In an article in early January, I explained the pitching stats we use regularly here at MLBTR. At the end, I briefly noted that I don’t use WHIP outside of fantasy baseball. Several commenters inquired about that choice, so I decided a separate article might be helpful.
If you’re reading this, you probably already know that WHIP is (walks + hits) / innings pitched. Hit-by-pitches aside, WHIP is a measure of baserunners allowed by a pitcher per inning. In 2020, Zac Gallen allowed 25 walks and 55 hits in 72 innings for a 1.11 WHIP. The calculation: (25 + 55) / 72 = 1.11.
In briefly researching how WHIP came to be, I found this fun Wall Street Journal article from 2009 by Nando Di Fino. WHIP was conceived in 1979 by Daniel Okrent, better known as the man who invented fantasy baseball. Okrent originally called the stat IPRAT – “Innings Pitched Ratio.” It was later renamed to the catchier WHIP. Though in his 11-year-old article Di Fino wrote that WHIP “is generally accepted as a legitimate baseball statistic,” he also quotes then-Rays director of baseball operations Dan Feinstein explaining why the team did not use the stat. In Di Fino’s words, this is “mostly because pitchers often can’t control the amount of hits that they give up.”
Sometimes, jamming together a couple of different stats into one can improve its usefulness. I don’t feel that’s the case with WHIP, because of that hit component. I’d rather see info about pitcher’s walks and hits allowed separately, because those are two very different things.
A pitcher’s ability to avoid walking batters is a real skill, and that’s why we cite BB% here at MLBTR. For pitchers with at least 100 innings in a season from 2015-19, the year-to-year correlation of BB% was 0.598. Knowing a starting pitcher’s walk rate in 2018 gave you a decent idea of what his walk rate would be in 2019.
Strikeout rate is even more of a concrete skill. K% has a year-to-year correlation of 0.753. If we know a pitcher’s K% and BB%, then almost everything else was a ball in play. So let’s talk about batting average on balls in play, or BABIP. Pitchers control BABIP to a small extent, and for a starting pitcher the year-to-year correlation is just 0.179. There isn’t that much variation pitcher-to-pitcher in BABIP skill. (As an aside, home run prevention matters as well, which is why we talk about groundball rate as a skill).
Going back to WHIP, its year-to-year correlation is 0.445. To the degree that WHIP is repeatable, that is mostly owed to the repeatability of K% (since a K is never a hit) and BB% (half of WHIP). The repeatability of WHIP is negatively affected by the hit component.
In my opinion, there isn’t a convincing reason to use WHIP. Resident stat expert Matt Swartz sums it up this way: “If the question is how a pitcher performed retrospectively, actual ERA is the more logical stat to use. If the question is how a pitcher will perform prospectively, WHIP doesn’t correlate that well with future ERA, and you can get to a better picture by looking at components.”
So, we’ll talk about what a player already did on a the field, and hits allowed are a big part of that. Trevor Bauer gave up 41 hits in 73 innings in 2021, and it’s a big reason he posted a 1.73 ERA. I’d rather see his walk rate (6.1%) and BABIP (.215) separated out, because I find that more informative both in considering what he already did and what he will do in the future. If I simply told you he had a 0.79 WHIP, that would be less informative.
My goal in this post is simply to explain why I personally don’t use WHIP to evaluate pitchers, and those are the same reasons you’ve rarely seen it on MLBTR in our 15 years. We’re all here because we love baseball. The stats you look at should be whichever ones increase your enjoyment of the game. Whether WHIP, WAR, wins, or something else does that for you, there’s no wrong answer.
Dock_Elvis
I believe Daniel Okrent also believes he and that restaurant bunch invented fantasy baseball as well. Interesting guy…but yeah, no.
cwsOverhaul
Strat-O-Matic was pretty genius way back when. Great game!
looiebelongsinthehall
Loved it. Still remember using my Dave Stapleton card the year after he broke in. I think it was the 80 season so 1981 card set. He was a surprising hit master that rookie season and the rules allowed for 120% of the prior year’s ABs so I had to be careful in deciding when to “rest” him. A lot of fun back in the day.
yick04
Have you played Out of the Park?
Unclenolanrules
There is a Strat O Matic league that has played since the 1950’s or 60’s. They are very particular about who they let in.
Santee Alley
It’s not really up for debate; he invented rotisserie baseball.
13Morgs13
U can use any of these stats(analytics) to guide u. Just don’t make them the be all end all.
realsox
What I’d like to read is an article on which clubs use which statistics to formulate bottom line evaluations of pitchers and position players. The number and variety of available statistics is staggering, and their sheer volume seems self-defeating, at least for the interested fan.
Idioms for Idiots
@13Morgs13
Agree. Others down below came up with a very good point with stats like WHIP and ERA are good comparison stats but not good for predicting the future.
