What Is Whiff In Baseball

WHIFF! Strikeout Rates Explained

Even while there are several methods for skinning a cat, training a fly, and murdering a man, among other things, it appears that there is only one solid and consistent method for striking batters out – to get them to whiff on pitches. Of course, this isn’t exactly mind-boggling, considering that the only ways to end a strikeout are with a swinging or a looking strike, and the only ways to acquire strikes are with swings, fouls, or looking strikes. Nonetheless, when I ran the statistics for 2012 as well as the period from 2007 to 2012, I was taken aback by how dramatic the findings turned out to be.

Because Whiff/Swing is new (at least, the leaderboards are), and because it fared somewhat better than SwStr percent at explaining K percent variance, I’ll be referring to BP’s version of whiffs in this post.


I ran regressions for 2012 for K percent against the percentage of fastballs thrown (FA percent), the percentage of sliders thrown (SL percent), the average fastball velocity (vFA), the overall strike rate (Strike percent), the overall swing rate (Swing percent), the first strike rate (F-Strike percent), the horizontal and vertical pitch movement (H Mov and V Mov, respectively), the walk rate (BB percent), and finally the SwSt.

In 2012, I set a threshold of 40 innings pitched as the cut-off point (or approximately 500 pitches).

I realize that for a strikeout study, I should have included all pitchers, and I can certainly do so in the future if there is a good justification for doing so, but for the time being, those were the cutoff marks I used, and I apologize for that.


My regression analysis skills are a little rusty at the moment, but this Baseball Prospectus piece from Matt Swartz from a few seasons ago appears to confirm my findings that swinging strike rates are highly correlated with strikeouts. I’m not sure how much of a skeptic I am, but I think it’s worth mentioning. Matt’s research was primarily concerned with projecting strikeout rates for the following year (he discovered that after a baseline K percent is set, SwStr percent doesn’t tell you anything else), whereas my goal was to simply describe the anatomy of strikeouts (that is, this is descriptive, not predictive, for now).


Whiff/Swing outperformed all other indicators in the single-year regression for 2012 pitchers, and it was the most consistent of the indicators that I looked at. Because the R-squareds for these other statistics are so low, I’m not sure where I went wrong, but it’s likely that pitchers may manage strikeouts regardless of their repertoire or general location ability, as long as they have swing-and-miss stuff on their repertoire.

The results of the regression analysis comparing K percent to each indicator are shown in the chart below.

Correlationwith K%

Whiff/Swing 0.656 0.034
SwStr% 0.636 0.034
FA% (pfx) 0.108 0.054
Strike% 0.047 0.056
O-Swing% 0.046 0.056
Swing% 0.039 0.056
V Mov 0.025 0.056
BB% 0.020 0.056
vFA (pfx) 0.020 0.056
SL% (pfx) 0.013 0.057
F-Strike% 0.004 0.057
H Mov 0.001 0.057
All 0.786 0.027

All of the variables taken together can explain nearly 80 percent of the variation in pitcher strikeout percentage, leaving about 20 percent to random variance or, in the case of strikeout rates from other means, pitchers on the tails of the distribution (or other elements I didn’t measure) to explain the remaining 20 percent. When I looked at all of the years from 2007 to 2012, the story was the same as before.

Correlationwith K%, 2007-2012

Whiff/Swing 0.687 0.025
vFA (pfx) 0.216 0.039
FA% (pfx) 0.130 0.041
BB% 0.077 0.043
O-Swing% 0.076 0.043
SL% (pfx) 0.025 0.044
Strike% 0.014 0.044
Swing% 0.013 0.044
V Mov 0.008 0.044
H Mov 0.004 0.044
F-Strike% 0.000 0.044
All 0.858 0.017

With our parameters, we can forecast even more of the variance in strikeout rate over the longer term, and Whiff/Swing is much more important, accounting for 69 percent (haha, 69 percent) of the variance in strikeout rate. However, I accidentally deleted the SwStr percent from this data set and realized it too late, but I re-ran it and got an R-squared of.667, which was still a good result but fell short of Whiff/lead Swing’s and hilarious R-squared result. The formula our model generates for utilizing Whiff/Swing to predict K percent allows us to build a “Expected K percent,” which is K percent =.007502 + (.85006*Whiff percent) in very basic terms.

We can detect some outliers by taking this a step further and utilizing our (admittedly rudimentary) xK percent to identify them.


Craig Kimbrel 50.20% 42.17% 36.60% 13.60%
Brad Lincoln 24.30% 14.45% 13.03% 11.27%
Jake McGee 34.40% 26.57% 23.34% 11.06%
Kenley Jansen 39.30% 32.85% 28.67% 10.63%
Tom Gorzelanny 20.30% 35.48% 30.91% 10.61%
David Robertson 32.70% 25.37% 22.32% 10.38%
Jason Grilli 36.90% 30.65% 26.80% 10.10%
Aroldis Chapman 44.20% 40.00% 34.75% 9.45%
Antonio Bastardo 36.20% 30.93% 27.04% 9.16%
Sean Doolittle 31.40% 25.78% 22.66% 8.74%
David Hernandez 35.30% 30.90% 27.02% 8.28%
Vicente Padilla 23.40% 17.85% 15.92% 7.48%
Ernesto Frieri 36.40% 33.22% 28.99% 7.41%
Casey Janssen 27.70% 23.10% 20.39% 7.31%

Due to the lower sample sizes, it comes as no surprise that all of the pitchers in this group are relievers. The chart below is restricted to only those who are just getting started.


