Last winter, I had interviewed Neil Hornsby, founder of Pro Football Focus for a piece with Patriots Football Weekly. That column never ran. Here is the raw material from the session I did with Hornsby:
First, PFF has become really big in Boston. Just about all of the major outlets here have used statistics or analysis from PFF regularly this season. Is this indicative of all NFL markets, or has Boston been a little quicker to grab onto your information? Are you bigger in some markets than others?
Lots of people use our information but it is more used in certain areas. I would guess the other markets where we have better traction are New York, Chicago, Buffalo, Green Bay, Miami and Philadelphia. Boston is probably towards the top end of those areas.
I guess it’s the sheer amount of data we collect. We work with over a quarter of the NFL teams in one way shape or form and are usually flabbergasted by what we have. In summary:
- Who was on the field – in 2010 this was 99.83% accurate but we didn’t double hand most games then – this year we do so I’m predicting well in excess of 99.9%
- What position they played (at a level which allows us to provide formation as well as package information)
- What they generically did (block, pass route, cover, pass rush etc.)
- A measure of how well they achieved what they attempted to do (obviously we don’t know their assignments so this is what we use)
In addition we clean up a lot of the problems in the base data provided by the NFL; idiosyncrasies between different scorers and inaccuracies in terms of tackles, assists, hits, passes defensed etc.
Any aspirations of latching on with a major media company/outlet such as what those sites have done? Or do you prefer to stay independent?
We do sell information to the media outlets but we don’t write content specifically for them. It’s nothing we’d rule out but if it stopped us from working with the teams and player agents I don’t think it would happen.
How do you view all the games – an NFL Sunday Ticket sub with lots of DVRs, or do you use the NFL.com GamePass – GameRewind?
We use Game Rewind. A number of people question how accurate we can be with this but I think 99.9% playtime accuracy pretty much obviates that criticism. In my own little world of innocence I still like to believe we are the reason the NFL changed it’s policy this year to allow teams to get playtime information directly after games (rather than having to wait to season’s end) but insiders tell me our work is more accurate on the first pass than the NFL’s before they clean it up later in the week.
One player question – Jerod Mayo, He’s got his critics in New England, who basically say he is all hype and no real contribution/production. Sure, he has a lot of tackles, but they’re all down the field. What do your stats say about this player?
Sam Monson has analyzed the most Patriots games so I asked him for his view on the next two questions. This was his response:
Before this season we might have agreed with you about Mayo. Despite making a lot of tackles he was rarely a downhill, impact type of player and seemed more of a clean-up specialist than a linebacker really influencing the game. In this regard he was not unlike the Jets’ David Harris. This season though, the move to the 4-3 seems to have helped him be more of a downhill force. He is making more of his tackles in the right area. 39 of his tackles have been defensive stops this season, a far better ratio than in previous seasons where he was aligned slightly further back from the line of scrimmage so that he could control both sides if necessary. The hype for Mayo has always outweighed the real impact he has on the defense, but it would also be somewhat unfair to say that he does little of substance, as his contribution and reliability are not to be underestimated, and he does get involved in a lot of plays, even if it is almost always fewer than the generosity of the New England scorer would suggest. The area where he still struggles the most is in coverage, where he has allowed 416 yards this season with 319 of them after the catch. Only four other 4-3 OLBs have allowed more YAC than Mayo has this season and throwing into his coverage yields opposing QBs a 92.0 QB rating.
OK, I lied, second player question. Chad Ochocinco. Anything you can tell us about his lack of production/involvement in the offense? Is he doing things incorrectly? Is at least his blocking acceptable?
As for Chad Johnson the first thing to note is obviously what kind of small sample size we’re dealing with. He has only played 331 snaps and been targeted just 31 times. On those 31 passes he has obviously only been able to haul in 15 of them, which is a pretty poor catch rate. He still runs some really good routes on occasions, but he doesn’t seem to have the sudden burst and change of direction that he once had. Teams don’t fear his athleticism anymore the way they once had to, and he hasn’t really been able to get open consistently on the intermediate routes that the Patriots looked for him on. You also still see a reasonable number of disconnects between him and Tom Brady regarding where he should ultimately end up on routes. Part of what makes the Brady to Welker connection so great is that they are always on the same page when it comes to reading the coverage and adjusting the option route accordingly. Ochocinco is still not where they would like him to be and you will still see Brady miss him because he didn’t make the adjustment to his route that Brady was expecting or wanted him to make. There’s no reason why he can’t learn that going forwards if they keep him around, but it probably highlights the subtle complexities to the Patriots passing scheme that people don’t really appreciate. As for his blocking, again we are talking small sample size, he has been on the field run blocking just 87 times, but he has certainly not disgraced himself and is a willing blocker if not exactly an accomplished one. He at least deserves some credit for accidentally blocking two defenders in the Denver game on a pass that Aaron Hernandez was able to take for a big game because of Ochocinco getting in the way of the Broncos defenders.
An observation/question – it seems to me at times that there is a subjective angle placed on grades – I’ve seen cases where it seems like a better player gets graded tougher than an average player, is more expected of the star player? Or speculation about the player sneaks into the analysis “player X doesn’t seem right” – does that somehow finds its way into the grade?
No it doesn’t. The mantra is: grade the play, not the player and let the dice fall where they may. We have so many people still writing and saying “your grades must be wrong because X is better than this” or “Y is worse than this” and we have to be able to provide an audit trail of our work – we have to be accurate and accountable. We might say (as I did earlier this week) “it was a shock Calais Campbell graded so poorly because he’s a had a great year but the grade stays and it’s that methodology that allows us to “discover” stars before anyone else; guys like Cameron Wake, Carl Nicks etc. Just because Evan Mathis isn’t a household name won’t stop us standing by the courage of our convictions and naming him our all Pro LG and if people want convincing we’ve got a whole database of plays explaining exactly why that’s the case.
Finally – you’ve got the individual statistics and grades down to a science, and I know you put them into cumulative team stats. Has there been thought or plans to trying to come up with more accurate team stats for offense and defense – we know that those units don’t always equal the sum of their parts?
First and foremost we are a player evaluation site and we are real sticklers for trying to completely nail something before we move on. We have an ethos of continuous improvement and I want to see maybe another year of getting better, talking to coaches, player personnel guys and the like to make sure we have our bread and butter down before we take on something new. That’s not to say I’m not proud of our information; it’s passed every test to date with flying colors and while new NFL teams come on board none have left.