I don’t knock anyone who uses metrics to predict players futures. I can definitely see why Tim Dierkes, Steve Adams, etc. use metrics and eschew stats like WHIP. But there are many of us who don’t use these stats to try to predict their future. Personally I don’t so much because there’s too many players who come out of nowhere to have great seasons or suddenly fall off the cliff.
I’ll state the obvious–it’s much easier to use stats to predict the future of players in their late 20’s and early 30’s than those in their early 20’s or late 30’s/40’s. I’ll use some White Sox players for example.
I don’t think anyone could’ve predicted what Giolito’s done in ’19 and ’20 right after his ’18 season (embarrassingly bad). Moncada was very bad at the plate (especially with the K’s) and brutal defensively at 2B in ’18. In ’19, he wasn’t all that far from being an MVP candidate with far improved defense moving to 3B. TA was a middling hitter (albeit with some power for a SS) batting around .250 and sub-.700 OPS until he exploded in ’19 with the batting crown (and .865 OPS), continuing upon that in ’20. There are several more examples throughout MLB.
My point is it can be rather difficult to predict the future of players in their early 20’s, because you never know when things will finally click (if they ever do). But I’m not about to discourage anyone with that daunting task–more power to you if you want to take that challenge on.
slider32
The hardest thing is to predict World Series MVP’s!!
YankeesBleacherCreature
What I would love to see are Statcast batted-ball exit velocities (barreled- and non-) for individual pitchers and their various pitch types. I’m sure teams are keeping this info to themselves.
mlb1225
You can see all that stuff on Baseball Savant. On a player’s profile, you can change the graph to show different things. For example, Joe Musgrove had a 3.1% barrell % and 88.9 MPH exit velo on his four seam fastball in 2020.
oldmansteve
Baseball Savant is the best site for statcast data
semut
GREAT site!! Thanks for the tip mlb1225
milkman
all StatCast data is readily available. I have recently created a program where you can create your own personal database
ericmvan
You can get that data and then some by searches at Baseball Savant, baseballsavant.mlb.com. (I apologize if just ate up hours per day of your life!)
My go-to stat at the present is one I created myself, pERA, which uses just K, BB, Exit Velocity, Hard-Hit Percentage, Barrell Percentage, and Launch Angle (for larger samples only; in fact, there are 3 different formulas depending on number of batters faced). It has an advantage over Stactcast’s xwOBA / xERA in that it doesn’t penalize pitchers for giving up bloop hits and swinging bunts, where both the EV and LA suck but the combo leads to a lucky hit from the pitcher’s POV.
milkman
there is absolutely no shot your statistic outperforms xwOBA, it takes LA , EV and sprint speed as the main three components, along with an array of others in the equation. it takes the outcome of the batted ball completely out of the equation. …..
chaseturrentine
I love these sorts (two) articles. I’m obviously not as deep into stats as you all so I almost always take for granted a stat’s value in determining the quality of the player. Keep these coming please.
bcjd
+1. I’m not a math guy, but I’m not a dunce either. Articles like this are just right to teach me how to understand and contextualize all these different numbers.
Hammerin' Hank
I’m one who loves advanced stats and still likes the traditional ones. I love your attitude about it and your desire to learn more, as opposed to the hard-headed traditionalists who just want to rail against all new statistics, without even realizing that they could learn from them and maybe enhance their enjoyment of the game.
californiaangels
the winner of the game is more runs (or given up less runs) ….so I still dont understand why ERA isn’t the go to stat. I’ll always look at that first.
AngelsAdvocate
Exactly
mlb1225
ERA is like batting average. Useful when paried with more information. Not all ERA’s are equal. Sure, Eduardo Rodriguez and Julio Teheran had the exact same ERA in 2019, but were way different in terms of strikeout rate, walk rate, exit velo, hard hit rate, BABIP, etc.
revolver
What does it matter if their ERA’s were equal? The point of pitching is to suppress runs, which they both did. How they did it isnt as important.
bcjd
It does matter, though. If one had a low ERA because he had a low BABIP, and the other had a high BABIP but made up for it in strikeouts, then we know the second pitcher’s success didn’t depend on a good defense behind him. Next year, pitcher A might have a worse infield, and is less likely to be able to repeat his success.
mlb1225
That’s like saying “I don’t care if he hit .290. All that matters is that he has a good batting average”. Now am I talking about Mike Trout or David Fletcher?
Hammerin' Hank
HOW they did it gives you an indication of who will be more successful going forward into the next year.
oldmansteve
ERA is a good stat for comparing production, but it is not a good stat for predicting future production as the variables that go into an ERA can vary to no fault of the pitcher.
looiebelongsinthehall
I’d still rather have the pitcher with movement who gets an out using a couple of pitches as compared to the strikeout pitcher who throws six or more pitches to get that strikeout. A perfect game with six or more pitches per batter and he’s out after five or six innings.
JoeBrady
Strikeouts are actually a good way to keep your pitch count down. You face less batters than if the batter put most of the pitches into play.
JP8
Ill be nice but blunt. No, Greg Maddox and the soft throwers that induce weak contact breeze through games. Many batters will foul off those fastballs and run the count up.
