I think we can all point to problems in the 2009 Patriots. Some have to do with coaching, some with play-calling, others to do with administration (as a sidenote: pay Vince Wilfork, please). Obviously, though, a lot of them have to do with the roster.

Bill Belichick is nothing if not a problem-solver. If he sees a gap in the roster, he’ll fix it through trades, free agency and the draft. In 2006 Brady ran out of receivers at critical times; in 2007 Belichick gave him a bevy of new toys to play with  - Randy Moss, Donte Stallworth, and some guy called Wes Welker. Yeah, I hadn’t heard of him either, but he seems pretty popular on Pats Pulpit. In 2007 the Pats were shown to have a slightly thin, aging secondary and linebacker corps, so you guessed it, Belichick went out and got a few guys in the draft. And in 2008, it was the O-line that was looking a little old and frayed, so what did Bill do? Drafted deeply for O-line guys and picked up a swag of rookies in ‘09.

So we come to 2010 expecting Belichick to fix the roster. We can all point out the individual positions to fill - WR, RG, DE, OLB, maybe RB or CB or T - and I’m sure we’ll see those discussed ad nauseum over the next few months. But when you analyse the overall roster - how Belichick arranged his 53 players, week in, week out - what was the root of most of the problems? I have a suggestion: the linebackers. Or, more to the point, the type of linebackers. I’ll explain why after the jump.

The Pats under Belichick have always had a particularly strong D-line. There's a reason why Hoodie uses so many first- and second-round draft picks on it - it's important. Sure, the Pats traded away Seymour and might have struggled a little because of it. But they have coped without him before - in fact, the Pats won a Superbowl with Seymour injured. So that indicates it's not (just) the D-line.

I suggest instead that the success of the early Belichick defences was predicated upon their strong, versatile, 3-down linebacker corps - names like McGinest, Bruschi, Pfipher, Johnson, Colvin, Vrabel spring to mind. That's not particularly controversial. But when I pondered why that previous success wasn't replicated by the 2009 Pats Linebacker corps, I came to a conclusion that I've never seen mentioned anywhere else - it's because the 2009 LBs weren't 3-down guys. I could be wrong, but I have three reasons why I think it's critical that they weren't.

The first reason is what it allows your Defence to do. In particular, having 3-down players allow you to disguise your plays - on one play, Tedy Bruschi might have bull-rushed inside, Willie McGinest might cover the flat, Ted Johnson drops back, and Mike Vrabel might blitz outside. Or they might all sit back in coverage. Or they might all blitz. We didn’t know. The opposition sure didn’t know. That was why it was so effective - it had a crucial element of disguise. And when Belichick chose to use guys off the bench - say, Colvin and Banta-Cain - the opposition offence had even less idea of what was going on. Colvin and Banta-Cain could play DE or OLB, they could put a hand down or stand up at either position, they could blitz, drop back, or cover the flats... all of which added to the confusion for the opposition offence. This versatility of personnel is played out in the excellent statistical analysis of The Hill: the years the Pats did well, there tended to be a higher number of players with 3+ sacks - 6 in 2001, 8 in 2004, 6 in 2007. That suggests that almost everyone in the front 7 was at some point used to blitz or create pressure - rather confusing for an opposition QB.

In comparison with those rather successful teams, the 2009 Pats were the polar opposite in terms of 3-down guys. While Tully Banta-Cain was admirable as a pass-rusher, he wasn't exactly devastating against the run. Adalius Thomas turned into a 2-down guy who was taken off for obvious pass plays. Derrick Burgess was another pass-rusher who didn’t exactly shine against the run. Jerod Mayo didn’t blitz. Gary Guyton didn’t either. Junior Seau? Umm… In other words, the opposition QB could predict exactly what the Pats D was going to do based on who was on the field. This is shown in the stats - the number of players with 3+ sacks? A paltry 4. And dropping into coverage is part of it, right? Zero interceptions by Pats linebackers in 2009. Zero. Bruschi got 11 all by himself during his career. It’s no coincidence that the two best non-Brady Quarterbacks around - Manning and Brees - annihilated the defence: they knew exactly what every guy on the field was doing because the D was easy to read. That’s the first problem.

The second problem is what it doesn't prevent the opposition Offence from doing. In 2009, not only were the opposition QBs not deceived by the D packages, but they could actually take advantage of the Pats' need to sub pass-rushers on and off. Teams with no-huddle or quick-read offences - the Bills, the Broncos, the Colts, the Dolphins, the Saints - are practically built to kill you when you're slowly subbing guys back on and off the field. Having three-down Linebackers prevents that vulnerability precisely because they don't need to be subbed off. Bearing in mind how often the Pats strike the Colts, being weak against the no-huddle is playoffs suicide - Manning will carve you up in a no-huddle in the playoffs if you need to sub guys on and off. He just flat out will. So the second problem is basically: it's weak against P. Manning.