Mike Fiers 25.10% 20.68% 18.33% 6.77%
Cliff Lee 24.40% 20.73% 18.37% 6.03%
Stephen Strasburg 30.20% 27.88% 24.45% 5.75%
David Price 24.50% 21.39% 18.93% 5.57%
Marco Estrada 25.40% 23.08% 20.37% 5.03%
David Phelps 23.20% 20.92% 18.53% 4.67%
Travis Blackley 16.00% 23.26% 20.52% 4.52%
Derek Lowe 8.60% 14.39% 12.98% 4.38%
Vance Worley 18.10% 15.36% 13.81% 4.29%
Alex White 13.90% 20.47% 18.15% 4.25%
Max Scherzer 29.40% 28.71% 25.16% 4.24%


Our own Glenn DuPaul has been doing a lot of study lately on how basic K and BB-based ERA estimators (including his newpredictive FIP) work, and it’s becoming increasingly important to understand what goes into striking out hitters. The number of swings that end in misses appears to be the most accurate predictor of strikeout performance, according to the data (aka “dominance”).

Items of Interest

This is maybe the least shocking finding of any research you’ll read this year, given the greatest Whiff/Swing rates belonged to Craig Kimbrel (42.17 percent) and Aroldis Chapman (40.00 percent). * Aaron Cook had the lowest Whiff/Swing rate of 8.92 percent, making him the only pitcher to have a percentage less than 11 percent in his career. Basically, if you swing with him, you’re equivalent to Marco Scutaro in terms of contact ability. Since a result, Tyler Chatwood, Bobby Parnell, and Ben Sheets are the poster boys for this method, as each of their real K percents was within.05 of their xK percent.

Overall, the R-squared value of Strike percent was only.56, indicating that there is no clear reason for BB percent other than “swing and miss,” which makes logical sense. Follow Blake Murphy on Twitter at @BlakeMurphyODC.

Whiff+: A Look at Which Pitchers Dominate by Generating Swings and Misses

Explainers of Metric Terms When it comes to assessing a pitcher’s ability, the whipff rate does not provide a realistic picture of the situation. Whiff+ makes up for the difference between whiff rate and other factors. While command+ gives a more comprehensive and accurate evaluation of a pitcher’s command, whiff+ may provide a more in-depth examination of a pitcher’s stuff by assessing the rate at which he creates swings and misses (swings and misses rate). Whiff+ allows analysts to compare players from multiple seasons on the same scale, which is something that other swing and miss rate measures are unable to do.

  1. Whiff+ is calculated based on the average pitch type used in the league for that season.
  2. Whiff+ compensates for this.
  3. As a result, we may determine that he has a whiff+ of 125, which is 25 percent more than the league average of 100.
  4. The purpose of whiff+ is to determine how effective a pitcher is in generating whiffs based on the sorts of pitches he throws.
  5. Having finished third in the majors in whiff+ in 2019 after going 20-5 with a 2.50 ERA and an MLB-best 326 strikeouts, Gerrit Cole ranked third in the majors in whiff+ in 2020 after going 7-3 with a 2.84 ERA and finishing sixth in baseball with 94 strikeouts in a 60-game season.
  6. Even though we’ve known for a long time that not all balls and strikes are equally effective (or unsuccessful), we now have additional methods to assess the difference in efficacy for each pitch.

MLB Swing Profiles

The speed at which a ball was struck by a hitter, measured in miles per hour. The angle at which a ball was struck by a hitter, measured in degrees. In other words, a hit ball with the ideal combination of exit velocity and launch angle A ‘hard-hit ball’ is defined as one that has an exit velocity of 95 mph or above, according to Statcast. Batted ball competition with a launch angle ranging from eight to 32 degrees is held every year. A Batted Ball Event is a representation of any batted ball that results in an event.

  • The movement of a pitch is measured in inches, both in terms of raw numbers and as a comparison to the mean.
  • The amount of spin applied to a pitch was measured in revolutions per minute.
  • The amount of time it takes for a catcher to get the ball out of his glove and to the base when attempting a stolen base or pickoff attempt.
  • Time from one base to another How much time, in seconds, it takes a runner to go from one base to another, for as from Home Base to First Base.
  • When a pitcher makes his first movement or releases his pitch, the distance a runner is ranging off the bag is measured in feet.
  • A range-based indicator of talent that indicates how many outs a player has saved in comparison to his or her competitors.
  • xBA is a statistic that indicates the chance that a batted ball will be hit.
  • xERA is a straightforward 1:1 translation of xwOBA that has been translated to the ERA scale.

What does whiff mean in baseball?

Parker Collier posed the question. Score: 4.6/5 (18 votes) for a fluke. A stroke with a swinging motion (referring to the bat whiffing through the air without contacting the ball).

What is whiff rate in baseball?

It is generally used in relation to pitchers, and it is defined as the ratio of the number of pitches swung at and missed to the total number of swings in a certain sample. In the case of a pitcher who throws 100 pitches at which hitters swing and the batters miss contact on 26 of those pitches, the pitcher’s whiff rate is 26 percent.

What does it mean to whiff in a game?

Whiff is a phrase often used by fighting game aficionados to describe to the act of completely missing an opponent in a combat situation. Whiff punishing is the act of an opponent effectively retaliating after a whiff, either by striking them back or by completing a combination of strikes.

How do you calculate whiff rate?

By combining these datasets, I have two data frames, one for hitters and one for pitchers, which I will use to investigate whiff rates, where a whiff rate is the number of whiffs divided by the number of swings.

What is whip mean in baseball?

