In my previous article, we began our conversation about consistency in fantasy football to explore its importance. We defined consistency, using both objective and subjective elements to evaluate how consistent players are, and uncovered how fantasy managers might be leaving points on the table with their lowered expectations of what constitutes a startable game.
Today, we take a deeper dive into the concept of consistency positionally, looking at quarterbacks and tight ends. With this, we can better address the question of how important consistency is positionally and how fantasy managers should incorporate the concept into their draft strategy.
Methodology Review
In our previous article, we used three years of data to examine individual player consistency. As we break down position groups, there is a benefit in looking at a larger season sample when examining broad-level data like points per game (PPG). Similarly, evaluating players in an individual season can help provide beneficial context between seasonal fantasy results and consistency metrics.
Fantasy managers often remember players in a very subjective way related to their experience rostering those players. Sometimes, those experiences can be skewed by one or two data points (games) that overshadow how the entire season went. Some historically memorable fantasy seasons don't correlate with consistent performance.
We used the Coefficient of Variation to establish a consistency score (CS) for each player within each season, which we were able to rank (CS ranking) by position and by season. We also looked at a more subjective evaluation method, examining how often players hit point thresholds that fantasy managers set as having “lost” their week, achieved floor-level starter status, or produced a week-winning fantasy point output.
Finally, you may remember when we compared the percentage of times players met these subjective thresholds to how often they actually achieved starting-level status in our league scoring format (top-12 for quarterbacks and tight ends, top-24 for wide receivers and running backs). For most positions, there was a large discrepancy between these numbers, with quarterbacks and tight ends failing to reach starting-level production roughly 13% and 17% of the time, respectively, when they hit fantasy managers’ subjective “Floor” and “Won” week thresholds.
With that, I wanted to add to our analysis a score/metric that incorporated this data. I created a measure that weighted players' consistency score (30% weight), percentage of weeks achieving starter level production (40% weight), and percentage of weeks achieving the subjective “Floor” level starter production (30% weight). The difference with this score compared to the standard consistency score is that the higher the number, the better.
Quarterback
The relevance of consistency varies by position, which will become clearer as we dig into the numbers. Let’s start with the field generals of the game. As laid out in my previous article, we are using the FFPC format, which is a 12-team league that starts one quarterback. In the data below, quarterbacks must have played and started at least nine games during the season.
The above graphic highlights details about roughly 180 quarterbacks with the highest CS rankings since 2010. Quarterbacks earning a top-12 CS ranking over the past 15 years averaged 21.62ppg, are drafted at QB14 on average (143 ADP), and have startable weeks just 50% of the time. Yes, you read that right.
Looking specifically at the top-30 most consistent quarterbacks (average CS ranking of 1.6), the picture becomes even more concerning. Not only do they maintain that same ~50% startable rate, but they actually deliver week-winning production less often than the broader sample, just 28.10% compared to 30.55% for all 180 quarterbacks.
While they do lose you fewer weeks than the general population when using our subjective “lost” week thresholds (11.76% vs 17.72%), they're still costing you nearly 12% of your matchups while failing to provide the ceiling games needed to win weeks.
Examining the names on this list, you see an interesting mix of some of the most highly regarded fantasy quarterback seasons in terms of PPG (2019 Lamar Jackson, 2011 Aaron Rodgers), as well as some very forgettable seasons (2018 Alex Smith, 2024 Patrick Mahomes).
Re-ranking this list in descending order by Starter Games, the new top-30 group averaged 26.57ppg, a 79.30% starter rate, but an average CS ranking of 5.6. On average, these players are drafted in round eight with a positional ADP of QB8.
These quarterbacks are hitting week-winning ceiling games nearly 60% of the time, while only sticking you with disaster week-losing games 6.63% of the time, ~11% lower than the overall sample.
Finally, when comparing this list to the first one, you will notice there are only eight quarterback seasons in common (27%).