Assdribble_Cabrera
Lots of K’s are great and chicks dig the flame throwing strikeout artist, but pitching to contact and letting the defense do their job is the most efficient method for keeping the pitch count down.
Assdribble_Cabrera
For clarification before anyone jumps me, the comment is to compliment what looiebelingsinthehall and JP8 said. Pitchers who control and command, who pitch to contact, which results in ground balls in the infield or fly balls in the outfield.
Hammerin' Hank
I’d rather have the flame-throwing pitcher who strikes out a ton of batters. These pitchers will have much more success through the years than soft-tossers with movement. It’s hard to be great pitching like Greg Maddux. He’s one of the few who could pull it off.
Assdribble_Cabrera
A pitcher can be both with a good mix of pitches and command of the strike zone.
FrankEttingChiSox
Or just RA if you think ‘errors’ are a useless statistic. Assigning blame for certain balls in play to fielders rather than the pitcher when we well know there’s as much to run prevention in the balls the fielders can’t get to as the ones they muff.
Just John
Here’s one way to look at it:
Take Rockie Kyle Freeland’s 4.33 ERA last year. Consider what his results would be if he was instead pitching for the A’s. There are several reasons why we might expect his ERA to differ at another altitude with more foul territory and different players behind him, perhaps substantially. While his BB% and K% may also end up different, there’s a much higher likelihood that these two stats stay closer to the same rates he pitched with in Colorado.. If you’re looking at this site’s content to decide what players you’d like to see your team sign or trade for (i.e., predicting future value) it makes a ton of sense to find better stats than ERA.
HalosHeavenJJ
I’m old enough to have been raised on ERA and still look to ERA+ as a kind of baseline stat. Just like hitting 30 HR in NY is not the same as hitting 30 in LA, giving up 30 runs in NY or Colorado isn’t the same as giving up 30 in LA.
Now that we can break apart how many batters a guy struck out, how many he walked, etc. there are much better ways to give context to his ERA.
looiebelongsinthehall
Mostly anyone with a low ERA in CO can pitch anywhere else.
Rangers29
I think that “projections” are the dumbest thing. You can show me all the stats you want, but in reality, there are so many other variables in play that would go into projecting a player for the following season (and those variables would get very personal too). You might have a string of hitting .247 for four years in a row, but there is no reason to expect something at or around .240 for the next season as well.
dejota
Yes, everyone understands statistical projections are rarely 100% accurate but go on…
oldmansteve
Teams constantly project players performance, that is how they make future decisions. How else would they go about doing this?
Rangers29
I think teams can rightfully look at a players previous stats, but to try and project what they will be going forward is flawed because players are human. Maybe the guy simply won’t be as good? Maybe he went to driveline and is now a star in the making? Projections can’t predict that though the teams can say “Well he’s had a consistent ops and K rate for 5 years now, and he’s shown no signs of slowing down, let’s sign him”. But the teams do that on their own along with the due diligence of talking to the player and getting his work ethic, etc. There’s more in play than what projections can predict.
oldmansteve
Projections aren’t absolutes, they are the most accurate guess people can make. We do that in all aspects of the world.
“Shown no signs of slowing down.” That is what projects do. They look at the signs of slowing down. Is his exit velos dropping? Is he pulling the ball more or less? Is he hitting more flyballs? Is he getting lucky? Projections try to get as much info on a guy they can. They aren’t designed to be accurate. If this were the case, teams wouldn’t make mistakes. But they are designed to be a better road map for future success than past production and this is definitely the case.
dan55
Rangers29 – Yes, projections are not perfect. That’s why they are called projections. Teams that use projections are going to be better off because they will be able to bring in more good players than teams that don’t use projections.
You used the example of a guy who has had a consistent ops and k rate for 5 years to show that sometimes the stats don’t show everything. However, signing a guy who has had a consistent k rate and ops over a long period of time is better than signing a guy who has never had a good k rate or ops. For example, would you rather sign a guy like Masahiro Tanaka(multiple seasons of success) or a guy like Chris Mazza(never had any success)? By your logic, Chris Mazza would be the better signing because projections don’t matter.
Just John
R29:
Q1: Who will have a better ERA next year, Lance Lynn or Jordan Lyles?
Q2: How confident are you in your choice?
Q3: How did you come to that result?
HalosHeavenJJ
It is just like the stock market: past performance is no guarantee of future results.
Just like stocks you try to figure out whose performance will sustain, which assets are under valued, and which are over valued. You’ll hit on some and miss on others. That’s the game.
And that’s why front offices are full of people with financial backgrounds now.
Rangers29
I’m not an idiot @dan55, of course I’d say Tanaka. Though you come to that conclusion through looking at his past success, not a ZIPS projection that says he’ll have a 4.30 ERA. Past success and other non-human factors are what come into play when deciding to sign a player. I’m not saying just to ignore stats, I’m saying that they have to be paired with the non-human factors as well.