WHIP (Walks and Hits Per Inning Pitched)21 questions were identified that were connected.

Who has the best WHIP in baseball?

Addie Jossis the all-time leader in WHIP with a lifetime mark of 0.9678 throughout her professional baseball career. In addition to Walsh, the only other player with a lifetime WHIP less than 1.0000 is Ed Walsh (0.9996).

What is a good whiff percentage?

Percentage of swinging strikes: Also known as swinging strikes per swing. This is modeled after FanGraphs’ “Contact percent,” which monitors contact every swing, except that it makes use of Baseball Savant’s total swings and total misses numbers instead of FanGraphs’. Over the previous two years, the MLB average has been 23.28 percentage points.

What is hard hit percentage?

A ‘hard-hit ball’ is defined as one that has been hit with an exit velocity of 95 mph or higher, and a player’s “hard-hit rate” is simply the percentage of batted balls that have been hit with an exit velocity of 95 mph or higher. To achieve real production, you must reach speeds of 95 miles per hour.

See also:  Who Won Baseball Olympics 2021

What is CSW rate?

Hard-hit balls are defined as those that are hit with an exit velocity of 95 mph or higher by Statcast, and a player’s “hard-hit rate” is simply the percentage of batted balls that are hit with an exit velocity of 95 mph or higher. 95 miles per hour is required for genuine productivity.

Why is it so easy to whiff in Valorant?

They’ll miss their shots if the sights are set too high on their rifles. When it’s low, they’re more likely to hit body shots and lose to opponents who out aim them for headshots. Another helpful approach is to record your matches and look at how frequently your crosshairs aligned with the locations of your opponents’ positions.

What is whiff slang?

1a: a brief puff or tiny gust of wind, especially of air, odor, gas, smoke, or liquid spray b: the inhaling of an odor, a gas, or a vapor In the form of a faint puffing or whistling sound. secondly, the presence of a faint odor or suggestion of scandal 3: a strikeout is called.

What is whip slang for?

What is the slang term for a whip?

Since the late twentieth century, the slang term “whip” has been used to refer to an automobile. It’s also used as a verb, which means “to operate (a vehicle).”

What does G stand for in baseball?

Games that have been played (G) Grand Slam is a series of victories in a single sport (GSH) Toss The Ball Into Double Play (GIDP) The Groundout-to-Airout Ratio (GO/AO) is the ratio of groundout to airout. Pitch-for-pitch (HBP)

What does R mean in baseball?

A run is awarded to a player if he crosses the plate in order to score a run for his side. When calculating the number of runs scored, the method by which a player reached base is not taken into account.

What is the most important stat in baseball?

Hitting average, runs batted in, and home runs are the most often used batting statistics in baseball. To this day, a player who leads the league in all three of these statistical categories is known to as the “Triple Crown” champion of the season. For pitchers, the classic statistics of wins, earned run average, and strikeouts are the most often reported.

What is the hardest hit ball in MLB history?

Stanton blasted a 122.2 mph single off Atlanta Braves pitcher Max Fried while playing for the Marlins on that particular day. Stanton also owns the record for the most home runs hit with the hardest swing. In 2018, he hit a solo home run with the Yankees, smashing a ball 121.7 miles per hour.

What is the longest home run ever hit?

Even the cameraman was fooled by the longest home run in baseball history.

  • The following players have walked 535 feet: Adam Dunn (Cincinnati Reds, 2004) and Willie Stargell (Pittsburgh Pirates, 1978)
  • 539 feet: Reggie Jackson (Oakland Athletics, 1971)
  • 565 feet: Mickey Mantle (New York Yankees, 1953)
  • 575 feet: Babe Ruth (New York Yankees, 1921)
  • 575 feet: Babe Ruth

Is a foul tip a whiff?

To be clear, SwStr percent refers to the percentage of total pitches that a hitter swings at and misses, whereas Whiff/Swing refers to the percentage of total swings that a batter fails to land on.

What does FIP mean in baseball?

The term “Fielding Independent Pitching” (FIP) is defined in the MLB.com glossary.

What’s more important ERA or WHIP?

The WHIP measures a pitcher’s proclivity for allowing hitters to reach base, with a lower WHIP indicating better performance than an upward WHIP. While the earned run average (ERA) indicates how many runs a pitcher allows, the earned run impact factor (WHIP) gauges how successful a pitcher is against batters.

Who has the lowest ERA ever?

Tim Keefe held the record for the lowest single-season earned run average in baseball history with a 0.86 ERA in 105 innings thrown for the National League’s Troy Trojans in 1880, outscoring his nearest challenger by 52 runs. Dutch Leonard set a single-season record in the American League with a 0.96 earned run average.

Statcast Hitter Studs and Duds – Whiff Rate for Week 4

Taking a deeper look at hitters who are making excellent contact and those who aren’t, we’re back in the flow of things, figuratively and figuratively. In this week’s Statcast, we’ll be looking at Whiff Rate as we reach Week 4 of the 2021 fantasy baseball season. Whiffs are simply determined by dividing the total number of swings by the total number of swings, which results in a simple calculation. There are several phrases that are used interchangeably in sabermetrics that can be confusing to those who are not familiar with the field, such as Swinging Strike Rate (SwStr percent) on Fangraphs, Swing and Miss percent, and others that might cause confusion.

Check out the throwing side of our Statcast analysis, which this week focuses on Exit Velocity, when you’re through reading.

Promotional item of the week: Get a 50 percent discount on your MLB Premium Pass for 2022.

