Strategy IV – Managing the High End of the Roster
The first three in the series are listed here.
The first article discussed filling the roster with mid-range mid-dollar role model FA to fill holes so one can use BPA in the draft. The second focused on how to use inefficiencies in the draft chart to minimize risk and maximize value. These two articles discussed how to add talent.
In the third edition of the series, the discussion noted how a team should manage its roster by looking at the evaluating all the players and ridding the lowest scores not necessarily the lowest scored players. The difference is that teams should use their evaluations of both current and future performance. Trends and inflection points are important considerations, not just the overall scores themselves.
In this article, we will discuss the other end of the spectrum the highest scoring players.
Everyone wants stars. We see this in the FA discussions every year.
We have noted that teams cannot have too many stars on the roster. Otherwise one tends to skimp on quality backups. When the inevitable injury occurs, that leads to holes that other teams can exploit.
Yet we also say that we want to use FA to fill holes and then upgrade the talent through the draft. What happens if through judicious scouting, we draft well and still end up with too many stars?
That is a problem worth having. Bring it on, especially with the memory of not so good drafts. Hopefully, Garrett has changed our process, but it is too early to say. Yet even the best teams will not be successful with every pick.
Statistics help show the relationship
As OCC noted, every team will have some stars, some of the worst, and all kinds of players in between. Several issues come into play.
First, in statistics we use two measures. One is the mean or average [measure of central tendency] and dispersion away from the mean. Two teams can have the same average but have wildly different dispersions.
Let us show a simulated 3 player team roster
A – 5, 3, 1
B – 4, 3, 2
C – 3, 3, 3
Team A could have a third of the team who are stars , a third average  and a third of the team who are terrible 
Team B has 1/3 of the players who are above average , average  and below average .
Team C could have everyone who is average  – average score of 3.
All three teams would average 3, but their measures of dispersion would differ.
Secondly, we have noted that stars get an abnormal high level of pay compared to even above average players as the salaries do not go up in a straight line. So team A would be a lot more expensive. In reality, the teams might look like this
A = 5, 3, 1
B = 4, 4, 2
C = 4, 3, 3
Teams that don’t spend so much money on stars can have better quality across the board and much better depth. I called this the DAT effect.
Further, the lower scoring players might be the youngest backups who are still in the process of developing as described in the last article.
Now the issue is what to do with a roster that has too many stars already. We need to consider several factors.
We have discussed earlier that teams have to let players walk if they get too expensive, even stars. The major issue is that if the star is a FA, the time for the GM to decide has already passed.
Stars are fan favorites. Lose them at your peril. A GM will consider the fans in constructing a team, but should focus primarily on the talent. Additionally if you lose them by letting them walk, you get nothing in return.
So a team should consider trading stars before they lose them in FA. Without a salary cap, this is the Walker Effect. Dallas got a lot better with the returns they got from the trade than what Herschel Walker brought to the team.
With a salary cap, you can’t trade players too early based on how many years that are left on their contract. You have to eat the money on their contracts as their bonuses accelerate.
The key is to have a budget for each player and for each position for now and for the future. If a player is greatly exceeding their future expectations, then the market is ripe to trade them and reap a Walker effect even on a smaller scale.
That bonus acceleration should be minimized if the players are on their rookie contract. So consider trading a player who has already shown their potential to be a star. You lose a great player but reap the returns. The fans should not be as attached to a relatively new player as one on their second contracts with a team.
So another issue is what contract a player is on. The second contract for rookies will get a lot more expensive for budding stars. Yet until that happens, they are relative bargains on their rookie contracts. Trading them a year before their contract explodes is a good option.
Typically, the first contract is for 4-5 years, but the length of the second contract can vary. Second and third contracts denote players who are already established as stars. Again, the key is the same; trade them before they get to FA, but not too early. The year before their contract year sounds like the highest value.
Teams should manage who they put on their roster, but also have to manage who do not keep. On the lowest end, we wanted to lose the lowest scoring players but who have little upside.
Bill Parcells upgraded the talent on the roster significantly. Yet he could rightly be criticized for the way he trimmed talent. Once a player was “no longer in the future of the team” they were gone. We often got nothing. Even if we could have gotten a 6 or 7th round pick that was valuable. At the worst, we could those extra late picks to trade for future more valuable picks. We used a 7th round pick for Cook this year. I shudder to think of the season without that trade.
We also need to consider how to reap the most value of the best players. I call this the Walker effect. Every team needs stars, if only for the fans. Yet GMs should manage the roster for most value of talent overall not individuals.
A team with lots of above average players allows for better quality and depth throughout the team. I call this the DAT effect.