And Steve, I mean projections like the ZIPS ones that say that x player will have a (blank) ERA and (blank) FIP. I totally agree with you that teams look at factors like groundball rate and velo to see if they are trending up or down, but those aren’t the “projections” I was talking about.
This might be a misunderstanding, but I don’t think I was talking about the same projections y’all were.
looiebelongsinthehall
Remember the original computer saying, GIGO (garbage in, garbage out).
dan55
Rangers29 – I do believe in ZIPS projections. Overall they do a good job of predicting how a player will perform in the future by looking at predictive stats such as bb rate and k rate. These stats tend to stay the same over a period of time, which indicates that a player can sustain his success.
I was just using Tanaka as a random example. I wasn’t saying that he is necessarily the best pitcher available or anything.
Rangers29
I guess I’m a hypocrite when I say this because *I* take all the data that I can – whether than be trends, stats, injury history, etc. – and put it all together into a table that I use to evaluate players. So you could say I create my own projections. I see where you’re coming from, but I guess I just don’t use the same “projections” that you do.
slider32
Agreed, I look at Frangraphs War over last 2 years, then go to K;s per nine, and BB per nine. Seems to work for a lot of pitchers.
fishco
Kris Davis disagrees with you
MoRivera 1999
Kris Davis batted .200 last year, .220 the year before, .247 before that. So, he agrees.
danny c
yes, yes there is, atleast you’d have an idea of where he’d likely end up in the next year. What do you suggest? teams dont have projections at all? how would you commit to a player long term? based on previous results? that’s factored into a projection system
JoeBrady
That;s quite a bit short-sighted, imo. Trying to project 1 player has a high margin of error. Trying to predcit the entire league’s average has very low variance. Take for example, the AL batting average for the past 5 full seasons:
2015 .255
2016 .257
2017 .256
2018 .249
2019 .253
As you can see, there is virtually no variance whatsoever. If you are projecting for 25 players, you are likely to be fairly accurate.
BlueSkies_LA
All statistics are projections. They are information about the past used to predict the future. Useful statistical models are the ones that predict the future more accurately. Even the best statistical models aren’t deterministic, they are probabilistic.
Lloyd Emerson
Can you repeat your story about two different lefties having radically different arm angles again?
Roll
I would prefer a pitch count per inning to know how efficient a pitcher is especially for starters.
I do think WHIP comes into play a little for relievers but i would prefer batters faces instead of innings pitched. You can get to that number based on WHIP. It gives you a better idea of a reliever as you have to judge relievers different especially middle relievers as they usually come in with runners on and they dont get hit with an er if the inherited runner scores. You can use stranded runners but that would be another calculations.
rovssss
WHIP is about as useful as ERA. No more no less.
It tells you what the results were. It doesn’t isolate what the pitcher actually did.
dejota
This isn’t a hard concept why so many hopping on the struggle bus in the comments lol?
ericmvan
My Dad taught me WHIP in about 1965; who knows when he invented it. I’m sure it had been invented thousands of times, since in those days the only available baseball stats appeared in two long columns in the Sunday paper, and the stats for pitchers were IP, W, L, ERA, H, and BB. There was nothing else you could do with the numbers but add the hits and walks together and compare them to IP!
SoCalBrave
I think WHIP works best as a comparison stat, rather than a defining one.
dejota
Great article, please consider more of these.
semut
I personally don’t care all too much for WHiP, I just think it’s odd to not just include all the stats for the readers knowledge.
Imagine someone not including batting average just because they feel like x___ is more valuable? Just let the reader decide
ilikebaseball 2
This was incredibly helpful. Thank you.
Far Beyond Driven
When it comes to judging relievers, I like WHIP a lot. When you come into the game, are you putting a lot of baserunners on? WHIP answers that question for me.
ASapsFables
WHIP isn’t a good stat to use, “mostly because pitchers often can’t control the amount of hits that they give up.” Guess what? You can say the same about walks since pitchers don’t control the number of the intentional variety that their managers call for.
If WHIP isn’t good because of ‘hits’, then I suppose H/9 is also a substandard stat. Same could be said of HR/9 since ballparks differ. Same goes for BB/9 because of IBB and so on.
When it comes to criticism of WHIP and the others, I say WTF!
Orel Saxhiser
WTF would be a great acronym for a baseball statistic. We just have to figure out what it will measure.
Rangers29
Wins The Fastest, the stat that measures the time it takes a starting pitcher to throw a complete game.
Orel Saxhiser
These days, it takes a starting pitcher two or three starts to throw a complete game. That means WTF would be measured in minutes, hours, days or, for a guy like Ohtani, weeks.
Rangers29
No no no my friend, you take the time it takes him to pitch 1 inning on average, and multiply it by 9. Big brain.
Pete'sView
WTF = How many times a player Walks To First. Thus, a player like Manny Machado would have a higher WTF quotient than a player who simply gets on by virtue of a base on balls.