Our exclusive preseason Draft Kit includes 1 accuracy rankings, ATC and CUTTER forecasts, 15 in-season lineup tools, DFS cheat sheets, Research Stations, Lineup Optimizers, and other valuable resources. Sign Up Right Away!

Whiff Rate Leaders

Whiff Rate is 18.7%, which is quite high (87th percentile) It turns out that Shaw did not have the bounce-back season we had hoped for in 2020, but he could have one in 2021. After a successful return to Milwaukee, Shaw has raised his hitting average to its best level since 2017. While his batting average of.260 is nothing to write home about, it will be useful for fantasy purposes if he is leaving the yard and driving in runs. As of now, he has three home runs and 13 RBI in his first 14 games, which is a promising start.

It’s possible that Statcast is to blame for this to some level.

In this particular instance, Shaw intended to do precisely that, but ended up making matters worse.

Shaun Shaw is in the lineup every day and is expected to play third base for the foreseeable future.

Ozzie Albies, Atlanta Braves

Whiff Rate is 19 percent (in percentage) (85th percentile) This may seem strange, given that Albies has been one of the most disappointing first-round choices in the history of the game, particularly among hitters. With only a.157 batting average and no stolen bases so far this season, Albies isn’t providing much in the way of production for fantasy managers. However, it is not an issue of deteriorating plate discipline in general. His whiff percentage of 19 percent is the lowest of his career, and he’s only struck out 16.4 percent of the time this season.

One of the most surprising aspects is that Albies is ranked close or at the top of the xBA rankings.

The fact that he’s hitting the ball harder and striking out less indicates that he’s simply been unlucky, as seen by his.146 BABIP.

Raimel Tapia, Colorado Rockies

11.1 percent Whiff Rate is a percentage of the total (97th percentile) The fact that Tapia is not as well-known as the other players on our list doesn’t diminish his importance in any way. Tapia is more well-known for his speed and contact-inducing swing, but he hasn’t quite made it into the fantasy conversation yet. As a result, he currently ranks among the top 10 in K percent (12.3 percent) and has reduced his whiff rate by 7.6 points since last year. Tapia overachieved significantly previous season, recording a.321 batting average after having a.251 xBA, indicating that he would have some decline.

The recent analysis of Tapia by Sam Chinitz revealed that the outfielder is a great To waiver wire possibility because of his stance modification, which might also result in additional power gains.

The leadoff batter in Colorado should cross the plate 70-80 times while also recording double digit steals and posting a batting average in the mid-twenties.

Whiff Rate Strugglers

51.2 percent Whiff Rate is a significant figure (1st percentile) Since joining the league in 2010, Baez has been successful at both uplifting and deflating opponents. In the majors, he is tied for second with five thefts, which is a welcome surprise considering he only stole three bases all of last season in 59 games. He also has four home runs and 12 RBI, which puts him in the top half of the league in both categories. However, he does it with a.214 batting average and a godawful 45 percent strikeout rate, which makes him more of a liability in points leagues than anything else.

From 2018, Baez’s whiff rate has increased year after year, with a more dramatic increase of 13.4 percent since the start of the previous season.

The fact that he dislikes taking free passes hasn’t stopped him from managing to lower his walk rate this season to 1.7 percent.

If he abruptly quits jogging, on the other hand, the negative effects may begin to exceed the favorable effects.

Eugenio Suarez, Cincinnati Reds

41.2 percent Whiff Rate is a percentage of the total number of whiffs (4th percentile) Cincy’s newly-minted shortstop, similar to Baez, has been missing the ball with increasing regularity as the years go past. Suarez was whiffing at a rate of 24.4 percent in 2017, but that statistic is steadily increasing. According to popular assumption, the process of learning a new defensive position is interfering with his performance in the batter’s box; yet, the strikeouts had been building up even before the switch.

Suarez’s xBA has been steadily declining as well, which is not surprising.

After 63 plate appearances in 2021, he has a.164 xBA,.358 xSLG, and.352 xwoBACON.

His power numbers should improve, so it is not nearly time to abandon ship, but be advised that he may no longer be an elite fantasy performer across the board, and a decline in his batting average is possible.

Sean Murphy, Oakland Athletics

Rate of 40 percent for Whiffing (5th percentile) As part of our investigation into who is hitting the ball particularly hard early on, we looked at the leaders in Max Exit Velocity to get an indication of who is hitting the ball particularly hard early on because it can be a signal of rising power. Murphy was not reviewed individually in the article, but he does feature on the leaderboard, where he is ranked in the top ten percentile of the group. But the good news does not last long, as he is towards the bottom of the xBA rankings with a.177 mark, and he is whiffing at a rate that is 15 percent higher than it was last season.

After 40 plate appearances, he has scored one run, hit one home run, driven in five runs, and has a.161 batting average.

Murphy was notable for having a collapsed lung during the offseason, which necessitated emergency surgery.

While he has not yet reached the IL and is considered to be “healthy,” the delayed start to spring and after-effects of the wrist injury might be having an impact on his swing. It’s best to be patient with Murphy, especially considering that he just hit his first home run of the season yesterday.

Download Our Free News and Analysis App

Like what you’ve seen so far? Take advantage of our free fantasy baseball app foriPhone andAndroid, which provides 24×7 player news, injury alerts, sleepers, and prospects, among other features.

