Writers of Pro Football Prospectus 2008

06 May 2013

Jaguars Start Anew

Some of you may already be aware that the Jaguars have brought in a lot of people and committed a lot of resources on the football analytics side. Albert Breer has some specifics on the internal thinking on their draft:

  • "Some of the studies Tony did helped us come to the conclusion that taking an offensive tackle (with the second pick) was a good route to go," Caldwell said. "He had the passer rating broken down from where the quarterback had 2.6 seconds to throw to 2.5, the amount of pressure, the sacks we gave up, third-most in the league last year. We used a lot of the stuff Tony put together. ... It's part of the process, not the whole process. Maybe we miss something, but we're trying not to."
  • The analytics also helped later in the draft. The Jags identified safety Josh Evans as undervalued -- a Round 2/3-type who gave up a completion percentage of just 35.0 last year at Florida -- and got him in the sixth round. They saw Tobais Palmer the same way -- a player who averaged 25.7 yards per kick return and caught 66 percent of the passes thrown at him by a 57 percent passer (N.C. State's Mike Glennon) -- and signed him as an undrafted free agent.

Posted by: Rivers McCown on 06 May 2013

12 comments, Last at 10 May 2013, 9:18am by lumberjack


by fb29 :: Mon, 05/06/2013 - 4:57pm

Is that second bullet really advanced thinking for an NFL team? That would really surprise me.

by Tim Wilson :: Mon, 05/06/2013 - 5:12pm

Agree. I don't even think the first bullet is that crazy. For an NFL team purporting to be doing something very different analytically, I guess I was expecting a bit more. Analytical research and statistical engines that larger budgets could make possible. Of course these are just public quotes, so maybe there's all types of regression analysis and algorithms that we're just not seeing.

by Theo :: Mon, 05/06/2013 - 6:13pm

Sometimes there's not much more than meets the eye.
I think some NFL coaches will use stats like these on FO, but for the most part I think they go with their collective eyes.

It sounds kinda obvious what the guy is saying.
"Let's protect the QB, he will become better that way".

by Richard Craswell (not verified) :: Mon, 05/06/2013 - 6:54pm

Although the original article did mention "football analytics," that wasn't the article's main point (for example, it's not even mentioned in the article's original headline). As a result, it wouldn't be surprising if the stats that impressed the reporter who wrote the article aren't the kind of stats that would impress most readers of FO.

That said, is it possible that one of the statistics mentioned -- the completion percentage allowed by a defensive back -- actually IS a little more sophisticated than what some scouts, or even some front offices, still use? I mean, it's one thing for a coach to look at the tape on a Josh Evans, and later conclude something like, "I don't remember him as being involved in a lot of passes." It's another thing to actually COUNT the number of passes (both completed and uncompleted) that he was in on, and then to compare that number to, say, the league average.

Okay, maybe I'm unduly pessimistic, and maybe this sort of not-very-sophisticated use of nontraditional stats goes on in every front office, all over the league. But if that's not true, and if there are many teams who haven't even gotten THAT far, then Jacksonville deserves at least a little credit for starting to move in the right direction.


by Karl Cuba :: Mon, 05/06/2013 - 9:50pm

I think your first paragraph is spot on, if Breer started by touting advanced metrics on NFL.com then frankly he wouldn't be doing his job. It just isn't accessible for the average peruser of that site.

I find the Evans pick interesting too. I really liked him as a prospect and was perplexed he didn't get more attention predraft. Knowing the 49ers would be in the market for a safety I had taken the time to have a good look at most of the top prospects at that position and while looking at Elam I kept noticing Evans. If you looked at the Texas A&M game and were asked which was the potential first rounder there's no way anyone would plump for Elam and not just because he couldn't tackle a thing, Evans was impressive.

by RickD :: Mon, 05/06/2013 - 5:02pm

It's funny how this article cites analytics as supporting the decision to take a tackle at #2, while FO/ESPN insider recently posted something arguing that KC had made a huge mistake by picking a tackle at #1.

I was suspicious at the time that the reasoning behind the other article was flawed, but could not read it properly since it is behind a paywall. I'd be interested in any comparisons of the two.

FWIW, this article seems sound. I'm intrigued by this argument that Gabbert is actually a decent QB if he gets 2.6 seconds to throw. And by "intrigued" I mean "a bit suspicious."

by artmac (not verified) :: Mon, 05/06/2013 - 6:30pm

I seem to recall reading that 2.5 seconds is the average time an NFL QB gets on a dropback? or maybe it's 2.8, but definitely under 3. either way I can't imagine an extra tenth of a second will magically make Gabbert not terrible, or that the Jaguars actually believe that. I think it's just a fancy, hairsplitting way of saying they think he can be better w/more time to throw, which is obviously true of any QB. whether Gabbert improve enough to become decent I share your doubts.

what were they really gonna do tho? it's not like there were any QBs this year worth investing the next couple years of your franchise on this year, even the teams that took QBs relatively high had some kind of safety net in place. granted next year's QB class doesn't look too hot either unless I'm missing something (which I could cos I don't follow college ball too closely) but giving Gabbert one more shot behind an improved o-line doesn't sound like the worst idea given the circumstances.

by erotica (not verified) :: Mon, 05/06/2013 - 9:13pm

The other article basically says the past conference champions (excluding this years 49ers) had average or at least not elite left tackles so its unnecessary. The other main point was that offenses today have changed to reduce the impact of elite tackles.

by bravehoptoad :: Tue, 05/07/2013 - 11:36am

Benoit is a scout, not an analytics guy, so his arguments weren't analytic at all. Counting the quality of tackles on championship teams, then counting the superbowl trophies of top-tier tackles, was as advanced as it got.

by Scott C :: Tue, 05/07/2013 - 12:22pm

As a scout he should have at least had an alternative, better pick to make.

Its not like they passed on Andrew Luck.

by Dean :: Tue, 05/07/2013 - 2:00pm

I get the impression that this article doesn't go high enough up the food chain. The fact that the team hired a new GM with an increased emphasis on analytics is only one part of a greater whole. The real story is that Shad Khan understands that he inherited a mess and is changing the culture of the organization down there. By his own admission, he had no idea what he was getting into when he bought the team, but he seems to be learning from his mistakes. If you believe that good owners set the tone for the entire organization, and that bad owners are barriers to sustained success (and I do), then it's time to keep one eye on the Jags. They might not be ready to turn the corner yet, but their days as a doormat just might be numbered.

by lumberjack (not verified) :: Fri, 05/10/2013 - 9:18am

I liked the OT pick, just because I wasn't a fan of any of the highly touted QBs, but this statement by the Jags seems odd. I have to wonder how large a sample size they had for plays in the 2.5-2.6 second range, during Gabbert's reign of terror. Maybe the data was there, but it seems peculiar.