Orel Saxhiser
Defensive WTF is when an outfielder climbs the wall to catch a potential home run ball that winds up landing well short of clearing the fence.
Granted, this doesn’t happen often, but it’s been a few hours since we’ve seen a story on Marcell Ozuna’s free agency.
BlueSkies_LA
Funny stuff, Cey Hey. Good timing, I needed it.
WTF = Where’s the Field?
A stat to measure DH reaction time when told to pick up a glove.
ilikebaseball 2
They actually explained already why they don’t use BB/9 and K/9. Check it out another great article.
fivetwos
WHIP it baby
WHIP it right….
its_happening
Whip is fine on hot chocolate.
GoLandCrabs
The simple fact it eliminates all balls in the field of play….Useless stat
Oddvark
I understand the logical arguments for why many “old-time” stats are not accurate reflections of a player’s true skill level and therefore may not be particularly good at predicting future performance.
But my brain hasn’t stored the baseline of information about stats like K% or BB%, so when those stats are used to describe a pitcher, I don’t immediately know if they mean he is good or bad or average at those skills. I suppose over time a 22.2% K-rate or a 10.8% BB-rate will mean something to me at first glance, but right now they are practically just random numbers.
On that note, I think it would be helpful if MLBTR writer’s could use more adjectives — like average, above-/below-average, elite, or atrocious — at least on occasion when listing a pitcher’s K-rate or BB-rate (or similar less traditional stats). It might help some of us olds adjust to those stats more quickly.
MetsFan22
Someone here use AVG to tell me Harold Ramirez is better than Nimmo. So yeah.
Mrtwotone
I remember!
phantomofdb
If you aren’t using WHIP because of BABIP, maybe it’s time to train your writers to stop always referring to BABIP as a “luck stat” lol
Overall, I don’t really agree with the assessment here, and I also think you’re a little better off to make projections using the same stats that teams actually use.
BUT, I am glad you took the time to write out your rationale. It’s much appreciated!
HubcapDiamondStarHalo
I always thought there should be a “BIP” stat for pitchers – bases allowed divided by innings pitched. Errors, of course, would not factor into that; walks, HBP, total bases, etc
Oddvark
Sounds kinda like an OPS-Against stat.
Orel Saxhiser
In using WHIP, the problem with tying walks and hits together is that all hits aren’t created equal. Some balls are obviously hit harder than others. WHIP also doesn’t take into account soft-hit balls that wind up being hits.
To make WHIP more applicable, why not tie it to the percentage of hard-hit balls? That way we’ll know more about the quality of the hits given up. The formula would be WHIP times Hard-Hit (or creamed) balls. The acronym would be WHIP-CREAM, which could be a heckuva promotion. At the World Series, two lucky fans would spray the AL and NL WHIP-CREAM leaders with delicious, creamy Reddi-Wip.
stymeedone
Not all outs are created equal. It doesn’t take into account that a hard hit ball can be an out or dp.
Deleted_User
LOL
bravesfan
I’m a big WHIP guy, I use it as a baseline to judge most pitchers much like I use OBP as a baseline to judge all batters. The key is using it as a baseline. WHIP is the OBP for pitchers. It shows me how often they allow players on base which is the main objective of every good offensive team. There is a direct correlation between getting on base and winning games. So it makes sense that pitchers who keep people off the bases would be more successful. But I totally get why you need to review other stats also. But I think it’s ridiculous to not use whip … again, it’s like obp. … any fan knows that’s extremely important
BlueSkies_LA
Not often someone comes here and admits to really being into S&M!
bravesfan
Clever lol
robluca21
I love obp probably to a fault but baseball in its simplest form. Object get to first . So obp till I die
mrgreenjeans
ERA is pointless .. relievers give up runs after starters leave games.. where AS WHIP is an individual stat to determine how pitchers keep guys off the base paths .. just another waste of time article by this site
Orel Saxhiser
Baseball needs to find a way to make pitching wins meaningful. Relievers who blow saves shouldn’t wind up with a victory if their team rallies to win in the next inning. Just have the “W” revert back to the starter. As things stand now, both wins and saves are useless stats. A reliever who gives up two runs while protecting a three-run lead shouldn’t get credit for anything. The dumbest thing in baseball is a manager not bringing in the closer because it’s not a save situation.
Jaysfan1981
I’ve always felt that the best pitcher regardless of IP should get the win.
Scenario 1. SP goes 5 IP gives up 6 ER , releiver goes 2 IP gives up 2, but your offense is clicking and your still ahead 9-8 releiver #2 comes in and shuts down the heart of the order 1 2 3 then gets the next 2 in the 8th but gets pulled for the closer bc of a match-up, closer finishes the 8th and gives up a single with 2 out but gets a pop up to close out the ninth, I say releiver #2 gets the win, he was the most integral P that game, closer gets the save where typically the SP who went 5 and exited with the lead (regardless of his poor performance) gets the win by today’s metrics
Scenario 2. Its game 7 of the WS SP gives up 5 in the first and can’t get 1 out, gets pulled with bases loaded. SP #2 on short rest comes in and strikes out the next 3 keeping it 5-0, goes 2 more solid innings and it’s 5-3 going into the 4th. Because it’s the playoffs you throw your #3 starter out there who’s a #2 on any other team, he gets it to the ninth with a few shakey moments and you’ve tied it up, you go extras, you put in the rest of the pitchers each inning and they all go 123 1 inning each until the 17th, where you win it dramatically. The win should go to the #2 SP because he stopped the bleeding and kept it within reach for the offense to make a comeback whereas the final P in extras is usually credited with the win (although I think the home team offical scorer can award wins to the SP who pitched longest in extras?)