More StatcastFantasy Baseball Analysis

Baseball Savant provides a “Statcast Search” page where one can query the MLB Statcast Database and get interesting datasets, each of which can be downloaded as a csv file and easily imported into R. Baseball Savant also provides a “Statcast Search” page where one can query the MLB Statcast Database and get interesting datasets, each of which can be downloaded as a csv file and easily imported into R. I downloaded four files from this page, using the following link:

  • For all batters up to and including the last game of the 2017 season, I retrieved summary statistics on fastballs by using the Player Type = Batter and the Pitch Type = Fastballs filters. I downloaded summary data for fastballs for all 2017 pitchers by using Player Type = Pitcher and Pitch Type = Fastballs
  • I repeated the above two queries by using Pitch Type = Offspeed, first for Batters and then for Pitchers
  • I repeated the above two queries by using Player Type = Pitcher and Pitch Type = Fastballs, first for Batters and then for Pitchers
See also:  Who To Draft In Fantasy Baseball

Fangraphs also provided me with Standard Measures for hitters and pitchers, which I downloaded as well. By combining these datasets, I have two data frames, one for hitters and one for pitchers, which I will use to investigate whiff rates, where a whiff rate is the number of whiffs divided by the number of swings. (The relevant variables in the StatCast csv file are swings and whipfs, to name a few of examples.)

Batter Whiff and K Rates

As a starting point, it should be noted that a batter’s whiff rate is significantly correlated with his strikeout rate. In this section, I create a plot of whiff rates vs K rates. Chris Davis, Keon Broxton, Miguel Sano, and Joel Gallo are four batters that have very high strikeout percentages (more than 35 percent) in the majors. Dustin Pedroia, on the other hand, has the lowest whiff percentage and the highest strikeout rate among batters who have seen at least 500 pitches.

Pitcher Whiff and K Rates

A similar pattern of correlation between whiff rate and K rage may be observed in pitchers, as well. In the scatterplot, five pitchers show out as being exceptionally good at getting whiffs and striking out hitters, while four pitchers stand out as being exceptionally bad at getting whiffs and striking out batters.

How Does Whiff Rate Depend on the Pitch (Batter View)?

Let’s take a look at the variables that influence whiff rates. Based on my extensive baseball watching, it appears that off-speed pitches are frequently used to strike out batters. As a result, I would expect batters to have higher whiff rates on off-speed pitches than they would on fastballs. This serves as the impetus for the defining of the distinction. Whiff Rate on Offspeed Pitches minus Whiff Rate on Fastballs equals the difference. In the scatterplot below, I plot the batter’s whiff rate against the difference between the two numbers.

  • Almost all hitters fall above the line Difference = 0, indicating that they do have higher K rates on off-speed pitches
  • However, there are two interesting outliers — Scott Schebler and Corey Dickerson — who actually miss a greater fraction of fastballs than off-speed pitches
  • I’ve labeled two high K guys (Gallo and Broxton) and three high Difference guys (Cesar Hernandez, Alex Gordon, and Aaron Hicks) as high K guys and high Difference The whiff rates for Hernandez, Gordon, and Hicks appear to be lower on off-speed pitches (compared to fastballs), however I am only examining the whiff rates.

How Does Whiff Rate Depend on the Pitch (Pitcher View)?

Pitchers can be represented by a graph that is similar to the one above. Practically all pitchers (with the noteworthy exception of Jake Odorizzi) have better whiff percentages on offspeed pitches than they do on fastballs. Luis Perdomo is remarkable for the fact that his whiff rate on offspeed pitches is more than 0.35 greater than his whiff rate on fastballs, which is a significant difference.

In terms of differential statistics, Clayton Kershaw and Corey Kluber are nearly identical; nevertheless, Kluber has a far greater whiff rate than Kershaw. Jacob deGrom has a high whiff percentage, yet he throws offspeed pitches and fastballs with nearly the same frequency.

Why Does This Matter?

I think that MLB clubs have access to this information regarding batters, and that this knowledge might have an influence on strategy — specifically, how pitchers would throw to hitters. In order to determine whether this is typically accurate, I plotted the Difference (whiff rate on offspeed pitches minus whiff rate on fastballs) versus the fraction of off-speed pitches delivered to the hitter on a line graph. There is a modest positive connection, which suggests that players with a large gap in skill level are more likely to get a larger proportion of off-speed pitches than other players.

Consider the careers of Aaron Hicks, Alex Gordon, and Cesar Hernandez, to name a few.

However, Hicks receives a “large” proportion of off-speed pitches, but Hernandez receives a “low” proportion.

Final Thoughts

One of the goals of this study is to demonstrate the relative ease with which the Baseball Savant database, which serves as a window to the new Statcast Data, may be accessed. Given the present surge in strikeouts (at the moment, 21.5 percent of PAs result in a strikeout), it would appear to be worthwhile to better understand the factors contributing to this high proportion. Having a high whiff rate may indicate that the batter cannot make contact with a fastball, or it may indicate that the batter is swinging at an unhittable pitch outside of the strike zone, which is not the case in most cases.

Modeling Whiff and BABIP Rates from Speed and Spin

The usage of sticky substances by Major League Baseball pitchers to improve the spin rate of their pitches has been the subject of much recent debate. What follows is a logical question: what, if any, advantage does additional spin on a four-seam fastball have on a pitcher’s overall performance? Specifically, in this post, I will examine whiff rates on four-seam fastballs thrown within the zone, as well as how the whiff rate and batting average on balls in play (BABIP) are affected by the pitch’s release speed and release spin rate.

Spin, on the other hand, is the most important component in determining the predicted batting average on balls placed into play.

Data Work

For this analysis, we will use a subset of the Statcast data that includes all four-seam fastballs thrown in the 2021 season through June 14.

