Every team should have a dome. Plain and simple.

To the naysayers who love seeing fields covered in snow, sure, I’ll get my T levels checked. But I’d bet there’s a fantasy manager or two in that crowd who would love to avoid a windy forecast before a game. Or another who still remembers making a last-minute roster move because of the weather that lost them a crucial matchup. 

We’ve all got a bad-beat story when the elements are a factor, and we’ll have more this season for two reasons.

First, we can’t reasonably plan for things like high winds. Weather models update well after we’ve made moves on the waiver wire. And second, if we want to adjust our lineup, the remaining options reduce our expected results. 

So, we need a better idea of what matters when the skies (literally) try to rain on our attempts at being a league winner.

Temperature, precipitation, and wind can individually or collectively play a role in a game’s outcome, but let’s start small. Using historical data, I’ll focus on wind effects to break down what we’ve come to accept and what matters regarding fantasy football.

Premium

Go FantasyLife+

Your league is unique. Your advice should be too. Get personalized tools, rankings, and projections tailored to your fantasy football league settings.

Learn More

What do we already know about wind effects?

Stop me if you’ve seen something like this on social media or heard it on a podcast.

Increased wind speeds mean shorter passes.

Strong winds force teams to run the ball more.

Because I can’t fathom playing football in windy conditions, I can accept statements like those on vibes alone. And folks can point back to gusts making an otherwise boring contest somewhat comical. But the data checks out.

ESPN’s Brian Burke wrote about weather effects in 2012, and the chart has served as a backstop for weather analysis.

Even I used it when looking for data to support my priors about the negative correlation between wind and the passing game. But since this study was over a decade ago, I updated it with data from 2018-22.

I didn’t adjust for interceptions or touchdowns but obtained a similar trend. As wind speed increases, passes get shorter. However, we’ve overlooked some fine print when pulling takeaways from charts like these.

Fantasy Life's 2024 Fantasy Football Draft Kit is LIVE (and free)! Click here to access all of the content you need to dominate your drafts.

What’s a better way to use this data?

To start, consider the sample size. Each point in either of the above charts is an average across however many games occurred at that wind speed. The more, the better. But if I recreate that same graph without averaging, there’s an obvious issue at the extreme.

From 2018 to 2022, there were 1,311 regular-season games, and only 48 (3.7%) had recorded wind speeds over 20 mph. Look at how few dots there are as the graph extends to the right. So we can’t automatically assume the trend observed between 10 and 20 mph (with more samples) will continue. 

But let’s say that it did: It still doesn’t tell us much about the fantasy implications for the affected QB.

Defensive coverages have put the onus on offenses to be more efficient in their short game. Patrick Mahomes’ 2022 passing aDOT (7.7) is down nearly two full yards from the high-flying attack he commanded during his first full season. Joe Burrow had to find more success in the intermediate parts of the field to keep the Bengals on track. 

So shorter passes aren’t necessarily a critical blow when receivers can work after the catch. 

Regardless, play volume seemed like a simple test to confirm whether the correlation between wind speed and the passing game existed.

I looked at play volume in two ways: total offensive plays run and pass rate over expectation (PROE). While total plays could get skewed by more runs, PROE accounts for the game environment to measure what a team should do—and usually, they should be passing. Nevertheless, if the wind affected passing games, there’d be a similar trend.

Simply put, there wasn’t. Well, at least not right away.

Downward shifts in either plot aren’t evident until after 20 mph. And, like I pointed out before, the lack of samples in this regime gives us an incomplete picture. Regardless, we can start to push back on past biases.

Shorter passes can be good if the right scheme and personnel are in place. However, more wind doesn’t always equate to a situation fantasy managers must avoid. Until you see the forecast get past 20 mph, hold tight.

What’s a better way to look at weather data?

For a quick review, disregarding wind speeds below 20 mph is a way to ignore most of the noise on game day. Or, put it another way, 900 of the 924 non-dome games over the last five years. Not bad! 

But we’ll all wake up on Sunday to a bad forecast or see a tweet thread (likely from me) about games to monitor and will need a way to reset our expectations. Luckily, there are two metrics that measure what our “expectations” should be.

As I mentioned, PROE blends how often each team calls a pass play with the expected probability of an attempt given in-game stats like down, distance, and the score. And when plotted against wind speed, we see typical behavior despite the lack of data points.

