The concept of consistency and its role in fantasy football is multi-faceted. A fantasy manager’s overall drafting philosophy will often dictate how much they value consistency in their player selections. Where in the draft players are being selected can play a factor as well.
Primary factors in the decision-making process are league format, league size, scoring, required starters, and bench size. Regardless of the motivation or reason for considering player consistency in fantasy football, it is discussed frequently.
The terms “boom/bust”, “high floor”, “low ceiling”, and “reliable” get tossed around during fantasy discourse, and they are all different ways of describing consistency. Sometimes consistency is purely subjective, often influenced by one’s most recent experience of rostering a certain player, which may provide a skewed perspective on that player’s actual consistency. On the other hand, consistency can be objective, evaluated across a player’s career, across an entire position group within a season, or multiple seasons.
Our analysis will have multiple parts, first looking at the concept of consistency and how we can measure it, and how we can incorporate both subjective and objective elements into our analysis. Then we will examine how the difference between the subjective and objective varies by position, looking at a specific example for each position.
Our future articles will provide more detailed breakdowns by position, diving more into specific players and examples of how well consistency translates into fantasy success.
What is Consistency?
As noted above, there are many ways to conceptualize and describe consistency. The terms used above, like “boom/bust,” evoke certain thoughts or reactions about players fitting this profile. Fantasy managers will sometimes associate consistency-based terminology like this with certain players (a bit of a throwback reference, but we’re looking at you, Tyler Lockett…).
Whether subjective or objective, fantasy managers use the concept of consistency to guide their fantasy drafts and roster construction. But how can we objectively measure consistency? With a little bit of statistics and simple division, we can create a Consistency Score (CS).
The first step is calculating the standard deviation across a player’s weekly performance in a given season. Standard deviation is a measure of how spread out the values in a dataset are from the mean (average). The lower the standard deviation, the closer all of the values are to the mean, and therefore, the more consistent. Here’s a simple example:
Player A scored a total of 73 points over eight weeks, which is 9.125 points per game (PPG). The standard deviation across those weeks was about three points. Now we can take these two pieces of information and create a consistency score, which in statistics is called the Coefficient of Variation (CV). It divides the standard deviation by the average. We will continue to refer to this as our consistency score (CS), though.
Consistency Score for Player A: 3.06 / 9.125 = .34
Great, 0.34. That’s good, right? Well, maybe, but we don’t know yet. The score doesn’t tell you anything by itself – the only thing we know is that the smaller this number is, the better. So we have to do this for ALL players in a season across ALL positions. Since we don’t have time to do that today (and I have done all the work for you already), I’ll provide one more example of Player B, who scored the same number of points over the same number of games:
Consistency Score for Player B: .599/9.125 = .07
So was Player A’s consistency score of .34 good? Not in comparison to Player B’s score of .07. While Player B’s game-to-game consistency is pretty extreme and not entirely realistic, I used these extreme examples to illustrate the point of how consistency scores work, and how they can be used to compare two players with the same raw fantasy output that got there in very different ways.
So now, using some fancy formulas within our data model, we can create consistency scores for all players for every season across each position. Next, we can rank these consistency scores – this allows us to rank players within their position and season relative only to players of that position in that same season.
As we look at the entire dataset, we can evaluate the top XX range of Consistency Score Rankings and use that range of rankings to evaluate players across their position group within every season. This will be used more as we dive deeper into the positions in future articles.
Though we have established how to measure consistency, our work is not done. Who was more valuable for your team, Player A or Player B? Player B was certainly more consistent, but did that help you in your fantasy matchups?
Consistency in and of itself tells just part of the story, not the whole story. The missing part is how consistency factors into players making your lineup and winning or losing you weeks.
Establishing Who Is a Starter and How Much They Score
The data we will be discussing is somewhat league format dependent – meaning, some of the terms we will refer to, like Starter Games and % Starter Games, Flex Games and % of Flex Games, are calculated based on the FFPC league format, scoring, and lineup requirements. FFPC leagues have 12 teams, and starting lineups use 1 quarterback, 2 running backs, 2 wide receivers, 1 tight end, and 2 flex positions (running back/wide receiver/tight end) and Points Per Reception (PPR) scoring. Additionally, tight ends have premium scoring, earning 1.5 PPR. To sum this up, a starter is:
Top 12 Quarterback Weekly
Top 24 Running Back Weekly
Top 24 Wide Receiver Weekly
Top 12 Tight End Weekly
Flex-worthy spots are calculated based on the NEXT BEST 24 skill position scorers that week – they are mutually exclusive from starter categories, so you are either a starter or a flex, but never both.
