Yahoo Sports 
Eric Tillman, Jim Barker, Joe Mack and Ross Hodgkinson: what matters for a GM?
Why the Edmonton Eskimos made the bizarre decision to fire general manager Eric Tillman ahead of their playoff game against Toronto hasn't become any clearer over the past week, but several other curious decisions around the league recently have raised the question of what CFL teams value in personnel staff. From the Toronto Argonauts' move to extend the contract of head coach Scott Milanovich while leaving general manager Jim Barker hanging to the Winnipeg Blue Bombers' decision to retain besieged GM Joe Mack while axing assistant GM Ross Hodgkinson, who had been in the organization for 27 years, there's plenty to wonder about. Perhaps the most interesting element of these moves is how it's come out that some of them aren't purely about on-field performance.
That's clearest in Edmonton, where president Len Rhodes' comments about why the team parted ways with Tillman only muddied the waters further. Rhodes said there was "no specific reason" why he fired Tillman and that the controversial Ricky Ray trade wasn't why he made the move, but the one specific element he did cite was...where Tillman lived?
Through the generalities of his reasoning for Tillman's dismissal — which comes eight days before the Eskimos travel to Toronto to face Ray and the Argonauts as the crossover opponent in the East semifinal — Rhodes did point to Tillman's lack of presence in the community as an issue. When he was physically in Edmonton over the last two seasons, Tillman stayed in a hotel. His family still lives in Regina and Tillman often spent time there and also in Mississippi, where he grew up and his parents still reside.
"If we're looking at a future GM, my preference would be that the GM lives here in Edmonton," said Rhodes, who moved to Edmonton last year from Montreal when the Eskimos hired him.
"We're a community-owned team. When I took the job last December I made a clear decision that I'm going to live here. I would hope that the GM will also live here in the future."
Sorry, but that seems like extremely specious reasoning. If a general manager's job is to show the flag in the community and attend public-relations events, fine, but if the job is to construct the best possible roster, why would it matter if the general manager does it from the Amazon? (They're improving those satellite phones all the time.) That's particularly true in the offseason; plenty of successful executives like Montreal's Jim Popp head south of the border for the winter, and that hasn't stopped Popp from building a dynasty. In fact, considering that there's more going on in the football offseason in the States (free-agent tryouts, spring leagues, the NFL combine, etc) than there is in Canada, being based there may even prove an advantage.
Simply put, while coaching certainly requires some face-to-face interaction (although many great CFL coaches like Montreal's Marc Trestman also leave for the winter), the general manager's job really shouldn't need any in the offseason and only needs a little in-season. From this standpoint, the job's about building the best roster possible; where you live in the off-season shouldn't have anything to do with that. However, with the Eskimos and Tillman in particular, it clearly wasn't all about on-field results. Otherwise, why would they fire him before seeing how the roster he built does in the playoffs?
There seems to be somewhat similar logic applying in Winnipeg. Retaining Mack despite the Bombers' 6-12 season (where many of the failures were clearly attributable to the personnel side) suggests Winnipeg certainly wasn't evaluating things based on this year's record (and it's worth pointing out that while a Mack-built squad went 10-8 last year and made it to the Grey Cup, he let most of the key components of that group walk for little or nothing, while his 2010 team went a league-worst 4-14). It's tougher to see exactly what they are evaluating him on, though.
The Hodgkinson move is also curious. Sure, history with a franchise isn't everything, and if they think they can find someone to do his job better, fine. What's bizarre is that there have been suggestions of that move being made in order to bring in an AGM who can eventually replace Mack, though; it's highly unusual to hire someone under those circumstances, and it gets weirder when you consider that "willingness to live in Winnipeg year-round" is also apparently a prerequisite (Mack returns to North Carolina for much of the winter). There are plenty of reasons for blasting Mack, but where he lives isn't the best one, and requiring a new candidate to live in town year-round may reduce the calibre of the talent pool they're able to draw from.
The Toronto situation is a little more ambiguous. There are arguments both to keep and remove Barker, as he's led the Argonauts to the playoffs twice in three years and acquired talents such as Ricky Ray and Chad Owens, but has also made unusual moves like releasing Cory Boyd and hasn't exactly built a dominant powerhouse yet. However, letting him twist in the wind while bringing back Milanovich is curious, and locking up Milanovich before seeing how he does in the playoffs is also weird. It's not hard to imagine there's more being considered here than just the Argonauts' on-field performance.
Of course, general manager evaluation shouldn't be strictly about wins and losses, either. For one, the state of the team before they show up has to be considered, as does the job their coach does with maximizing the talent they provide. There are cases where there's little talent and okay utilization of it, which reflects poorly on the general manager (Winnipeg this year), and others where there's great talent but a coach can't figure out how to effectively use it (see Edmonton this year). For another thing, draft records matter, as do free-agent signings and trades, and there are sometimes circumstances that aren't particularly under a general manager's control (such as injuries to top players). Still, there's a much stronger case for evaluating general managers based on how they build a roster than where they live and how much they do in the community. CFL teams should do community outreach, of course, and it's great if the GM wants to chip in there. Any general manager's key job has to be to build a winning roster, though; everything else should be merely a nice extra, not a core criterion for the job.