We concentrate on the pitches on which the batter swings, and we establish a Miss variable that is 1 if the hitter whiffs on the pitch and 0 if the batter hits the pitch perfectly. Here are some examples of preliminary work that I completed.

  • The release speed and spin rates are on different scales, so to standardize each variable, we subtract the mean from the standard deviation and divide the result by the standard deviation. The standardized variables S Speed and S Spin are named after the standardized variables S Speed and S Spin, respectively. This standardization will make it simpler to comprehend and compare the estimations of the regression coefficients
  • Also, Whilst we know that the placement of the pitch has an effect on the whiff rate — hitters are more likely to whiff on pitches that are high and outside of the strike zone — a pitch x is created such that negative values correlate to inside pitches and positive values correspond to outside pitches for batters on both sides, taking into account that the position of “outside” varies depending on which side is throwing the ball.

The Model

Let us signify the likelihood of catching a scent. For this, we employ a GAM logistic model in the form There are additive components in the model that correspond to standardized values of pitch speed and spin, as indicated by the expression s(a plate x, plate z), which implies that the logit of the whiff probability is dependent on a smooth function of the pitch location (a plate x, plate z). Following the fitting of this model to the data from the 2021 four-seamer, the coefficient estimates and associated standard errors for the intercept, speed, and spin variables are shown.

Errorz value Pr(|z|) (Intercept) -1.676550.01521 -110.2452e-16 S Speed0.170070.0155810.9182e-16 S Spin0.108090.014787.315 2.58e-13 EstimateStd.

Goodness of Fit

In order to determine whether this is a fair model, I compiled a list of pitchers who threw at least 150 four-seamers throughout the 2021 campaign. (I’ll take another look at these pitchers in the near future.) I calculated the observed number of whiffs W, the expected number of whiffs E, and the related Z-scoreZ = (W – E) / sqrt for each pitcher (E) As a result, almost all of the Z values fell between the numbers -2 and 2, suggesting that this model provides a reasonably good match to the data.

Location Effects

The model estimates demonstrate how the whiff probability varies depending on where the pitch is thrown. To begin, let’s look at the average values of speed and spin for the cases in which both S Speed and S Spin = 0. The following is a contour graph depicting the estimated whiff probability over the zone in question. The left side of the graph correlates to inside pitches, as depicted on the graph. The likelihood of whiffing on a four-seamer is low (around 10 percent) for pitches low and inside, but steadily increases as the pitcher travels outside and up in the zone, as seen in the chart.

Speed and Spin Effects

The estimations of the regression factors for speed and spin are modified for pitch location in order to obtain more accurate results. Looking at the output, we can see that the estimations for S Speed and S Spin are respectively 0.17 and 0.11, which is consistent with the data. It follows from this that for each standard deviation increase in pitch speed, the risk of whiffing increases by 0.17 on the logit scale (on the logit scale) (for a fixed value ofspin). In a similar vein, the likelihood of whiffing (measured on the logit scale) increases by 0.11 for every standard deviation increase in the spin of the pitch (assuming a fixed speed).

At (0.67, 2.5), the whiff rate is around 0.20, which corresponds to the whiff rate for a pitcher throwing at an average pitch speed and with an average spin rate.

Because both inputs are standardized, we can compare the coefficient estimates from the two inputs (0.17 and 0.11).

Because the speed coefficient estimate is bigger than the spin coefficient estimate, this shows that speed plays a more essential role in explaining the difference in whiff probability than spin.

A New Metric

With the fitted logit, the general formlogit(p) = location effect + 0.17 S Speed + 0.11 S Spin may be written as logit(p). As a result, the linear predictor (LP)LP = 0.17 S Speed + 0.11 S Spin is calculated. The ability of a pitcher to elicit a whiff on a four-seam fastball after correcting for pitch location is measured in this category. This LP can be used to determine the whiffing ability of a substance.

The 2021 Starters

I compiled a list of the 2021 pitchers who have thrown at least 150 four-seamers between June 14 and June 21. There were 58 pitchers in this group, which equates to nearly two starting pitchers per Major League Baseball team. For each pitcher, I compute the mean standardized speed and mean standardized spin data, as well as the mean standardized velocity. I’ve created a scatterplot of these numbers, which you can see below. LP measurements are shown by dashed lines, which correspond to constant values of the linear predictor (LP).

Jacob deGrom is the top pitcher in the majors according to the LP criteria (no surprise).

  • On the four-seamer speed and spin scales, one can observe how each starter performs in terms of speed and spin. Bauer has an average pitch speed with a significantly higher-than-average spin, but deGrom has a very fast pitch speed with a slightly above-average spin. DeGrom is the better pitcher. Greinke’s four-seamer has below-average pitcher speed, but it has above-average spin
  • A pitcher may compensate for a lower pitch speed by increasing the spin rate of his pitch. For example, deGrom and Cole both have whiff rates that are identical to one another (after adjusting for location). Even though Cole doesn’t throw quite as hard as deGrom, his fastball has a greater spin rate, and their LP values are comparable.

Effect of Speed and Spin on BABIP

We have concentrated on whiff probability; however, a similar modeling method may be used to analyze the influence of pitch speed and spin on the likelihood of hitting a ball in play. With that in mind, let’s talk about the 2021 balls put in play on four-seam fastballs. An analogous model was developed, in which the logit of the hit probability is a function of the pitch location, and additive factors for standardized pitch velocity and spin rates are included. Here are a few highlights from this session.

The red zone represents the batter’s most active region.

