Cracking the Code: Pitching at Coors Field


Since the Colorado Rockies played their first game in 1993, the organization has never put forth a team ERA under four (sporting a home ERA of over six the last three seasons). While the organization has put together formidable lineups for multiple years, the magic touch on the hill has eluded them. Many attribute this struggle to the location that the club plays its games- Coors Field in Denver, Colorado. 

The “Mile High City” has long been known as a pitcher’s graveyard, attributing this doom to physics — there is less dynamic air pressure at altitude (i.e. the air is thinner). In fact, a study by the University of Illinois validates that the air density in Denver is only eighty two percent of what it is at sea level. Thus, the ball tends to fly. 

It’s not a secret, the ball flies at Coors Field

In order to counter this, the Rockies have oftentimes fielded rotations full of sinker ballers, with the idea of counteracting, or at the very least limiting Coors Field from being their achilles heel via inducing ground balls at a high rate. 

This inherently makes sense. The weather and air data at altitude statistically benefits a sinker baller. The significant drop in dynamic air pressure lowers the amount of drag on the ball mid flight (additional velocity as ball reaches the plate). Additionally, the dry air found in Denver (hydrogen and oxygen molecules are heavier than water vapor molecules) also greatly reduces the magnus force on the ball as well. This is critical when it comes to figuring out how to be a successful pitcher in Colorado. The lack of magnus force greatly alters the break profile of pitches relative to how they would move at sea level. According to the aforementioned University of Illinois study, the average magnus force drop off in Colorado mimics that of the air thickness, sitting at around eighty two percent of what is typical at sea level. So, what does this all mean? Lower spin reigns supreme. 

Allow me to elaborate on the effects that less magnus force has on typically dominant fastball pitchers. The Illinois study came to the conclusion that pitches with backspin at Denver’s altitude will generally end up dropping about four more inches on average (gravity included) than at sea level. For an arm like Trevor Bauer, who leads the MLB in fastball rpm, this would mean death. Bauer’s average fastball rise sits at only 3.9 inches above the MLB average per Statcast. If we apply four inches of drop to that pitch, his 93.5 mph average fastball now plays average in the break department, negating the advantage that he would normally have. In other words, he has gone from elite to mediocre at best.

A fine example of high rpm fastballs not doing what they are meant to (staying up via magnus effect). Buehler was 3rd in Major League Baseball in average fastball spin amongst starters in 2020. While Buehler did execute in terms of release in order to generate a whiff up/above the zone, the loss of upward lift due to the air characteristics sees it end up at the top of the zone, instead of above the hands and out of it- the result speaks for itself.

Not only that, but the study explains how the good old magnus effect destroys the breaking ball at altitude as well. While the fastball (backspin) gains in drop due to less magnus force pulling it upward, the curveball (the typical compliment for a fastball), which showcases frontspin, loses around four inches of break. Magnus force has an equal and opposite pull to the direction of rotation on the pitch. Meaning, the curveball will sit up, being rendered mostly useless, or at the very least far less effective. Think the changeup will do any better? Not a chance. Most consider the ideal changeup tilt to sit at 3:00 for a right hander and 9:00 for a lefty. This means that the pitch will showcase a lot of sidespin, helping to garner some armside run, and depending on the seam orientation and spin rate, some potential drop. While the changeup may drop more at altitude, it will lose a lot of its run due to the magnus pull that affects both fastballs and breaking balls, irrespective of their difference in spin. This is worth noting for multiple reasons; the most significant is that sinkers (more horizontal tilt than fastball; backspin) also lose run, making them more hittable. 

So far, this sounds like a very difficult nut to crack. However, there are some positives that are often passed over, and not even considered, that are swimming within this sea of pitching despair. In fact, these previously underappreciated characteristics could be the key to solving the enigma that is pitching at Coors Field. Here they are.

Solving the Breaking Ball:

I am explaining my theory on breaking balls first, as it is the most straightforward when it comes to direct application in Colorado. As mentioned prior, the curveball’s purpose, for the most part, is defeated at altitude due to a drop in magnus force hurting its vertical break profile. Unless a pitcher has a beyond elite curveball (3000 rpm +), it is likely to regularly get hammered. Thus, I feel that it is pretty safe to scrap curveballs altogether and substitute sliders instead as the ideal offering. While sweeping sliders will lose some run, the general idea of the pitch is to kill vertical break, while throwing a higher velocity pitch with some glove side break. Spin induced movement is not necessarily the most important aspect of the slider in Colorado, as the horizontal tilt and heavy gyro spin promote horizontal movement via its natural ball flight and gravity. This is important to note when building a slider for Coors field. Slight gloveside break is just enough, though the decrease in magnus force will see a heavy gyro pitch tack on some additional vertical movement. While this is counterproductive relative to a typical “ideal” slider’s profile, it does create a unique slurve-like action that is more beneficial than that of a 12/6 curveball with lost spin induced break (from altitude).