This reveals that quarterback 'consistency' in fantasy football is often consistency around mediocrity rather than reliability. While the most consistent quarterbacks lose you fewer weeks than the general population (as we showed above), the most startable quarterbacks lose you even fewer weeks while providing far more weekly value. The consistent quarterbacks aren't giving you a safe floor, they're giving you predictably average production that still results in “lost” weeks nearly 12% of the time.
To further illustrate the difference between a high consistency score and start-worthiness, we will take 2014 Andrew Luck, who had a CS ranking 11th of that year, yet was startable in 14 out of 16 games he played. In comparison, we will use Alex Smith in that same year, who had a CS ranking of 3rd but was only startable in four games.
Looking at Luck’s game log, he was stellar the first 14 games of the year, averaging 28.73ppg. It was the final two weeks that killed his CS ranking; if you run the CS on just the first 14 games, he comes in at .21 (with a standard deviation of 6.13), which would have been first that year.
But you can’t just remove games when it’s convenient, or you would have to do that for everyone (which means you would have to go through and flag every case for every player over the last 15 years that you think should be removed).
Now contrast that with Smith’s game log; he was very consistent, averaging 17.31ppg, with a standard deviation of 4.95. Despite lower PPG, this much lower standard deviation produces a better consistency score.
So, what are the implications for your draft strategy? If you want a weekly starter, you may have to pay up earlier than the QB14 range, where the average highly consistent quarterbacks are drafted.
The only way to reliably lock in a startable quarterback or someone putting up 26ppg+, you should be looking at the top 7-8 quarterbacks in rounds 5-8. While they may not win a consistency contest, they will win you weeks with ceiling games (and their floor will often be at or higher than the ceiling of your Steady Eddies).
The last thing we can look at to help us capture how consistency over a larger sampling of games translates into weekly fantasy success is evaluating the three-year sampling of weighted consistency scores.
If you recall, my weighted consistency score incorporated how often players were top-12 during the week as well as how often they met the subjective “Floor” level starting threshold set by fantasy managers. The value of this score is that it aggregates across all of the selected seasons, giving a single weighted score rather than one for each season.
This final graphic for quarterbacks reinforces that traditional consistency metrics poorly predict fantasy value. While three of the top four most consistent quarterbacks over the past three years (Jalen Hurts, Josh Allen, and Lamar Jackson) do deliver among the highest start-worthy rates, they're all mobile quarterbacks, suggesting their consistency comes from rushing floor production rather than traditional passing consistency.
Even with these three elite performers included, the entire group of 30 most consistent quarterbacks averages just 21.01ppg with a sub-50% starting rate and a significantly higher week-losing rate of nearly 22%. This shows that even when weighted consistency does identify some valuable players, the overall group still underperforms compared to quarterbacks selected purely for startability.
One last observation when comparing this list to the previous one, which was broken down into individual seasons. At an individual season level, it was not uncommon to see quarterbacks achieve start-worthy status 80%+ of the time. However, using just the last three seasons and looking at the aggregate data, no quarterback exceeded 74.47% start-worthy games.
In fact, this same list, when sorted by start-worthy game rate and evaluating the top 30 players, the average starting rate is still only 51.19%, and only 13 quarterbacks in the list were startable 50% or more of their games. This makes the elites at the position like Hurts, Allen, Jackson, maybe Jayden Daniels, and Joe Burrow the clear exceptions. Beyond them, it is a crapshoot.
Of the top-10 quarterbacks based on percentage of start-worthy games, six of them offer significant rushing upside (Hurts, Allen, Jackson, Daniels, Justin Fields, Kyler Murray). Three of the remaining four (Burrow, Brock Purdy, Mahomes) are still mobile quarterbacks who have scrambling upside. Dak Prescott, while once known as a mobile quarterback, has lost much of that upside due to recent injuries and is nearing 32 years of age.