As long as your league format is similar (single quarterback, starts 2-3 running backs, starts 2-3 wide receivers, starts 1 tight end, starts 1-2 flex spots, PPR), this analysis should hold pretty true for you as well. Best-ball and Superflex formats will be a different story.
We are almost done with our objective analysis and will take a look at the subjective elements shortly. The subjective part of the conversation includes creating points per game (PPG) thresholds for each position. This is where we start to unpack how PPG is evaluated based on season-long production compared to how it looks being evaluated based on weekly usability. We will use the quarterback as the example:
Going back to 2010, if you take the average PPG for quarterbacks who finished the season as top-12 quarterbacks based on their cumulative points, those players averaged 22.12 PPG.
However, if you look at the weekly data, evaluating only the quarterbacks who finished as a top-12 quarterback during the week, the PPG for those players is 26.93.
The large gap of nearly five PPG represents the significant difference between season-long rankings and weekly usability. While it is great to say that your quarterback finished as QB10 on the year, this doesn’t tell us much in terms of what they did for you every week.
We can assume they were likely pretty consistent, but QB10 production probably isn’t league-winning material. It would be great to know how often that quarterback “caused” you to lose your week, contributed to help win your week, or gave you that week-winning outburst that single-handedly took your opponent out.
Fantasy managers have differing views on what this means, especially between the different positions, which will be evident as we dive into the subjective part of our analysis. These differences reveal a lot about the psychology of fantasy football.
The Positional Psychology Trap: Why Your Floor Expectations Are Capping Your Ceiling
Evaluating consistency is a multi-faceted process, with a thorough approach incorporating elements of both objective and subjective components. Up to this point, we’ve examined the objective parts of consistency, rooted in widely accepted statistical analysis. The interesting part comes when we start looking at it from a more subjective point of view, adding how fantasy managers view positional point production and what defines a successful week.
Like a panicked participant on “Who Wants to Be a Millionaire?”, I used one of my lifelines and asked the audience. I wanted feedback from fantasy managers about what level of point output from each position they would use as thresholds to define three categories of weekly success (or failure): “Lost Your Week”, “Floor Starter”, and “Won Your Week.” How many points would you need from each position before you would label that performance as having contributed to you losing your week, being worthy of starting, or having production that helped win you your week? The results were quite interesting.
Since there is a point gap between “Lost” and “Floor” thresholds, this is a “Neutral” area where a player neither lost you your week nor achieved floor starter status. These are infrequent, but still worth mentioning.
Overall, fantasy managers are much more lenient in their expectations for weekly point production compared to how often those players actually should have made it into starting lineups. In three out of the four positions, the number of times players hit at least a subjective “Floor Starter” level point production exceeded the number of times those players found themselves inside the top-12 (for quarterback and tight end) or top-24 (for running back and wide receiver) each week.
Running back was the one exception as fantasy managers are much more dialed into what level of production justifies weekly starter status.
One last note as we move through the positional breakdown – when we are looking at player statistics, a single season is a very small sample size with a maximum of 17 data points. Two seasons are certainly better, but one may have been shortened due to injury. Therefore, we will use three seasons as our sample when making these evaluations, understanding that the 2024 rookies will only have a single season in their evaluation.
Going forward, I will use “Lost”, “Floor”, or “Won” to refer to the various thresholds fantasy managers chose when setting their subjective expectations by position.
Positional Breakdown
Quarterback
Starting with quarterbacks, fantasy managers settled on 15 points being the “Lost” threshold, 18 points being the “Floor” threshold, and 25 points being the “Won” threshold. Applying these thresholds to the weekly performance data over the last three years, it was apparent that fantasy managers didn’t expect as much from their quarterback as they should.
The combined subjective “Floor” and “Won” week percentages exceeded the actual starter week percentage 12.71% of the time. In other words, quarterbacks were startable (top-12) in nearly 13% fewer games than fantasy managers set as their “Floor” expectation. We will use Jalen Hurts as an example to illustrate what this looks like, who had a “Floor + Won” to top-12 starter differential of 10.64%:
Hurts played 47 games over the past three seasons
* Was only a top-12 starter in 9 of 14 subjective “Floor Starter” games = 64.30%
So to summarize the info above, 29.79% of Hurts’s games hit fantasy managers’ definition of a “Floor” week, and 55.32% hit week-winning status, totaling 85.11% of his games. Conversely, the data shows that Hurts was only a top-12 quarterback in 74.47% of games, providing the 10.64% differential between expectation and reality.