The drop in magnus force at altitude gives the slider a bit more of a dipping motion. There is little horizontal movement, though it is just enough to make it an effective pitch that can bite a corner, generate a whiff, or freeze a bat (as seen above)

If used correctly with the right spin and tilt characteristics (enough spin to garner average spin induced horizontal movement and solid horizontal tilt), it can still be a successful out pitch- though the changeup in this proposed arsenal will be the more important offspeed weapon – more on this a bit later…

While this was a hanger, it is still a fine example of why curveballs are not optimized for Coors Field. The ability for this mistake pitch to drop enough to where it may have been a double instead of a home run is not there- hangers become even easier to hit, and top of the line curveballs become average ones.

Solving the Fastball and Ending the Fastball/Sinker Debate:

In the Primer, I discussed how after analyzing air and weather data in Colorado, the “ideal” sinker benefits more than the “ideal” fastball at altitude. This happens because the magnus forces bearing on the ball promote more drop on pitches with backspin (both the sinker and fastball have backspin). 

Science clearly shows that what is seen as “ideal” or “conventional” at sea level does not apply whatsoever at altitude. While the sinker/slider combo is still effective at Coors Field, I believe that there is an oftentimes disregarded outlier that will be far more effective and dangerous: the low spin, high velocity four seam fastball. 

Again citing the University of Illinois study on ball flight at altitude, the thinner air in Denver makes a ball’s velocity lose about eight percent of its initial velocity at release by the time it reaches the plate. The same pitch thrown at sea level loses about ten percent of its initial velocity, or about one to two miles per hour upon arriving at home. Thereby, a hard fastball will play even harder at Coors Field. As previously discussed, pitches with backspin (fastballs) will drop about four inches more than typical due to there being less magnus force on the ball. As a result, rather than throwing a high spin, high velocity fastball (extremely effective at sea level), a high velocity fastball with low spin should be lethal in Colorado. The lower spin fastball already sees more gravity included break as less spin means less magnus force keeping the ball up. This break profile will only be enhanced via the eighteen percent drop in average magnus force being applied to the ball at altitude. The enhanced break and high velocity will make this fastball a very dangerous primary offering- forcing a lot of whiffs. While he has yet to throw in Colorado, Sixto Sanchez’s fastball profile is pretty much exactly what I am looking for. He averages 98.6 mph on the four-seam, with a spin of 2164 rpm. Even though Sanchez has to lean heavily into his velocity rather than spin to generate whiffs at sea level, the boosted vertical break profile would make his primary offering extremely difficult to hit at elevation. 

Game data on Rockies arms also backs this fastball theory. German Marquez lead all Rockies starters in fastball whiff rate in 2020 (15.1%). Unsurprisingly, his fastball traits mirror the exact profile I just described more than any other Rockie arm (95.9 mph average velocity, 2156 rpm). John Gray, who had the second highest whiff rate on the starting staff (12.7%), also throws a four seam with low spin as his primary offering (2148 rpm average, though the lower velocity- 94 mph average may limit whiffs more). The two also lead the club in xwOBA amongst starting pitchers throwing their high velocity pitch over thirty percent of the time (.382 and .323 for Marquez and Gray respectively in 2020). While Marquez’s overall fastball profile may be more “ideal” following my formula, he located it higher up in the zone than Gray on average (2.56 vs 2.48 feet). While an elevated fastball is ideal for high velocity or spin, we are intentionally selecting low spin arms with the intent of them picking up additional vertical break on the pitch (at Denver altitude). Thus, the lower xwOBA belonging to Gray makes sense, as his risk of a ball dropping right into the hitting zone is lower due to his looking to throw the ball a little bit further south than Marquez in the zone (Gray fastball xSLG also 85 pts lower than Marquez). 

Marquez is able to generate whiffs on the fastball lower in the zone more than typical due to his low spin rate and additional lack of magnus force (altitude) allowing the ball to drop at a greater rate. This approach is more ideal than up due to the aforementioned lack of excess magnus force that would normally push the fastball up- giving it “rise” (how we see most high velocity, high rpm putaway fastballs thrown). Video via MLB

In other words, results are already proving that the low spin four seamer at higher relative velocity plays best in Colorado. When paired with a slider (John Gray), rather than a curveball (German Marquez), it can become even more effective. However, a unique, elite changeup, as I will describe below, is what could really turn an arm from plus into elite at Coors Field.