All of this is to say that quarterback startability over the last three years is highly tied to mobile quarterbacks and less to consistency. Mobile quarterbacks will often give you some floor production with their legs, which is worth more than equivalent production through the air. If they are a competent passer as well, then you likely have a combination capable of giving you a strong weekly starting rate.
Later-round (high consistency) quarterbacks aren’t winning that many weeks because their consistency is rooted in floor outcomes, with ceilings being very close to their floors. You can cobble together starting weeks by streaming these guys and playing matchups; this is fundamentally what the late-round quarterback strategy is all about. But these players, in a vacuum, won’t offer the ceiling you need to beat the competition, and their consistency does not equate to you winning your weeks consistently.
2024 Player Comparison
Using FantasyMojo average draft position (ADP) data for FFPC FantasyPros Championship leagues, we can look at two players drafted in similar spots who offered two very different consistency profiles and very different results for your team.
Patrick Mahomes was drafted as QB2 at an ADP of 51, and Jalen Hurts was drafted as QB3 just one spot later at 52 overall. Mahomes had a CS ranking of 2nd with six start-worthy games, whereas Hurts had a consistency ranking of 11th with 10 start-worthy games.
It’s unclear what drove these ADPs, as Mahomes ranked 4th in consistency in 2023 with seven start-worthy games, and Hurts ranked 9th with 13 start-worthy games. So, despite the 2023 results, fantasy managers were banking on a bounce-back from Mahomes while possibly not fully factoring in the ceiling of Hurts.
Looking more subjectively, Mahomes had one game in 2023 where he “Lost” you your week (15 points or fewer), whereas Hurts had two games. While both had minimal disaster games, Hurts's came at the worst possible time during week 17 (fantasy playoffs for many managers).
Considering the stark difference in these quarterbacks' 2023 results, one would think Hurts would have been drafted higher than Mahomes in 2024. How close these ADPs were coming off of two very different seasons in terms of startability could be related to the psychological impact of Hurts’s week 17 dud, Mahomes’s history of solid production, or his perception of being stable (consistent).
Either way, if you took the consistent guy here, you lost out on four additional startable games. This real-world example demonstrates how consistency metrics can mislead draft decisions. Fantasy managers who chased stability over startability paid a significant opportunity cost in 2024.
Tight End
While quarterbacks showed little correlation between consistency and fantasy value, tight ends present a somewhat different story, though the relationship remains weaker than many fantasy managers assume. Despite the perception of tight ends evolving over the years, their overall role within team offenses has remained pretty stable.
League-wide since 2010, tight ends have averaged just under a 21% target share, with last year representing the largest target share for the position during that span at 22.47%. They have consistently seen north of 21% target share over the past several years.
Many fantasy analysts argue that the tight end is not a real position, and for fantasy purposes, we should do away with the position altogether and lump them in with wide receivers. This argument is rooted in how tight end usage has changed over the years, and how different archetypes of players are being deployed as “move” tight ends, accounting for the recent increase in target share for the position.
While player archetypes are an important factor that deserves a larger discussion, today, we are focused on how consistency translates to fantasy success. And like quarterbacks, since there is only one starting tight end spot on our roster, we will examine the top 12 tight ends in the CS ranking.
Unlike quarterbacks, where we found virtually no correlation between consistency and startability, tight ends show a slightly stronger relationship, though it remains far from reliable.
Tight ends with a top-12 CS ranking since 2010 were startable in 52.68% of games (slightly higher than quarterback) with a flex rate of 10.69% and overall, should see your starting lineup 63.37% of the time. On average, this group of tight ends is taken at the beginning of round nine at TE13 and averages just under 14ppg. So the for tight end apologists, we’re off to a rough start if you are looking at consistency to predict startability.
Selecting the top 30 players shown from this list, it starts looking a little better. This group averages just over 15.6ppg with an average CS ranking of 1.5. They are taken mid-round seven on average as TE10. They are startable 64.33% of the time and are flex-worthy 8.80% of the time, making them worthy of a weekly roster spot 73.14% of the time. This is nearly 10% more than the overall group of top-12 CS ranked tight ends.