Scenario 3, and this happens often, SP goes 8 or 9 innings and exits after a stellar performance tied 0-0 , not because the other SP was better but because your offense has a bad night. Ends up winning in extras but the win goes to random releiver who might have even given up a run and almost lost it, but the team get a 2 run HR to steal a win. The SP who went 8 strong should get the W
dan55
Agreed. I think the official scorer should get discretion in regards to who gets the win for a game.
tigerdoc616
WHIP it, WHIP it good!
etex211
Limiting baserunners is an important pitching skill. If I look at two pitchers, and all I can see is that one has a WHIP of 0.80, and the other has a WHIP of 1.30, I’ll take the 0,8 guy every time.
All of these guys telling us the stats we’ve used for decades are useless, and we should be using all of these new, made up, intricate stats that nobody understands, are growing very tiresome.
Tim Dierkes
Trying to figure out how strikeouts, walks, and groundballs are new, made-up, intricate stats…
EnjoyCrush
Nice to hear someone else that feels the same way. I think sometimes we forget the big picture and get so bogged down in statistical minutiae. People completely dismiss RBIs now which I find very sad. Driving in a run is an incredible skill and I could argue the most important skill a batter can have since the point of baseball is to score more runs than your opponent.
dan55
RBIs are a bad stat because they rely on team success over individual success. The top three RBI hitters last year were Jose Abreu, Marcell Ozuna, and Freddie Freeman. Now, these are all really good players, but not because of RBIs. Their RBI totals are inflated because they play on good teams. If you took Jose Abreu and stuck him on the Pirates, his RBI total would drop like a stone.
Using RBIs to judge a hitter is similar to using wins to judge a pitcher.
Michael Chaney
I still like WHIP. I basically look at it as the amount of baserunners a pitcher allows per inning (with slight variations because of guys reaching on errors, etc.), and although hits allowed and walks allowed are two different skills, I think it’s important to look at the total picture to understand how many guys are reaching base every inning that someone pitches. The defensive metrics of the defenders behind the pitcher matter so WHIP isn’t an end all, be all stat — but it shouldn’t have been to begin with. I still think it’s helpful.
That said, I don’t like OPS. This site doesn’t really use OPS much anyway (and I like league adjusted metrics like OPS+ and wRC+), but to me there’s too much difference between slugging percentage and on base percentage for them to be grouped together in a single stat. A guy could have an elite walk rate and reach base 37% of the time, but if he doesn’t provide much power his OPS will be relatively low. Or you could be a Mark Trumbo-type where you hit for a lot of power but rarely get on base, and your OPS will be inflated because of a much higher slugging percentage. A .330-.340 OBP means practically the same as a .440-.450 slugging percentage, so why should two stats where the standard is so different mathematically be combined in a single stat?
5toolMVP
W-L is still the gold standard of pitching stats!!
Change my mind.
JoeBrady
In no particular order:
It’s been a few years since I used Whip. My preference is adjusted OPSa. I take a pitcher’s OPSa, and subtract or add the difference between the picher’s BABIP and .300. Using this for Ottavino’s last two seasons, you can see that he didn’t vary that much, even though his ERA tripled. The advantage is that it’s quick and you don’t need a calculator.
That said, there is no such thing, imo, as a single stat that anyone can or should rely on. Every single one of them should be thought of as a clue.
claude raymond
Good stuff Tim. Could you please explain the year to yea r “correlation” you referred to. How is it calculated, etc??
Tim Dierkes
I just used Excel’s correlation function. Basically one column would be a list of pitchers’ 2018 WHIP, and the other column would be the 2019 WHIP for the same pitcher.
BlueSkies_LA
I think what you are measuring is the ability of WHIP to predict WHIP. I understand the value of that as somewhat of an expression of variability, but it doesn’t get you to any idea of whether it’s a useful statistic for predicting a pitcher’s effectiveness. I’d also like to see this broken down between starters and relievers. All the stats are going to be more stabile for starters, if based on nothing other than the larger number of batters they face in a season.
Tim Dierkes
That’s correct. If your goal is predicting a pitcher’s effectiveness next year using one stat, WHIP’s correlation with next year’s ERA is 0.35.
The highest correlation I’ve found is K-BB%, which has a 0.47 correlation with next year’s ERA. Or you’d be better off just citing K%, with a .46.