Solving the Changeup:

As a reminder, the loss in magnus force absolutely kills sidespin. While this is beneficial from a vertical break standpoint, it kills horizontal run on changeups (four inch average loss in horizontal run per Univ. of Illinois). With the typical “ideal” changeup featuring a massive amount of sidespin due to a horizontally shifted axis, its profile is simply not optimized in Colorado. However, there is a solution. In order to find it, one does not need to look any further than one, Ian Anderson. While Anderson’s changeup is elite, its uniqueness — specifically its tilt and seam orientation, make it a match made in heaven for Rockies arms. 

As seen above, Anderson does not pronate on his changeup, releasing the ball off of his middle and ring fingers after stuffing it into his palm. This slows the ball down enough to garner a lot of whiffs while keeping a relatively similar tilt and spin as the four seam fastball. Video via Michael Augustine

As seen in the clip above, Anderson’s seam orientation on the changeup grip exactly mimics that of a four seam. This is especially crucial when watching how he releases the ball. Typically, a pitcher will pronate his hand in order to shift the axis horizontally, creating sidespin. However, as seen above, Anderson does not pronate at all, leaving the tilt of the pitch extremely vertical- nearly identical to his fastball. 

The end result is a pitch that mimics the spin characteristics of the fastball, while moving a lot slower due to the ball rolling off of his three outside fingers. The index finger, which applies the most pressure on the ball, generating the most velocity, is not at play here. While this pitch won’t generate the horizontal break of a 3:00 tilted change, it will still see some vertical break- which will be exaggerated further in Colorado (similar to the fastball). Essentially, the pitch will fool hitters based on the spin, tilt, and release profile looking the same out of the hand (no pronation means no picking it up). Instead of worrying about generating maximum movement while working against the laws of physics (Coors Field), max deception will be much more effective (there is a reason Anderson’s whiff rate is 24.5% higher on his changeup than his fastball- 39.8% vs 15.3%).  

How this Arsenal Plays on the Road:

While physics and game data back the above mentioned ideal arsenal for Colorado, there is still a massive elephant in the room: how will it play on the road? Afterall, only half of the Rockies season is played at Coors Field; it needs to work at sea level as well. Thankfully, the answer to this question is very straightforward: it will be effective, though the approach and sequencing of the pitcher will need to change. At sea level, magnus forces on the ball return to normal. This means that the low rpm fastball will not break quite as much, and will sit in the zone more due to the lack of spin generated magnus force (higher spin/magnus fastballs are better for whiffs at sea level). This means that the whiff rate on the pitch will likely plummet. However, this need not be an issue if the arm is willing to shift his sequencing.

The increase in magnus force will give the slider more break, making it, and the already proven effective at sea level changeup (thank you Ian Anderson), the out pitches on the road. The natural high velocity of the fastball will still garner whiffs, though as seen via Sixto Sanchez, it is best used to generate soft contact when thrown at an even or similar rate as the offspeed pitches. This is contrary to the more fastball heavy sequencing at home. Due to the shortcomings of the fastball, and the aforementioned uptick in spin based movement, the offspeeds should be used as the putaway pitches on the road. It is a very different approach away from Coors Field, but you need to work with what you have, and in the case of Colorado, this seems almost necessary.  

Notice where Sixto locates the fastball. This 1-2 pitch is designed to induce soft contact rather than a whiff, which is reflective of what I would expect from Colorado arms showcasing similar pitch traits away from home (less frequency of use, jumps on the bat more- not squared up in the zone). Instead, the slider and changeup become the putaway pitches due to the bolstered movement profiles via plus magnus force/spin at sea level. Video via Fox Deportes

What the Rockies are Doing Now/How They are on Their Way/Conclusion:

It is clear that the Colorado Rockies are at an inherent disadvantage on the bump. Since 2007, they have only finished in the top half of the league in team ERA twice, and have finished with a home ERA above six in both 2019 and 2020. However, there are clearly ways to shift the disadvantage into something more than effective, which would turn the tables completely at Coors Field. Moving forward, I would expect the Rockies to pick up on the low spin, high velocity fastball being more effective than the sinker at home, and hope that the Player Development team looks into alternative offspeed options, such as the Anderson- esque changeup. I feel that Mason Black (Lehigh) would be a fantastic draft candidate for 2021 to emphasize this idea and trend in their system.

Mason Black is an ideal draft prospect to look at in regards to identifying who fits the mold of a potential top arm at Coors Field

Black has topped out at 102, and showcases a very low spin fastball, along with a hard slider. The addition of a unique, vertically tilted changeup would literally see him fit the exact profile  described in this piece, and could translate into him being a top of the line type of arm for Colorado in the future. While this altitude dilemma is an organic challenge that is immutable, the onset of the data age and innovative thinking can help give the Rockies more overall pitching success.

Leave a Reply

Powered by

Up ↑