Sorting by startable games and selecting the 30 shown in the list, this group of tight ends has a starting rate of 79.32%, a flex rate of 5.06%, and a starting lineup rate of 84.39%. To get a tight end in this range, you will need to draft them by the end of round five at TE7 on average. This group averages just over 18.8ppg (an increase of ~5ppg over the total pool) and has an average CS ranking of 4.1.
Going back to the quarterback, if you recall, the average CS ranking of the top startable quarterbacks was 5.6. What this tells us is that tight ends with a higher start-worthy number of games tend to have a higher level of consistency than quarterbacks using the same evaluation criteria.
Top start-worthy quarterbacks had an average CS ranking of 5.6, which is in the top 46.67% of the most consistent quarterbacks (top-12 ranked), whereas the best start-worthy tight ends, at an average CS ranking of 4.1, are found within the top 34.16% of most consistent tight ends.
Also noteworthy with this group of tight ends is that they only return week-losing production in 11% of their games, compared to over 24% of the total tight end sample. This is a 54.47% reduction in the percentage of disaster games, though not quite as impressive as the 62.58% improvement with the same group of quarterbacks.
As we did with quarterback, let’s sort this list by weighted consistency score to see how the most consistent tight ends over the past three years translate into start-worthy weeks and how they compare to quarterback.
The above graphic for tight ends uncovers the volatility at the position over numerous seasons, especially for those considered more consistent. Your top two tight ends with the highest weighted consistency score (Gerald Everett and Dallas Goedert) were only startable in 43.33% and 50% of their games, respectively.
After that, some unsurprising names show up like Travis Kelce and Sam LaPorta, as well as some more surprising names like Zach Ertz. These top-30 tight ends in terms of weighted consistency over the past three years only have a start-worthy game percentage of 46.07% and 9.87% flex rate, bringing their overall start-worthy rate to 55.94%.
A name of note on this list is George Kittle, who, despite having one of the highest starter game percentages at 65.22%, was middle of the pack in terms of weighted consistency at 19th. This is due to his raw consistency score for two of the last three seasons being ranked 11th and 14th, though last year he was ranked 2nd. His fantasy point totals game to game, while inconsistent, still resulted in the third-highest start-worthy game percentage, though he was only flex-worthy in one additional game.
Further evaluating this list, sorting by the most starter games, the names should not be surprising. Kelce, Kittle, Mark Andrews, David Njoku, and Evan Engram round out the top five of most startable games over the past three seasons, with Trey McBride and Brock Bowers making the list when sorting by percentage of startable games.
In my article from back in May, where I discussed rookie tight ends, we discovered that the top tight ends in fantasy are typically top two target earners on their team. As we discussed the importance of quarterback mobility earlier, earning targets will remain a primary factor for tight ends to have weekly startability. Injuries to other players on their teams throughout the season can offer that contingent weekly opportunity if you find yourself streaming the position and looking for a spot-start.
Much like quarterback, weekly consistency at tight end is not the end-all be-all, though it is slightly more important. While historically the most consistent tight ends return start-worthy performances about 53% of the time and flex-worthy production nearly 11% of the time, even the most consistent ones over the past three years are only worthy of your lineup 56% of the time.
2024 Player Comparison
Last year, Dalton Kincaid was drafted as TE4 towards the end of round two at 22nd overall. George Kittle was drafted two rounds later as TE8 at 46th overall. While not as close in ADP as our quarterback example above, they were both taken in the top four rounds of drafts.
We have discussed Kittle a bit already, as he is an interesting case study for tight ends. I mentioned that his weighted consistency score was barely top-20 over the past three years due to his raw consistency being subpar in two of those three years. 2024 was the one good year of the last three, where he ranked 2nd in consistency and returned 13 start-worthy games (86.67% rate). However, Kittle did not have any additional flex-worthy games.
