity. I’m not sure who it is, but I will say that I’ve never seen RotoViz Staff and Agatha Christie in the same room at the same time. I recommend all the writers there, many of whom have their own websites or books, that are well worth your time.

The SportsWunderkind podcast with Davis Mattek and Coleman Kelly is a good time. Those guys are funny and pretty insightful, despite the generally messy rooms they keep. Overtime Ireland is fun for the accents and foreign take on our favorite game. There are a lot of great sites and writers, but I especially look for anything by CD Carter, Rich Hribar, Pat Thorman, Chris Raybon, Chase Stuart and Brian Burke.

Look at this guy, giving kudos to others in his own interview. All your stats belong to him. Follow James on Twitter and thanks for answering my questions.

, , , , , , , , , , , , , , , ,

Trackbacks/Pingbacks

  1. The CIA, the NFL, and Using Data Driven Models to Make Good Predictions | RotoViz - May 3, 2014

    […] one variable should receive more weighting than the other two. This is what I meant when I said “the evolution of stats based analysis is always falling forward.”38 This is also where a lot of data driven analysis can be improved; we may have a good idea of […]

Leave a Reply

Powered by WordPress. Designed by Woo Themes