Granted, the “Floor” expectation for quarterbacks includes the entire pool of players, so you may expect more from a player like Hurts compared to everyone else. Regardless, only 12 quarterbacks put up start-worthy performances each week, and you want one of those.
If you enter your draft being content with a player giving you floor-level production, you are leaving quite a few points on the table. The data tells us that for quarterbacks, a 50th percentile outcome for a QB1 on any given week is 25.82 points. If your floor expectation is 18 points, you are short seven to eight points compared to what a mid-level QB1 outcome normally is.
Even a 10th percentile outcome is closer to 21 points, so in reality, this is what the floor expectation should be for quarterbacks. An 18-point week is equivalent to a 60th percentile outcome for QB2s, so somewhere between QB16-18 weekly.
Having consulted with the late great Sigmund Freud, we have dubbed this the “Quarterback Premium Delusion”. Fantasy managers appear to have dramatically lowered expectations for quarterback performance, setting their “disappointment threshold” at 15 points when historical top-12 QBs rarely score that low. This suggests managers have internalized the idea that QBs are inherently valuable regardless of actual production – a classic case of positional bias overriding performance evaluation.
Wide Receiver
Fantasy managers settled on a “Lost” production from wide receivers at 10 points, “Floor” at 12 points, and 20 points being the “Won” threshold. Like with quarterbacks, fantasy managers underestimate what to expect from their starting wide receiver spots. The combined Floor and Won percentage weeks exceeded the starting week percentage by 13.20% for all wide receivers over the last three years.
So again, even when they are hitting the thresholds fantasy managers have deemed reasonable, they may not hit top-24 weekly-level production. We’ll examine Amon-Ra St. Brown with a Floor/Won to Starter differential of 16.33%:
St. Brown played in 49 games over the past three seasons
* Was not a top-24 wide receiver in 8 of 21 “Floor Starter” Games = 38.09%
Recapping St. Brown, he achieved “Floor” or “Won” status in 81.63% of his games, yet was only a top-24 wide receiver in 65.31% of games, resulting in a 16.32-16.33% differential (rounding factor). He failed to be a top-24 wide receiver in eight of the 21 games that fantasy managers subjectively labeled him a starter.
The data shows that the median (50th percentile) WR2 performance (ranks 13–24) is nearly 17 points. Even the 10th percentile WR2 score is 14.10 points, meaning that 90% of WR2-level starters score above the 12-point Floor Starter threshold in a given week. This suggests that a “Lost” week threshold should be closer to 12-14 points, a “Floor” threshold 16-17 points, and based on the data, a “Won” type performance should be closer to 24 points.
Freud and I went back and forth on this one and settled on “Wide Receiver Volatility Acceptance”. The 10-point minimum expectations before considering a wide receiver having lost your week indicate managers have become conditioned to accept boom-or-bust performance as normal. This likely stems from the “home run” mentality around wide receiver play – managers remember the big weeks and mentally discount the duds, leading to unrealistic low-floor expectations.
Tight End
Based on learning how fantasy managers view tight ends, the responses had all the enthusiasm of a root canal consultation. Tight ends are the most underestimated position among the big four.
This one deserves a bit of nuance, as the most popular scoring format is points per reception (PPR), and as I mentioned earlier, this data comes from TEP leagues, which award 1.5 points per reception for tight ends. The top-36 tight ends' PPG average is roughly 18-20% higher in TEP leagues than in standard PPR leagues, so I will provide context on these thresholds as we uncover the numbers.
Fantasy managers set the "Lost" threshold for a tight end at eight points, the "Floor" threshold at nine points, and the "Won" threshold at 15 points. As if it weren’t any more obvious that tight ends are the most hated position in any sport, the fact that tight end Starter rates exceeded the "Floor and Won" week thresholds by 17.43% just reinforces the hate.
Nearly one out of five times a tight end hit the subjective "Floor" outcome, they still did not achieve a top-12 performance that week. Next up is Evan Engram, who had a differential of 13.96% of "Floor and Won" games exceeding Starter games:
Engram played 43 games over the past three seasons
* Was not a top-12 tight end in 5 of 14 “Floor Starter” games = 35.71%
* Also was not a top-12 tight end in one of the games where he hit “Won Week” status
Recapping Engram’s three-year sample, fantasy managers' threshold for Engram being worthy to see their lineup was 69.77% of games, though he was only a top-12 tight end in 55.81% of games. This nearly 14% differential, while the largest of our positional examples thus far, is still below the 17.43% average for tight ends.
The most alarming part of this was in the 14 “Floor” games, Engram was not a top-12 tight end in five of those. Looking at both TEP and PPR leagues, Engram failed to reach top-12 status in either format during one of the weeks where he hit the 15-point “Won” threshold (2024 week 8).
