Optimizing Your Dynasty Roster with the Pareto Principle:

Applying Economic Principles to Dynasty Fantasy Football

The Law Firm Corollary

A friend of mine, who is an excellent fantasy football player, once made a mistake that has influenced him to this day. In week 1 of 2013, he was an Adrian Peterson dynasty share holder. Coming off a 2,000 yard season with the Vikings, Peterson’s value was at an all time high. At the same time, my friend was offered Bengals wide receiver A.J. Green – in his third year – and then Seahawks running back Marshawn Lynch. The only catch? My friend also would need to give up Bengals running back Benjarvis Green-Ellis, aka The Law Firm.

As hard as it may be to fathom now, Green-Ellis was too significant for my friend to give up, and the deal was spiked. Lynch went on to outscore Peterson, while Green put up the first of a long stretch of 1000 yard seasons (remember that Green?). Green-Ellis meanwhile finished with just over 700 yards and was out of the league a year later.

Since then, he and I have pledged to no longer allow insignificant pieces to blow up otherwise significant deals. I call this, ‘The Law Firm Corollary’.

Introducing the Pareto Principle

Our made up term actually has a basis in economic theory, it is known as the Pareto Principle. This is a guiding principle by which I evaluate my decisions in dynasty fantasy football. The Pareto Principle states that in a given setting, 80% of outputs are typically generated by 20% of inputs. In other words: the best workers in your company, players on your team, stocks in your portfolio, etc. matter a lot. The rest does not matter a whole lot.

In this article, I will outline how the Pareto Principle applies to fantasy and dynasty football in particular. In future articles, I will go over how you can apply the principle in startup drafts, rookie drafts, and top players to trade for or trade away for based on this frame of analysis.

Before we get into the good stuff it is time for some boring math:

From 2017-2019, I looked at the points above replacement (PAR) of each season long finish in points per game at each position. I defined PAR as the points scored per game higher than the lowest ranked starter at each position in a standard league (1 QB, 2 RB. 3 WR, 1 TE, 1 FLEX).

I then determined the share of points above replacement within the top percentile ranges at each position. I used 80^th, 60^th, 40^th percentiles for running backs and wide receivers. For quarterback and tight end 75^th and 50^th were used due to the smaller total sample of players. The top percentile grouping included six running backs, eight wide receivers, three quarterbacks and three tight-ends.

These results are not particularly surprising: The top 20-25% of starting assets at each position are responsible for an outsized proportion of PAR. This shows only 14-16 running backs and wide receivers combined will be responsible for half the total fantasy points above replacement in a given year at those positions.

When you expand the pool of players from only those who produce above replacement to all players rostered, the application of the Pareto Principle becomes very clear. For now, ignore the quarterback and tight end position. In a league with typical roster settings, there will be a total of 144 running backs and wide receivers rostered. If I take the highest scoring 29 (20%) from 2017-2019 their PAR share per game is 75%. 

To make this simple to understand: The best 29 running back/wide receivers in your fantasy league are responsible for 75% of the value. That leaves room for only two to three players on average per team who truly matter.

The Principle at Work in Dynasty

Most sane people think Fantasy Football is a game about points, or football, or players scoring points while playing football.

But, as a nerdy former economics student who comes to fantasy football without ever having played a down, I define it as a game about accruing the largest share of value in a given league. In redraft fantasy football, value is accrued in terms of fantasy points. In dynasty however, you operate in a duel market system where you are often simultaneously aiming to accrue a larger share of points in the given year, as well as a larger share of the total market value across the duration of the league.

The second component – adding more value to trade with and/or use in future years makes the application of the Pareto Principle more applicable beyond just adding up points and plotting them on a spreadsheet.

It does so in two big ways:

A Duel Market Environment Reduces the Number of Elite Assets

In 2020 the following players were included comfortably within the top percentile range in terms of points per game:

–   Myles Gaskin, Running Back, MIA

–   Chris Carson, Running Back, SEA

–   David Montgomery, Running Back, CHI

–   Adam Thielen, Wide Receiver, MIN

–   Will Fuller V, Wide Receiver, HOU

–   Julio Jones, Wide Receiver, ATL

All of these players – either due to age, future ambiguity, or market devaluation provided a high level of PAR but held far less market value. If you have ever had the pleasure of being offered Myles Gaskin and a 2021 third round pick for Philadelphia’s running back Miles Sanders being told you’re getting a “steal”, while rolling your eyes into your skull, you know exactly what I mean.

Consider this, even truly elite running back producers like Tennessee’s Derrick Henry or Green Bay’s Aaron Jones: these assets are universally respected. But, they are worth far more to contenders than to a rebuilding team. If you are trying to maximize your chances to win in a given year, while also optimizing the long term value of your roster, the pool of true difference makers in PAR who also have elite long term value and top end market value to you is very limited.

A Duel Market Environment Increases the Threshold of Production Required to Win

This one is slightly less intuitive. In one sense, because there are elite near term assets which are devalued in dynasty, it can actually make it easier to stack a roster in a given year with undervalued short-term only assets. Think about the guy or gal in your start-up who trades all their rookie picks to stack up veteran stars. They may have a better chance to win in 2021 but if everyone valued these players based on their 2021 outlook alone, owners wouldn’t be able to build this style of team so easily. 

Essentially, the manager doing this is trying to build the same beautiful house you are, but cutting out the hard work of building the proper foundation and just holding it together with duct tape, yarn and Julio Jones. 

My advice is focused on the foundation builders. 

I want your teams to win now, and the next year, and compete for decades (unless you’re in my league.)

So what do I mean by increasing the threshold of production to win?