BlueSkies_LA
Neither of these are very strong correlations, and to me it only points out the high degree of variability in all of the stats from one year to the next. I’d like to see this presented in a scatter chart with multiple years of data. The distribution around the mean is bound to be tighter, I think,
Tim Dierkes
I think it points out that ERA itself is so noisy, that predicting it has a certain ceiling.
Plenty of pitching stats correlate well to themselves year-to-year, suggesting they are controllable skills:
GB% – 0.781
K% – 0.753
BB% – 0.598
That’s why we focus on those three. There is always going to be part of ERA we can’t predict because of the variability of BABIP and HR/FB.
BlueSkies_LA
You make an excellent point. I’d like to see more work in this area.
Now how about those defensive metrics? I’ve always had a problem with them. Even with thousands of data points every season, they are still way too noisy to be trusted.
claude raymond
So a correlation of 1.00, for whip for example, would mean consecutive years with identical WHIPs?
Tim Dierkes
Yes. The closer it is to 1, the stronger the correlation between the two variables.
claude raymond
Got it. Thanks
goob
Thanks for giving me a new perspective on WHIP, TD.
I won’t just “WHIP and chill” as much anymore.
Matter of fact, I’m finally gonna try out that FO service of yours.
Thanks again.
Cap & Crunch
Whip is a perfectly fine measure to use- I wish it would include hbp and discount IBB but those are so minor you get a pretty good initial read from just using whip
Ofc its not an end all be all but its going to lead you on the right path from the outset if you look over multiple years
Another oldy but goody is simple k:bb . Statcast ofc gives you a whole other set of variables for deeper dives
MoRivera 1999
I don’t really care if WHIP is predictive. Sometimes I just want to know what has happened. Yeah, I would prefer it if WHIP excluded intentional walks and included hit by pitches. And I might like to see a version that used Total Bases instead of Hits, as someone suggested earlier. But in lieu of those changes I will still glance at WHIP as a quick indicator of what a pitcher has done, even if it’s a weak predictor of what he will do. Just as I continue to look at ERA even while using ERA+.
MikeyHammer
You can’t have a pie without Cool Whip. – Stewie G
Sabermetric Acolyte
Your article gives a very convincing argument but I’d like to counter by suggesting the problem is looking at all pitchers the same way. I personally find WHIP is a great way to look at relievers while saying very little about starters. Which I personally think is the better use of the stat. A reliever throwing 50 innings in a season versus a starter throwing almost 200 innings is going to see a greater spike in their ERA due to a single bad outing and relievers are more likely to be kept in the game during a bad outting if the goal is to save pitching for the next day. While the same could be said about WHIP if you remove bad outings from a reliever and relook at their numbers then you’ll find ERA is a more susceptible stat.
But to each their own. One last thing to point out, no single stat works on its own. One of the things most misunderstood about statistics in general is they require context and some times the context is other statistics.
mjones650
The authors rationale absolutely makes sense. WHIP is so 2009. Now we’re talking about barrels and hard hit contact. With better tracking we can come up with better metrics, and we are!
Pauly2112
Have you ever noticed Sabermetricians always pontificate about twice as much as required?
Tim Dierkes
I guess? I wouldn’t know…I’m just a dude who thinks strikeouts, walks, and groundballs tell the story.
reflect
If you don’t use a whip how do you get to work?
SwingtheFNbat
Use WHIP, you should use it, and don’t let anyone tell you that it’s not useful. This is a stat just like any other good stat, There all best when put next to each other.
Just because this site hates on WHIP and doesn’t use it, doesn’t mean u have to hate. USE IT! JUST LIKE ALMOST EVERYONE STILL WILL.
dkhits20
“Going back to WHIP, its year-to-year correlation is 0.445”
Hmm… that doesn’t seem right. Unless I’m misunderstanding.
Tim Dierkes
Why do you think it doesn’t seem right?
dkhits20
Are you using the MLB average? If “1” means a positive correlation then .445 would mean that league average WHIP changes pretty drastically from year to year, wouldn’t it? Seems like it would be closer to .90, but again, maybe I’m misunderstanding.
Tim Dierkes
No, this is looking at any individual pitcher who pitched 100 innings in back-to-back seasons. So it’d be running the correlation of Zack Greinke’s 0.84 WHIP in 2015 vs his 1.27 WHIP in 2016, getting a sample of 353 players from 2015-19.
dkhits20
Ah ok. Makes more sense now. Thanks Tim.
LetGoOfMyLeg
In my mind what all this is avoiding is the actual batter vs hitter stat. WHIP is a comparison between pitchers; BA for example is a comparison between batters.. A particular pitcher may have a higher WHIP simply because of the batters he faced and the circumstances therin.
Dr Obvious
Ugh… No individual stat in a team game is ever going to be perfect.. Completion percentage for a NFL QB doesn’t count drops, but to totally avoid a stat that is a solid guide is dumb… How about eliminate ERA since bad fielders can’t get to more balls or eliminate wins / losses for teams since BABIP went against a team? I can pull stats that prove Dodgers didn’t win 2020 WS if we want to go that route
Tim Dierkes
We’ve got basic, important counting stats for pitchers. Two of them are walks and hits.
