1/n The news of Judd Brackett leaving got me thinking about something that often boggles my mind – the degree to which ideas that are common in the private sector look like fire to a caveman for professional hockey teams.

2/n The case got me thinking about William Thorndike’s book, The Outsiders (not ponyboy). The book is about managers whose companies significantly outperformed their peer-groups over the long-term. I believe there’s lessons for Jim Benning here (and the rest of the NHL).

3/n The central premise of the book is that in organizations with large, unrelated processes, the CEOs who outperformed embraced three things:

4/n 1) they have a general, high-level knowledge of each process in the company 2) they put competent, specialized people in charge of each process (with of course, some understanding of unrelated processes) and maximized their autonomy

5/n And most importantly: 3) They adopted capital allocation (basically managing resources) as their main focus and job requirement. Effectively, this entailed judging internal processes and giving resources to those they felt would create the most value.

6/n All this culminated in lean head-offices with some input, but little role, in lower-level decisions.

7/n Now a lot of you out there will say that hockey is a different domain and that these ideas have no merit. I couldn’t disagree more, but I’ll outline three general differences:

8/n 1) The assets allocated in hockey include the internalized budget, cap-space, draft picks and roster spots as opposed to cash

9/n 2) The domain is constant. Significantly, this results in more overlap between the functions of a hockey team. Talent evaluation is the most notable skill that encompasses almost all aspects of a hockey operation (player development, coaching, scouting etc).

10/n 3) In hockey, the value of an asset is dependent on the stage of the team’s “cycle” and to a degree, positional need. Your goal is not just to have good players, but instead to have good players at peak output at roughly the same time.

11/n Keep in mind, as a result of #3, two values for assets can be derived 1) the value of the asset to the team 2) the value of the asset to the “market” (other GMs). When 2 > 1, the team makes a trade.

12/n So with this backdrop, here are the three roles that I believe should be embraced by NHL GMS:

13/n 1) Maintain a high-level understanding of the domain you’re competing in and the functions that support success in that domain (player development, contracts, scouting etc). Talent evaluation is the most transferable skill in an operation, and a manager needs that

14/n 2) Put good lower-level managers in charge of these functions with autonomy. Why? No person has the capacity to effectively perform every function in the organization because it's too highly specialized. But the most important function of a GM, IMHO is:

15/n 3) Focus on developing a plan around the allocation of resources (including draft picks, cap-space, roster spots, internal budgeting, and current roster players) to maximize roster output at the same time.

16/n Now how does this relate to Brackett? Well first off, anytime an organizational function meaningfully outperforms over a significant sample size, that function is worthy of attention. The Canucks amateur scouting department has certainly done that.

17/n Brackett's request for autonomy and subsequent firing is clearly in violation of managerial job #2.

18/n As I said before, the individual operations in a hockey department are too specialized for any one person to do effectively. The problem is that the overlapping skill of "talent-evaluation" deludes GMs into thinking specialized managers are unnecessary.

19/n Of course, this couldn't be further from the truth. The drivers and nuances of the amateur draft are very different from those in (for example) contract negotiation. While talent evaluation is necessary in both, both also bring specialized skill requirements.

20/n Now, having said that, the ability to transfer talent evaluation does give room for outside input to mitigate groupthink. However, this should not lead to overstepping.

21/n With the meaningful outperformance of amateur scouting, it seems unlikely that Brackett's group is conducive to groupthink. As a result, Brackett is correct to demand autonomy, and Jim Benning is failing dramatically at role #2.

22/n But Benning has been awful at his primary role as a manager, role #3. Worse than any in the NHL imho. It's not even close.

23/n So in sum my thesis is this: a hockey operation requires different specialized skills to maximize output. They also have a collection of resources that need allocation to create a team that performs well at the same time.

24/n In the private sector, the situation is almost the same. The best CEOs primarily focus on resource allocation/strategic direction, and give autonomy to lower-level managers to make specialized decisions. This is transferable to hockey and should be embraced.

25/n I don't know why this isn't studied/embraced in the NHL, but I can think of three reasons:
1) No profit loss in Canadian markets when management sucks.
2) Managerial skill in hockey needs a broad sample-size to measure
3) Wins are more based on luck than profits are.

26/n To close, today I'm presenting a decentralized operation with a heavy focus on resource allocation as the fire to the cavemen that are hockey guys. Have a great day .