In most dynasty leagues teams are on diverging timelines. Typically three or four teams in your league have no intention of competing in the given year and are storing most of their value in youth and future picks. If some teams are producing well below replacement, and some of your fellow competing teams are selling the farm for an all out push, the level of production you need from your team to win is even higher than in redraft when all teams are trying to win and nobody can deal away future picks. It’s a difficult task to win a dynasty league while trying to keep your future intact. Here’s how you do it:

Applying our Principles

So you need elite players to win leagues. 

While the existence of this principle is rather obvious, what’s important are the ramifications. Understanding the ramifications of the Pareto Principle model of roster building should change the way you prioritize assets and build your rosters for sustained success.

Sorting Your Assets

The first thing to internalize is most assets on your team are not directly relevant to your success. As per the Pareto Principle, the vast majority of the value of your roster is a function of only a small number of your assets. Only about two-three assets per team on average (more if you’re playing superflex) will make the level of outsized impact necessary to deem them ‘irreplaceable.’

If you want to win a dynasty league you probably need five to six of the above . If a player isn’t one of those five to six in a given year, he only serves a tangible purpose to your roster value if he is able to net you one of those top five or six assets either via trade or via development.

Asset Fluidity

The biggest key to maximizing your roster value in dynasty is understanding asset fluidity. All of the data discussed to this point is based upon looking back rather than projecting forward. While we know the top percentiles of assets will provide outsized value, we do not necessarily know who they will be. While following the Pareto Principle reactively may suggest trading boatloads of picks and depth for Carolina’s running back Christian McCaffery is always a good idea, that would be learning the wrong lesson.

Among the non-elite assets on your roster you need to evaluate critically the path it would take to convert them to elite assets. The way you double up your opponents in terms of irreplaceable assets isn’t simply gutting your entire roster depth. That’s not a sustainable plan. If you have scarce resources, you need to find ways for your existing assets to accrue value themselves. That typically happens either via a player improving or a draft pick hitting, so it is important to keep the non-elite assets fluid so they can accrue value.

For example, Jets wide receiver Denzel Mims is likely to accrue value simply by developing. Moving Mims for a much proven player(s) in Rams wide receiver Robert Woods is a way to secure the value Mims currently holds but is unlikely to accrue additional value for your team. Playing by the Pareto principle, I would prefer to hold Mims.

If I rostered Woods, I would look to either consolidate him with other pieces on my roster to make a trade for an elite asset, or convert him into an asset of equal value but a wider range of outcomes and higher ceiling probability such as a mid-late first round pick.

Consolidation versus Conversion

So, you have a productive veteran on your team with limited upside and you want to make a move… how do you know when to consolidate verses when to convert? This requires self evaluation. Is your team above the median value of the league? If so, look to consolidate your value among an elite starting lineup. If your team is average or below, you don’t want to be consolidating. Star players have the highest market value but the least fluidity. 

Consolidating all your value into non-fluid assets may sentence you to eternal mediocrity if you don’t have enough value on your team to start with. Instead, break apart your veteran assets into assets of equal or greater value with the widest range of outcomes – typically picks and young players. This is the best chance for your team to accrue more value until you can eventually begin to consolidate.

Assume You Will Hit. Your Misses Don’t Matter

Not all picks and players are created equal. One of the biggest mistakes the dynasty community makes is associating age / draft year with upside as a 1:1 correlation. A 22 year old running back with a 99th percentile speed score and receiving ability but who had 33 carries in college has more upside than a plodder with better production but middling athleticism. 

Buffalo running back Zack Moss was probably more likely to be a ‘hit’ than Washington’s Antonio Gibson this year, we have to take into context what a ‘hit’ means for each player. Is Zack Moss’ ceiling outcome ever top 5 dynasty running back? In a future article I will expand on how to forecast the odds of this but likelihood is not high. For Gibson however, while there were red flags in his prospect profile, his ceiling outcome – based on his athleticism and receiving ability – was and is absurdly high.

Bringing this back to economics, a player like Gibson had a high degree of asymmetrical upside: any asset that is inefficiently valued due to the gap between their market value and ceiling being far greater than the gap between then market value and floor. For example: if you were given a one in five shot of $100 vs. a guarantee of $5,  the one in five would have asymmetrical upside. Even though four out of five times you get less money, taking the risk is still the choice with a better overall expected value.

When you remember a vast majority of players on your dynasty roster will not ultimately determine whether you win or lose, the median outcome of a player becomes far less important than their ceiling outcome and you should always be chasing upside.

The best example of asymmetrical upside paying off this year in dynasty was Philadelphia quarterback Jalen Hurts. Drafted near the end of the 3^rd round of 2020 SUPERFLEX rookie drafts, the ceiling outcome of Hurts – a highly mobile and productive quarterback prospect – was of far more value than any of the players selected around him, and far greater than the market value of his pick slot.

In a future article I will look in depth at which archetypes and specific prospects in the 2021 draft present the most asymmetrical upside right now.

Conclusion: New Rules

Implementing the Pareto Principle into your dynasty management means adopting the following new rules into your heart:

(i) Accept that only a small number of players will be responsible for you winning or losing the championship in a given year.

(ii) Understand that the highest market value players have the highest likelihood of producing at an elite level, but the least ability to accrue value for your roster.

(iii) Evaluate your rosters to determine which players provide elite value now, which could develop into elite assets, and which primarily function as stores of value.

(iv) Critically analyze the overall value of your roster to determine whether you should consolidate your assets to minimize your range of outcomes or expand it.

(v) Consider the asymmetrical upside of low market value players while knowing that your big hits are far more impactful than your misses.

I hope this serves as a useful overview and I look forward to expanding further in the future.

You can find Jakob Sanderson on twitter at FF_RTDB!