Estimate the standard error of the z value Pr(|z|) (Intercept) -0.6520820.018187 -35.8552e-16 *** Pr(|z|) (Intercept) -0.6520820.018187 *** S Speed-0.0049020.018798-0.2610.7943 S Spin-0.0456590.018900-2.4160.0157 * As one might expect, after accounting for pitch location, the projected likelihood of a hit decreases as the standardized speed and standardized spin readings get more accurate.

See also:  How To Flare A Baseball Glove

However, there is a large effect on spin rate.

Each unit increase in the standardized spin rate results in a drop in the hit probability of approximately 10 points (or 0.010) on the probability scale.

Wrap Up – Spin Rate is a Big Deal

Despite the fact that we have concentrated on whiff probability, a similar modeling method may be utilized to determine the influence of pitch speed and spin on the likelihood of hitting a ball in play. In this section, we’ll look at the 2021 balls that were put in play on four-seam fastballs in 2018. Another model, in which the logit of hit probability is a function of pitch location, as well as additive factors for standardized pitch velocity and spin rates, was developed. Listed below are some of the highlights from this match: Here’s how the position of the pitch affects the likelihood of a hit: This is the batter’s hot zone, and it is represented by the red zone.


We can see from the result that the influence of pitch speed is not substantial in this case.

Moreover, by narrowing your attention to a single pitch position and spin rate, you will see that the logit of the hit probability reduces by -0.046 for every standard deviation rise in the spin rate.

Each unit increase in the standardized spin rate results in a drop in the hit probability of approximately 10 points (or 0.0010) on the probability scale.

Sabermetrics Glossary: Called Strike Plus Whiff Rate (CSW)

It is possible to discover prospective breakout pitchers before others do so by using a statistic known as Strike Plus Whiff rate (CSW). The Strike Plus Whiff Rate (CSW) is a new statistic that has emerged in recent years that goes beyond traditional pitching stats to provide a more complete picture. A pitcher’s Swinging Strike Rate is something that almost everyone is aware with. It is just the number of swinging strikes a pitcher receives divided by the total number of pitches they throw. This computation has been updated to account for called strikes, which are strikes thrown that are not swung at by the batter.

  • Given that batters often opt to accept a number of pitches before the ball even leaves the pitcher’s hand, it seems reasonable to create a statistic that rewards those who take a called strike rather than those who swing at the ball.
  • As the season proceeds, we will have monthly updates on this statistic available.
  • Sleepers, cheat sheets, and more To provide some context, the following is a breakdown of all pitches thrown in 2019 as they relate to the CSW rate: 1.
  • On those pitches, the CSW rate was 27.7 percent.
  • The following are the top scorers from the 2019 season.
  • Using the pitch-by-pitch data from Statcast, we will be able to isolate each and every individual pitch for our investigation.
  • Make sure to come back during the season for in-season updates on this, as it might provide you a significant advantage in spotting breakthrough pitchers before the rest of the world realizes what you’ve discovered.
  • Subscribe on the following platforms: Apple Podcasts|Google Play|SoundCloud|Stitcher|TuneIn Jon Anderson writes for FantasyPros as a featured author.

wPDI & CSW: Whiffs

This is the second piece in my series comparing the wPDI with the CSW. For those who are unfamiliar with either metric, I will briefly educate you. Alex Fast of PitcherList examined his website’s pitching statistic, CSW, in last year’s FSWA Research Article of the Year, CSW Rate: An Introduction to an Important New Metric: An Introduction to an Important New Metric. The following is a brief and straightforward definition of the CSW formula: This is referred to as Strikes + Whiffs.

Total Number of Pitches I came up with the notion of the Weighted Plate Discipline Index on my own initiative (wPDI). When we use wPDI, we just ask three questions, or generate three binary events, for each pitch:

  1. Whether or whether the ball was thrown in the striking zone Whether or not the ball was swung on
  2. Whether or not the batter made contact with the ball

Based on the information shown above, each pitch may be categorized into one of six different pitching outcomes. The following is the definition of each of the outcomes: wPDI: Classifying the 6 Pitching Outcomes based on their importance

Outcome Outcome Outcome Outcome Outcome Outcome
Zone? Out of Zone Out of Zone Out of Zone In Zone In Zone In Zone
Swing? Swung On Swung On No Swing Swung On Swung On No Swing
Contact? No Contact Contact Made No Swing No Contact Contact Made No Swing

After that, a weight or an index is applied to each outcome. The following is the formula for the Weighted Plate Discipline Index, often known as the wPDI: If A percent is the percent of pitches thrown in each outcome, then wPDI = Index A* A percent + Index B* B percent + Index C* C percent + Index D* D percentage + Index E* E percentage + Index F* F percent If A percent is the percent of pitches thrown in each outcome, then wPDI is the aggregated, sortable metric.

  • The most striking similarity between CSW and wPDI is that they both use the same denominator – total pitches.
  • CSW is based on statistics from Baseball Savant, but wPDI is derived from FanGraphs numbers.
  • As a starting point, I have divided CSW into two components: the Called Strikes (CS) and the Whiffs (W).
  • In today’s essay, I’ll go through the smells and smells.


First and foremost, in order to concentrate on the whiffs portion of CSW, let us begin with a simple definition. The Whiff Rate (W percent) may be computed using the following formula: WhiffsTotal Pitches (in a row) To begin the process of formingulizing W percent utilizing the wPDI framework, my first and most immediate thinking is to specify the following wPDI parameters: The wPDI (Whiffs Pitching Deficiency Index) is a new pitching outcome index that has just been introduced.