The third problem is the crux of the roster problem - it prevents roster flexibility. The dearth of 3-down linebackers means you have to fill the roster with so many part-timers and situational players that a lot of crucial roster spots are taken up. There were up to 10 bit-part Linebackers on the roster; in the Superbowl years the total snaps went to only 4 or 5 or 6 starting-quality guys. Why are those roster spots crucial? Overall roster flexibility. The lack of available roster slots is why Benjamin Watson and Chris Baker were both on the field for 60%+ of the season snaps, yet got so few receptions - there was no blocking TE or FB on the roster, so they had to do it all. In fact, it was explicitly because of the LB problem - Belichick had to cut the only blocking TE on the starting week roster, Michael Mathews, to make room for Junior Seau. Cutting a TE solely to get another linebacker in a packed LB corps isn't ideal. Those lack of roster spots also contributed to Slater and Aiken being bounced from special teams into WR - they couldn’t keep an 'actual' WR (say, Greg Lewis) in the 53. Too many roster spots were eaten up by multiples of 2-down and 1-down linebackers instead of a single 3-down guy, and it impinged on all the other positions. It's anathema to how Belichick has operated in the past - he likes guys who can cover multiple positions, not having multiple guys cover one position. Once Belichick realised that he didn't have 3-down linebackers, his hands were tied for the entire season.

Bearing that major problem in mind, what would an ideal roster breakdown look like? Here is my offering.

QB: 2 slots.
Brady is obviously top-dog; Hoyer a useful and seemingly talented backup. While a third QB is always a possibility, I see it as unlikely; there are two other Quarterbacks on the roster - Edelman and Stanback. Stanback has filled in as emergency QB before, and could do so again.

RB: 5 slots.
The last three years the Pats have taken 5 running backs into the season, and every year there have been a few injuries to deal with. Five backs was just enough to cover injuries, so I see no need to change the status quo. A proviso: should one of them should be a full-time FB?

WR slots: 5
I don’t think the number is bad, despite the issues with the Wide Receivers last year. It was more to do with the personnel than the number of slots they took up, so I’ll stick with 5, especially since Slater and Aiken are considered core Special Teams in this analysis.

TE slots: 3

This is one change I’d definitely like to see to the 2009 roster arrangement. The Pats actually started off with 3 Tight Ends, but the young blocker Michael Mathews was cut early on. In hindsight, this seems like a mistake - given the O-line injuries to come and the lack of a fullback, Baker and Watson spent most of their time shoring up a weakened blocking scheme than catching passes. A blocking TE would've lightened the load.

OL slots: 10
Hoodie went into 2009 with 10 O-lineman, and it seemed like just enough - the Pats coped with an extraordinary number of injuries, but never completely ran out of willing and able linemen. While it might not have been pretty all the time, it was enough.

DL slots: 6
Again, no change from 2009. The Pats had just enough guys to deal with injuries to all three starting linemen at one point, so no change is necessary.

LB slots: 8
Preferably 4 OLBs and 4 MLBs, with the backup LBs contributing to special teams. Those every-down Bruschi types are worth their weight in gold, and are absolutely critical to a Belichickian scheme. Tully Banta-Cain is a situational guy, and I'm not sure Burgess is more than a situational pass-rusher, either. Pierre Woods was supposedly going to be a three-down guy and was due for a breakout year, but he disappeared into the nether-regions somehow. Adalius Thomas turned out to not be much of anything.

CB slots: 4
No real change from the playing roster as-is - the Pats took 4 corners to each game and never ran out. Special Teamer Arrington can cover the spot in an emergency.

S slots: 4
Again, no change - the Safety corps was effective enough and was never troubled by injury, so it had enough depth.

ST slots: 3
The Pats had two particularly good ST Gunners who didn’t see a lot of the field outside of special teams - Matthew Slater and, in particular, Kyle Arrington. The status quo worked well. Sam Aiken was a good general purpose, all-units Special Teams guy, and if he's bumped out of a WR slot, he'd would earn one here. All three could find the field in other modes, but keeping them fresh as specialists might work wonders.

K slots: 1
Only one kicker necessary, especially if he can drill deep kickoffs regularly. Welker can cover it in an emergency (if he’s available).

P slots: 1
Only need one; there are always unemployed Punters around.

LS slot: 1
Again, one, especially since Rob Ninkovich can cover the position as a backup.

Now we've got a rough roster to fill, I'll come up with a following post examining each position to determine who will stay, go, and if they go, how they could be replaced. It's mock-draft time! 

Thanks for reading.