Who was 1st in consistency last year? Yep, it was Kincaid who slightly beat out Kittle’s raw consistency score, but returned only four start-worthy games (30.77% rate) while also having three more flex-worthy games (23.08%). Much has been made about Kincaid’s lackluster season, with many speculating that numerous lingering injuries throughout the year were the primary factor for his underwhelming production.
Regardless of the reason, Kincaid returned the worst starting game rate of any top-12 drafted tight end by ADP, and at his second-round cost, it was a massive loss for many fantasy managers. However, he still was the most “consistent” tight end last year, which emphasizes that raw consistency, even for tight end, should not be a primary consideration for why you draft them.
Aside from Kincaid’s and Kittle’s objective consistency rankings, looking at the percentage of games they subjectively “Won” your week for you (15 or more points), Kincaid achieved this one time (7.69%), whereas Kittle did so 11 times (73.33%). As we discussed in our tight end therapy session in the last article, you should expect more from your tight end. Don’t settle for consistent mediocrity.
Conclusion
Whether evaluating objectively or subjectively, consistency is a very nuanced concept in fantasy football. It requires even more perspective when evaluating it positionally, as there are different thresholds for successful production and varying opportunities for players to make your starting lineup.
Analysis of the top-12 consistency-ranked quarterbacks over the past 15 years reveals that consistency may be less important for weekly fantasy production than commonly assumed. Your most objectively consistent quarterbacks are only startable half of the time, and produce under 22ppg. You will recall from our introductory consistency article, top-12 weekly starting quarterbacks average nearly 27ppg.
When examining the 30 quarterbacks with the most start-worthy performances, their average consistency ranking was only marginally better than the overall sample at 5.6 versus 6.0. More tellingly, nearly 37% of these high-performing quarterbacks (11 of 30) actually ranked in the bottom half for consistency (6-12), demonstrating that less consistent quarterbacks can still deliver frequent start-worthy weeks.
This suggests that for fantasy purposes, the ability to produce elite weekly performances may matter more than maintaining steady, predictable output, as volatile quarterbacks who alternate between boom and bust weeks can still provide substantial value if their ceiling games occur frequently enough. And we discovered that most of the top start-worthy quarterbacks have an element of mobility to their game, capable of achieving the elite weekly output needed to win your matchups.
The tight end position shows a stronger correlation between consistency and weekly startability compared to quarterbacks, though the relationship remains modest. The 52.68% start-worthy rate of the top-12 consistency-ranked tight ends suggests they are only slightly better bets than their quarterback counterparts.
The top start-worthy tight ends averaged a weighted consistency ranking of 4.1, representing a more meaningful improvement versus the overall sample average of 6.0 than what was observed with quarterbacks (5.6 average). This 1.9-point difference versus the average suggests that consistent production is somewhat more valuable for tight ends than the marginal 0.4-point difference seen with quarterbacks.
The marginally enhanced importance of consistency at tight end likely reflects the position's inherently lower scoring ceiling and smaller talent pool, where steady, reliable production plays a slightly larger role for fantasy relevance. Of course, players like George Kittle can buck this trend, producing a top-3 start-worthy rate over the past three years while barely being top-20 in terms of weighted consistency. His talent and status as a top two target earner on his team provide the upside that tight ends need to weather their inherently volatile fantasy production.
While fantasy football is a wildly inconsistent game by nature, influenced by elements like weather, luck, injuries, or any other myriad of factors, we can look at past data to help provide some insight into future outcomes. Knowing certain players may have a reputation for being consistent, but understanding how that translates to start-worthy production is important.
Balancing the perception of a player as boom-bust while also factoring in their weekly startability (and week-winning upside) can help give you the edge you need when considering consistency in your drafts.
Next time, we will explore the heart of our teams, the positions that typically fill over half of our starting lineup spots any given week. Since running backs and wide receivers often fill our flex spots, having a better understanding of how their consistency factors into top-24 weekly production as well as flex-worthy starts can help you on draft day and throughout the season as you set your fantasy lineups.