Nearly 70.0% of the dataset resides in the first three velocities. From that sample, PROE falls at an average of 2.6 percentage points. If you applied that to a team like the Bills, we should expect Josh Allen (who had a PROE of 7.8% at 20 mph) to have a -25.2% PROE at 30 mph. 

But the trend isn’t linear and, as the data indicates, wind affects each QB differently.

So just knowing an offense’s passing tendencies in poor conditions isn’t enough. Wind may impact them collectively, but we can’t roster 28 QBs. We’d need to look at any changes due to high-speed airflow at an individual level. To do so, I calculated a baseline for each passer.

I set a baseline as any game with a wind speed less than or equal to 10 mph. However, there are clear flaws in this approach. Other elements, such as rain or snow, aren’t disqualifiers. Personnel considerations, like injuries to star players, are also ignored. 

But the calculation still yields the core concept: how a QB would operate under “normal” conditions. And, compared to their PROE at higher velocities, we can see a change at the player level instead of a global trend.

But surprisingly enough, the outcome is the same. Most of the data points still exist between 20 and 23 mph, and the average drop from a QB’s baseline was 2.8 percentage points. Even with the consistency, we’ve switched the focus from a league-wide trend to a passer and their team. Looking at total plays run tells a similar story.

Teams in the full sample averaged 62.6 offensive plays on a non-windy day. Their mean dropped to 60.9 plays at 20+ mph. Two attempts (likely) won’t alter our fantasy matchups, but the wind has been a factor in how some teams have approached their playcalling. To zoom in further, we can use accuracy metrics to focus on the signal-callers themselves.

Completion percentage over expected (CPOE) works like PROE. A throw’s distance, its depth and location, helps set the expectation, and the outcome (a catch or failed attempt) determines the over or under. And using the same baselining process, the results highlight how QBs fare as the wind speeds increase.

Passers experienced an average 1.6 percentage-point drop in CPOE across the 34-point sample. Of course, there are some major dips (Baker Mayfield, Week 6 in 2020), but the trend doesn’t paint as dire of a picture as we’ve come to think. If anything, our takeaways should come from a combination of the drifts we see in the data to inform our fantasy decisions.

How can we use this data?

Fantasy managers use multiple sources to draw their conclusions. Weather effects shouldn’t be any different. In fact, because of the small sample sizes, a push for more variables to pivot on should be anyone’s starting point. And since we’ve got three (PROE, CPOE, and total plays), we can see which passers become problematic against the wind.

The list looks daunting, with a few names you usually wouldn’t see paired together (Jake Luton and Josh Allen). But two overall pictures give us a better jump-off point.

The average fantasy point total from the group was 16.0 points. Six of the 11 above-average QBs had positive PROEs, and nine had a CPOE greater than zero if weather weren’t a factor. 

In short, good passers stayed good. At least for fantasy purposes, they did. 

However, the table emphasizes that each was affected differently and still came through for fantasy managers. A closer look showed why.

I split the group of 28 QBs into three buckets (the numbers are through 2022):

• Aided by Weather: PROE, CPOE, and Plays Run were either positive or negligibly off (+/- 3 percentage points) from their baseline
• Unaffected by Weather: PROE, CPOE, and Plays Run were slightly off from their baseline
• Affected by Weather: PROE, CPOE, and Plays Run were somewhat or significantly down from their baseline

It would break apart an already-limited population, but the names and offenses could help determine why each succeeded or failed.

Unsurprisingly, the “Aided by Weather” category had the fewest games. But the leader of the group made the analysis simple. Russell Wilson’s 32-28 dog fight with Cleveland in 2019 featured a 1-point spread against a Browns offense that was top 10 in neutral passing. As a result, we got classic Wilson with nine rushing attempts. The matchup dictated the outcome and not the weather.

Conversely, Fields' run-in with Bills Mafia on Dec. 24, 2022, while already nursing a separated shoulder, highlighted the need for QBs to have a larger, more efficient passing component. Fields entered Week 16 dead last in EPA per dropback, and Buffalo’s defense was a top-10 unit. Again, a deep dive into the two teams would’ve set similar expectations for the game. Let’s look at the second group.

