Kirk's Hammer: Roster-Building Philosophy

Yesterday I tweeted about roster building philosophy and the Micah Parsons trade, and everyone was so encouraging about it and I kept getting messages like “please tell me more!” so I decided I had to put it into a newsletter. OK fine, that might not be exactly how it went.

Here was the tweet if you missed it:
“I've never really understood the roster building philosophy of trading a lot for a player to give him a market rate contract. Seems like a pretty great deal for Dallas IMO, understandable but not great for Green Bay.”

                            ________________________________________


What Are We Maximizing?

When it comes to roster building, the way I think about it is this: if I was the owner of any sports team, every move should be looked at through the lens of “does this increase my future Superbowl equity”, with a slight extra weight on winning the first one. I’d rather have a 50% chance of winning one Super Bowl than a 25% chance of winning two, even though the math is the same. The jump from 0 → 1 is simply more meaningful than 1 → 2.

That seems obvious, but it’s not actually what many owners and GMs are maximizing for. And honestly, that’s not even a criticism. If an NBA team goes 20–62 four years in a row, tanking a fifth year might be the right roster-building move, but it could also mean losing fans, revenue, and relevance.

Then there’s the GM incentive problem. Most GMs only get so long before they have to show “progress.” At a certain point, job security becomes more important than long-term championship odds, and that’s when fake progress moves happen. I think those perverse incentives are under-discussed and often explain the worst trades and signings you see in pro sports.

                            ________________________________________


Different Leagues, Different Philosophies

Roster building isn’t the same across sports. The NBA, NHL, and NFL all work differently.

NBA: The Superstar Economy

The NBA is the most unique league when it comes to roster building for a few reasons, one of them being its soft salary cap. Unlike the NFL and NHL, which operate under stricter hard caps, NBA teams can exceed the cap in certain cases — most notably with “Bird Rights,” which allow you to re-sign your own players even when you’re already over the cap.

That creates some strange math. For example, if the Lakers traded LeBron into cap space for nothing, they’d go from $58M over the cap to $5M over. Despite shedding $52M in salary, they’d still have basically no new flexibility to sign free agents (other than a little additional room in some exceptions). This makes NBA transactions very different — you can’t just dump money and reallocate it elsewhere like in other leagues.

On top of that, fit matters differently in the NBA than it does in the NFL or NHL. Those sports are more about raw accumulation of talent: if your roster adds up to the most talent, you’re probably very good (with the caveat that in the NFL you almost always need a top QB). The NBA doesn’t work that way.

Even if you had the 30th through 42nd best players in the league all on the same roster, you’d almost certainly not be a championship team. There’s only one ball, stars dominate usage, and players can stay on the floor nearly the whole game. The vast majority of NBA champions are built around a few elite superstars, surrounded by role players who complement them without needing the ball.

That’s why the “meta” path to winning in the NBA usually looks like one of two strategies:

1. Tank, draft high, and hope your picks hit.

2. If you’re in a market where stars want to play, stay competitive and be ready to add them.

Exceptions exist, but they’re usually tied to big anomalies — like Giannis being found late in the first round or Jokic in the second.

NHL: The Surplus-Value League

The NHL is almost the opposite of the NBA. It operates under a very strict hard cap with almost no wiggle room. The league has even cracked down on loopholes like the long-term IR trick teams were abusing.

Because of that, roster construction is much more straightforward: you need to fit as much talent under the cap as possible. Fit doesn’t matter nearly as much here as it does in basketball — unless your roster is wildly unbalanced toward one position group, you can usually just focus on adding talent.

That’s where surplus value becomes everything. If you’ve got a rookie making $2M but producing like a $10M player, that’s $8M in surplus value. Stack enough of those contracts under the cap, and suddenly you’ve got a contender.

And this is why a Parsons-type trade rarely makes sense in hockey. In the NHL, you’d be giving up valuable assets just for the right to pay a player full market value. In other words, you’d be trading assets for zero surplus value, which is basically the opposite of how you win in that league.

NFL: Somewhere in Between

The NFL sits in the middle. It has a hard cap, but money can be pushed into future years using contract restructuring. This allows for teams to load up when they’re contenders, and pay the price later on. Almost always, you need a good QB. The ideal path is hitting on a QB on a rookie deal — because that surplus lets you stack the roster everywhere else.

                            ________________________________________


So, the Parsons Trade

From Dallas’ perspective, I think it was a great move. They were a distant third in their division and unlikely to contend this year, which is Parsons’ most valuable season: he’s at his youngest and cheapest. Even next year, they’d need a lot to break right to be a contender. After that Parsons is 28, making $45 Million a year and very unlikely to be worth two first round picks. The fair criticism of Dallas is that they waited until right before the season where most teams had no cap space — if they’d moved him earlier, they maybe could’ve gotten more.

Before I get into the Green Bay side, lets talk about why picks matter. People say, “The odds of two late firsts being as good as Parsons are basically zero.” And sure, that’s true. But those picks aren’t being paid $45M a year.

Over the last 20 years, picks 15–30 have produced roughly (according to CHATGPT):
 

• ~13% All-Pros
• ~30% Pro bowler
• ~55% starters
• ~24% busts

The point isn’t that you’re likely to draft Parsons. The point is that you might land elite production at rookie-scale prices. Would you rather Micah Parson’s for $45 million, or Brian Thomas for $3 Million? Obviously, that’s not the real question, but picks are an extremely positive EV gamble, because of the real possibility of getting huge surplus value.

From the Packers perspective, I got a lot of criticism for saying “I've never really understood the roster building philosophy of trading a lot for a player to give him a market rate contract.” I completely stand by that frame of thinking, but I probably could have allowed for a bit more nuance.

Generally, trading assets to get a player and then paying them what they’re worth is a bad way to build a team. But the Packers are in a pretty unique spot. They’re young, and Parsons absolutely moves their odds of winning a Super Bowl this year. The best argument for why the deal makes sense for Green Bay is that they get the Parsons upgrade immediately — any other move they could have made, or drafting those firsts, wouldn’t have given them that kind of bump this season unless they swung a deadline trade (which is unlikely).

That’s why I said the deal was “understandable but not great for Green Bay.” I think I’m maybe a little higher on it now, but I still believe it’s a big gamble — and it’s a sizeable bet that the Packers are a Superbowl contender right now, if they are, they probably got it right.

                            ________________________________________
 

Check out Kalshi and get signed up today: https://kalshi.onelink.me/1r91/hammer