Outcome Description Index
A Out of Zone / Swung On / No Contact 100%
B Out of Zone / Swung On / Contact Made 0%
C Out of Zone / No Swing 0%
D In Zone / Swung On / No Contact 100%
E In Zone / Swung On / Contact Made 0%
F In Zone / No Swing 0%

Pitches that are thrown out of the zone, swung on, and missed are referred to as outcome A. Pitches in the zone that are swung on and missed are covered by Outcome D of the game. According to theory, assuming all data providers are in sync, just adding up Outcomes AD using FanGraphs data should accurately depict the Baseball Savant counterpart. Let’s take a brief look at how similar Outcomes AD is to Whiff percent in terms of performance. The following are the top 20 Whiff (W percent) performances from 2019, as well as the percentages connected with their Outcome A + D (A+D percent) results: percent change from outcome A+D percent change – 2019 Leaders in terms of percentage In 2019, a minimum of 250 pitches must be thrown.

When looking at the top 20 players, the average W percent to A+D percent ratio is 107 percent, and when looking at the overall player population, it is 108 percent.

There is a strong linear link between the two variables.

The point is, why are there an additional 7-8 percent whiffs included in the CSW calculation?

According to Alex Fast, the distinction between the two lies in the foul tips. When calculating whiffs, Baseball Savant has opted to incorporate foul tips in its calculations. The following is a list of all of the whiffs that were created in the year 2019. Whiffs for the year 2019

All Swinging Strikes Foul Tips
Whiffs 88,527 82,030 6,497
% of Total 100.0% 92.7% 7.3%

ORIGINAL RESOURCE: Baseball Savant There you have it! All of the whiffs divided by the number of swinging strikes is 107.9 percent, which is the multiplier we were hoping for. For the time being, I will not engage in a debate over whether foul tips should or should not be included in the computation of whiffs and penalties (and CSW). That will be addressed at a later time, possibly later in this series. For the time being, let’s take Savant’s inclusion at its value and continue working on modeling whiffs using the wPDI framework.

The following would be the regression multipliers for whiffs if all six outcomes (A – F) were taken into consideration: Regression Indexes for wPDI (weighted percentile distribution): Second Attempt

Index A Index B Index C Index D Index E Index F
102.0% 1.3% -0.4% 107.3% 2.4% -1.2%

As was predicted, the results and the AD are the most important factors in the equation. All of the other considerations are insignificant. However, there is one component that I find fascinating. The indices for outcomes CF are somewhat negative, but the indexes for outcomes BE are slightly positive, as seen in the table. According to intuition, the signed findings are correct. The non-swinging outcomes are denoted by the letter CF, whereas the swinging outcomes are denoted by the letter BE.

  • It would seem reasonable to assume that foul tips might hypothetically result from FanGraphs contact occurrences.
  • That was a real eye opener.
  • To the contrary, the observed correlation coefficient between the two variables is greater than 90 percent.
  • For the next regression effort, I will split out outcomes AD from the other outcomes.
  • The related R-Squared is still at a stratospheric 99.89 percent, which is still quite impressive.
  • This will have no influence on the total calculation’s efficacy, and it is still more accurate than using a single multiplier for A+D percent when paired with a single multiplier for B percent.
Outcome Description Index
A Out of Zone / Swung On / No Contact 105%
B Out of Zone / Swung On / Contact Made 0%
C Out of Zone / No Swing 0%
D In Zone / Swung On / No Contact 110%
E In Zone / Swung On / Contact Made 0%
F In Zone / No Swing 0%

DPI = 105 percent * A percent + 110 percent * D percent WPDI = 105 percent A percent + 110 percent * D percent According to the Wformula to Whiffs for individual pitchers in 2019, the resulting X-Y plot is as follows, and it is highly linear – much more so than the called strikes graph:


Let’s finish by putting everything together. In order to obtain the following result, we may combine the so-called strikes weighted PDI regression (wPDI CS) with the current whiffs formula (wPDI W): wPDI CSW: Pitching Outcome Indexes for the CSW World Series of Poker

Outcome Description Index
A Out of Zone / Swung On / No Contact 105%
B Out of Zone / Swung On / Contact Made 0%
C Out of Zone / No Swing 10%
D In Zone / Swung On / No Contact 110%
E In Zone / Swung On / Contact Made 0%
F In Zone / No Swing 90%

A percent + 10% * C percent +110 percent * D percent + 90 percent * F percent WPDI CSW = 105 percent * A percent + 10% * C percent +110 % * D percent + 90 % * F percent That’s all there is to it! CSW may now be completely approximated and represented using the FanGraphs wPDI component data, which is now available for download. In this case, the resultant X-Y plot between the two variables is as follows, producing a 92 percent correlation coefficient: One more point to mention: the formulation of the whiffs section of this equation resulted in far higher accuracy than the regression for the called strikes portion.

Still, there are several additional questions that we hope wPDI will assist us in answering.

  • What is the relationship between outcomes and the effectiveness or deceptiveness of a pitcher
  • Is it possible to make changes to wPDI CSW that will increase the connection between CSW and strikeouts
  • I’m not sure how walks fit into any of this
  • To what extent is the inclusion of foul tips in whiffs beneficial? Do they improve the predictability of CSW? For which pitchers does the difference between Alex Fast’s CSW and my wPDI CSWformula make a significant effect
  • What about the power hitters? In what form does the offensive equivalent of wPDI CSW manifest itself

It is my intention to address many of these concerns in subsequent publications. Keep an eye out for more information.

Leave a Reply

Your email address will not be published.