It has never been more clear that fantasy managers hate tight ends and wish they didn’t exist, and would prefer to just ignore the impact of their performance on weekly outcomes.
Freud and I agreed immediately on this, as we coined it “Tight End Learned Helplessness”. The eight-point floor for tight ends is particularly telling. Fantasy managers have essentially given up expecting consistent tight end production, setting thresholds so low they're meaningless. This reflects the widespread "tight end wasteland" narrative that has become self-fulfilling. Managers expect so little that they don't properly evaluate actual performance.
Running Back
We’ve saved the best (at least most different) position for last. It may come as no surprise that fantasy managers were the most dialed in on what level of performance made running backs startable weekly. Admittedly, many draft strategies and terminology are named after the position: “Zero-RB”, “Hero-RB”, and “RB Dead Zone”. With as much attention as is paid to the position, we better know what the good ones look like.
Fantasy managers set the same thresholds for running backs and wide receivers at 10, 12, and 20 for "Lost", "Floor", and "Won" week thresholds, respectively. With running back, they were on point, as running backs hit starter-level production nearly 4.5% more often than the combined "Floor" and "Won" thresholds. This is the opposite of the previous three positions we looked where fantasy managers were too lenient.
While fantasy managers may overestimate the number of points needed for startable running backs, the 4.5% delta is much closer to reality than the other positions. Alvin Kamara will serve as our final example, whose "Floor and Won" to starter differential came in right around the sample average at 4.76%:
Kamara played 42 games over the past three years
* Was a top-24 starter in all but one of the “Floor Starter” weeks (95%), and was a top-24 starter in three games where he hit “Lost” or “Neutral” week production
The highlight for Kamara is that he had three startable games below the subjective “Floor” starter point threshold, which represented nearly 10% of his total subjective start-worthy games, and over 7% of his total games played.
The primary takeaway from the running back position is that fantasy managers have a pretty good feel overall, as a 4.5% variance between their subjective production thresholds and true weekly starter status is much closer than any other position.
Sig and I took a bit longer to come up with this one, as it varied from the other positions, but we determined this was the classic case of “Running Back Realism”. The accurate running back threshold calibration suggests managers understand this position better, likely because running back performance is more tied to obvious metrics (carries, touchdowns) that are easier to evaluate objectively. There’s less positional mythology surrounding running backs.
Why Your Floor is Capping Your Ceiling
So, how are fantasy managers’ low expectations capping their ceiling? To summarize what was discussed above, fantasy managers are setting their expectations way too low for “Floor” level production for quarterbacks, wide receivers, and tight ends. The number of points resulting in starting-level production is higher than what fantasy managers have settled on in their minds. They are dialed in on running back, and while they may expect slightly more than what is actually needed, at least it's a step in the right direction.
As we look at how the concept of consistency factors into the idea of lowered expectations, the results may be detrimental on draft day as well as throughout the season, as you make your weekly decisions. Based on what we just learned, let’s look back at Player A and Player B and examine how their respective consistencies translated into weekly outcomes (subjectively, of course, since we cannot compare these to real-life weekly scoring since they are fictional players). For this example, we will assume these fictional players are wide receivers.
By adding a new row to our existing table with the subjective week status, we see that Player A, while less consistent, hit “Floor” level twice, while Player B, while a lot more consistent, failed to achieve a “Floor” level performance once. When comparing the individual weekly output between the two players, the largest point differential is in Week 3, where Player A outscored Player B by six points. The largest differential in which Player B outscored Player A in any given week was three points.
So, as a fantasy manager, choosing between these two players, your decision will be greatly influenced by how you want to run your team. Are you more risk-averse or risk-seeking? Do you choose the one with a guaranteed floor, or the one with a higher ceiling? Whether it is on draft day or when locking in your weekly lineup, that decision could help determine your success or failure.
Therapy Session Recap: Rewriting Your Fantasy Mindset
Now that our session on the couch has concluded, don’t you feel a lot better? Yeah, me neither. Better understanding how we, as a fantasy community, tend to underestimate or overestimate starter-level production for each position may help us find the answers we need when dealing with our fantasy anxiety and making better decisions than our league mates.
Of course, the trick is to know ahead of time whether a player will be the consistent floor-type or the inconsistent boom/bust-type. While that type of analysis is far from perfect, our future articles will aim to help identify player consistency tendencies, allowing you to better categorize players and help you build your team based on your desired level of risk tolerance.
If that interests you, our series will continue next time as we take a look at the two positions that typically take up the fewest spots on our starting lineups, the quarterback and the tight end.