WHIP advocates: Jam those together into one stat! It will tell you less about the pitcher, but it will…I dunno, save you time somehow? Come on, we’ve been doing this since 1979 for some reason!
Me: Don’t jam them together
Dr Obvious
I am fully OK with them being separate. If I misunderstood that your view was that – I apologize.. I thought you were against the general concept of both combined as being valid..
Tim Dierkes
No need to apologize. I actually am against the general concept of combining hits and walks into WHIP, yes, for the reasons explained in the article. But, only against it in the sense that I’m not going to do it.
Hammerin' Hank
But ERA is a fluky statistic as well. I think WHIP is a little better at determining future performance. It’s basically measuring OBP allowed, minus the HBPs of course.
BlueSkies_LA
This is also a pretty solid argument against WAR, in its various flavors. It’s the ultimate kitchen sink stat, a complex formula full of all kinds of assumptions. It makes WHIP look like driven snow by comparison, and I don’t know that it has ever been shown to correlate closely to team wins, the very thing it is supposed to predict.
nbresnak
I like the WHIP stat and it is NOT obsolete. I do like seeing newer stats to use as comparison as well too.
As a math teacher, I like looking at and evaluating all different types of statistics but not to push WHIP aside, this article actually makes me like the WHIP stat, not the opposite as it seems to have intended. K% and BB% is a solid stat as well.
For all the sabermetric stats out there, which are good for evaluating, actually watching and seeing the talent is more important! Sabermetrics should be used as another source involved with the whole process, but as a secondary source or used when evaluating 1A and 1B type of players.
Hammerin' Hank
WHIP may have its flaws, but so does ERA. I like them both, though, although I’m glad we now have the advanced measures like FIP and ERA+.
jorge78
“Pitchers can’t control the amount of hits they give up.”
WHAT THE FUDGE! That’s only their job-and they’re blameless?
Please tell me no major league team has come within a mile of this “genius.”
Hammerin' Hank
Guess you missed that famous study done years ago by Voros McCracken that demonstrated this to be basically true.
Moneyballer
Cool article! I think you are right here. Perhaps the inclusion of whip into fantasy baseball metrics has given the average fan a false sense of statistical importance. When you really break it down, I agree, you can get a much better picture of a pitchers value by digging deeper into seperate metrics. Let’s say you have a relief pitcher who for some reason has a propensity to allow base runners maybe his stuff is just a little on the wild side…the result would be a high whip…but the same pitcher also has the ability to strikeout hitters at a high clip getting himself out of the jams he created. This pitcher would have a high whip, high k% and a low era. The whip stat alone would say he’s ineffective but everything else would say the complete opposite! So there it is.
dougsolo1
But if a good pitcher gives up less hard hit balls, wouldn’t that correlate to less hits (and thus a lower WHIP)? I don’t agree with the premise that a pitcher has no control of what happens once a ball is in play. If one pitcher gives up more hits because hitters always hit hard line drives against him, while another gives up less hits because he is tough to square up, why wouldn’t that be reflected in WHIP?
claude raymond
I’ve become a “per nine inning” guy. I think that comes from 30 years of stratomatic where I’d look at a pitcher’s card and mentally calculate his per nine inning rates.
Give me walks per nine inning and ks per nine innings. From that I get walks/k. So I then know 3 rates that are meaningful. I also can then guess how many pitches are put into play. Now give me hard hits against the pircher and I think thats enough. Whats the acronym for that again Tim?
Last thing, many great pitcher’s, HOFers and active, have given up a lot of homers so I dont really care about homer rates. They gave up a lot of solo shots. Niekro comes to mind. I’d listen to games and the announcers would mention Robin Roberts when Niekro pitched. Roberts gave up solo shots.
Anyway Tim , please refresh my memory regarding hard hit balls against acronym and ill leave you be. Btw, kudos to you for these dialogues
ffrhb14Sox
The only problem w per nine inning numbers is nobody goes 9 innings anymore unfortunately. Per 9 was one way of equalizing numbers with different innings pitched and it made sense going to per “game” numbers. Going to per inning is also a way for fair comparison and closer to what pitchers do these days.
claude raymond
Thanks but the rates per nine are not meant to mean the pitcher will go nine. Just like ERA is earned runs per nine. The stats are meant to tell how the pitchers would look if prorated to nine. A relief pitcher with a rate of 13.5ks per nine is the same as saying 1.5ks per inning. I just like seeing it related as per nine.
A pitcher with an ERA of 5.50 doesn’t mean he’s throwing 9 innings. And giving up 5.5 earned runs. Dude , I know few pitchers go 9 innings anymore and they sure as heck can’t give up 5.5 earned runs
Moneyballer
So you don’t WHIP but the big question is can you NAE NAE?