Joe Burrow, Josh Allen, Kyler Murray, and Ben Roethlisberger were the only QBs with a positive PROE as their baseline. So passing alone wasn’t the differentiator. It was scrambling.

While unplanned, scrambles are highly efficient as QBs turn a would-be sack into a positive gain. Six of the seven signal-callers who averaged more than 16.0 fantasy points in 20+ mph winds were top 10 in total scrambles during their respective seasons. 

So while high-end rushing talents like Allen and Murray will always garner our attention, we should consider passers with some wiggle in inclement conditions.

Meanwhile, pure pocket passers like Derek Carr, Teddy Bridgewater, and Josh Rosen already had questionable arm talent (as it relates to fantasy) without even a minor rushing role to raise their floor. Their ADP or DFS price would already reflect their ceiling potential, but this archetype of QB doesn’t fare well in high winds.

Yes, Chiefs fans, I know. Patrick Mahomes doesn’t belong in this group. 

He had the seventh-highest fantasy-point average in the group, and he’s sitting next to Taylor Heinicke. But don’t blame me. Blame the teams he faced.

• 2021, Week 14 vs. LV—Final Score: 48-9
• 2021, Week 16 vs. PIT—Final Score: 36-10
• 2019, Week 13 vs. LV - Final Score: 40-9

If gravity doesn’t affect Mahomes, neither does the weather. And while Rodgers led the group in average points, he did it in true Rodgers fashion. Across his three games with wind speeds over 20 mph, four of his six touchdowns came on passes of seven yards or less. They converted one from the goal line. 

Regardless, most of this group (10 of 14) led run-oriented offenses that only leaned further into their strength. And still, without a scrambling element to their game, the probability of them failing to meet expectations only increased.

Are there other factors to consider?

I’ve outlined reasons for going against traditional thoughts on offenses in windy conditions. Plus, I put together profiles for QBs we can target with better chances to excel if high winds occur. But in redraft, we’re stuck with the guy we drafted or whoever’s available on the wire. So, if game day comes and the wind speeds aren’t favorable, two aspects of the location are worth noting.

Air flow direction

Wind velocity gets used to describe the weather, but we focus on only one part of the measurement: the speed. But direction also plays a role. And it’s not just the direction of the wind but how it’ll flow across the stadium. Let’s go back to the Chicago-Buffalo game I mentioned earlier.

Recorded wind speeds were 25-26 mph around Soldier Field at kickoff. Like you, I saw the number and started checking my lineups. But I also glanced at the direction in the forecast. Winds came from the west, generating air flow directly across the stadium and causing a cross breeze. And, for a passing attempt, that poses a problem.

A properly thrown football has two kinetic forces acting on it: forward momentum coming from the quarterback to get the ball down the field, and a rotational part. The rotation, or spiral, keeps the ball on its intended course with less surface area for the air to change its direction. But, if the wind is already pushing against the ball and the QB can’t execute a tight spiral, there’s a greater chance of a passer’s accuracy declining. 

There were 22 occurrences of a QB’s CPOE difference (baseline CPOE subtracted from their in-game CPOE) deviating by more than three percentage points. Within this sample, 45.5% happened when the angle between the airflow direction and the stadium’s orientation approached 90 degrees. 

Stadium height

I said it at the start of this article, but every team should have a dome. Owners make their stadiums look so large and lavish. Let’s try and keep everyone warm too! But I’ll credit a few who made things easier for fantasy managers when a windy day comes.

Seating areas can act as natural barriers, forcing the wind to flow around the playing field or up the structure's walls. MetLife Stadium has an ideal design, with the sides primarily closed and walls extending to an estimated height of 227 feet. 

Meanwhile, kickers become public enemy No. 1 at Gillette Stadium if the wind billows through the open section at the north end. So, a stadium’s design has been a factor, and its effect appears in the data.

The current average stadium height is approximately 180 feet, and 19 games happened at 20+ mph winds in stadiums below the league average. Of the 38 QBs in those games, 20 fell more than three percentage points below their baseline CPOE. While the limited sample size negates any direct conclusions, a game’s location can play a role in the wind’s effect on the QBs.

Join Fantasy Life’s Discord here!

Premium

Go FantasyLife+

Your league is unique. Your advice should be too. Get personalized tools, rankings, and projections tailored to your fantasy football league settings.

Learn More