FIP IS THE MLB’s NEW ERA: The Best Pitching Statistic

[@grahambutterfield7450]

Great video Jake! Love being able to learn more about different stats and being able to expand a deepen my baseball knowledge!

[@SimpleSabermetrics]

Thanks for the comment, Graham! I’m glad you find these types of videos are helpful.

[@wi54725]

As a former employee of three different Major League organizations, I will admit that FIP was an improvement over things like linear weights, but it also quickly had to be altered due to its weaknesses. 1. Hit by Pitch was left out, and there are enough HBP to matter. Bruce Kison's stats change immensely. 2. Ballparks affect the home run surrendered. If a left-handed pitcher plays for the Red Sox, he may give up weaker fly balls than a left-handed pitcher playing for the Yankees, but the Yankee pitcher will have better home run numbers strictly on the massive difference in left field power alley lengths. 3. There was a quick alteration for knuckleball pitchers that did not stack up well in FIP. But, then the thought was that other off-speed specialists, and stars like Mariano Rivera, might be hurt because they were much better at making hitters hit balls that were classified as soft by BIS, whereas other flamethrowers gave up more hard hit balls and more fliner liners, which tended to drop in for hits at a much higher rate than any other type of batted ball. Hard hit fliner liners carried something like a .730 batting average one year that I saw the stat. Then, an astute analytics intern came up with the answer that these "junk ball" pitchers tended to coerce a lot of batters to hit a higher percentage of short popups--the type that would qualify for the infield fly rule. Popups short of Texas League territory are almost converted to outs at the same rate as strikeouts. There are only a small number more dropped popups in fair territory as there are strikeouts with passed balls that allow runners to get to first base. So, Popups needed to be included in the formula. 4. The constant is constant for all pitchers, so it artificially lessens the variance in the stats. Forget trying to make it look like ERA, and allow the result to stay as is. 5. Finally, IP should not be used as the denominator. The stat tries to calculate effectiveness, so it needs to be calculated per plate appearance. Of course, now pitchers are evaluated with Trackman and similar technologic breakthroughs. FIP has become like many other analytics a 21st Century RBI. Tango Tiger does deserve a lot of credit for creating this stat. FWIW, the managers of every team I worked for in that period all had "The Book" on their desks as one of three must-haves (The Champions' Mind and The Power of a Positive Team).

[@SimpleSabermetrics]

This is an AWESOME comment! Thank you so much for sharing some of your experience with the community. At the end of the day I couldn’t agree with you more - the use of tech to evaluate players far outweighs that of any single statistic out there. Is FIP a step in the right direction? I’d say yea. Feel free to check out the rest of the videos on the channel more heavily focused on the player development and technology side of the game as well!

[@sammalec8467]

what is the name of this stat?

[@lynndavis6872]

Glad I found your channel I know what to look for in a bad pitcher

[@SimpleSabermetrics]

There you go! Thanks for the comment.

[@cedricgist7614]

Following the emerging analytics from Bill James up to the present has given me a different eye on the game past and present. Your videos are in the vein of the informative and enlightening content I have strived to digest over the decades. I like the game of baseball but don't find myself as much of a spectator or follower as I was when younger. But I like the numbers and I constantly remind myself that numbers don't tell the whole story.

[@SimpleSabermetrics]

Absolutely, the game within the game has been made utilizing the latest and greatest stats and player development technology. But there has been, and always will be a human element to the game. Incredibly honored to be delivering the high quality content you’ve been searching for. Let me know if there’s ever a topic you’d like to see covered in a future video!

[@johnnywholestaff4909]

Thanks for shedding light on this. FIP is better than what some normal fans realize.

[@SimpleSabermetrics]

You know it! Even threw in how to calculate the constant for you.

[@mikealaskavo]

I'm new to baseball and diving into understanding stats. I love the depth of it all! I have a question about the equation. Where does the 13 come from in "13*HR", and where does the 2 come from in "2 - K"? :)

[@SimpleSabermetrics]

Great question, Michael! These numbers were put in place by the creator of FIP (Tom Tango) and are based off of the average run value of each event occurring multiplied by 9. He found these run values by calculating the average runs scored after the point of the event (I.e. single, double, triple, home run) to the end of the inning from the years 1974-1990. If you’d like further reading on this, you should check out the article here: www.google.com/amp/s/www.athleticsnation.com/platform/amp/2010/4/27/1446531/statistically-significant-fip Thanks for the awesome comment!

[@madams3478]

I like the fact that BB is added in multiplied by 3 whereas K is subtracted out multiplied by 2. Because even if I put the ball in play, there’s still an excellent chance I’ll get out.

[@briangeller5193]

I just love these videos - great job! For someone who is a newbie and looking to learn more about data would your suggest learning R first? SQL?

[@SimpleSabermetrics]

Great question, Brian - I’d recommend diving into R first. I’ve got a few videos explaining some resources you can turn to (links below) and a blog that one of our contributors does a bunch of introductory tutorials on here: www.simplesabermetrics.com/blog/categories/sam-bornstein Thanks for the comment! Videos: THE BENEFITS OF CODING FOR BASEBALL | With Sam Bornstein kzbin.info/www/bejne/jJTIY2R_rbOBjKc 3 SKILLS TO LEARN NOW TO HELP ADVANCE YOUR CAREER IN BASEBALL kzbin.info/www/bejne/eWTYmZl4psydjbs ANALYZING BASEBALL DATA: A Skill Jobseekers Want kzbin.info/www/bejne/jWiqXnSKnNmKbaM

[@winningmessenger8817]

do you have a plan for player projection system? thanks always!

[@SimpleSabermetrics]

Great question - this is definitely something I’ll look into to share with you all. So stay tuned!

[@winningmessenger8817]

@@SimpleSabermetrics Thanks, I have good interest in this part and you as a good relationship with this considering your career of player development. Thanks again.

[@izanmaciasgallardo1658]

hi! is there any kind of FIP that takes in account only 1 pitch?

[@johnnysmoke612]

Getting on base, though required, only contributes a 50% win probability and a hit with RISP runs a 70-80% win probability. Your example, for the single to home run sequence is a measure of win probability. Only the home run is used in FIP, but not the singles. Measuring a pitcher's sequence pitching for win probability would be a better measure. ERA is not only measuring earned runs allowed but also has a strong correlation to how a pitcher pitches with RISP. The pitcher that gives up a HR, single, single has given his team a better win probability than a pitcher that gives up a single, single, HR. For years, the argument against RISP hitting or pitching wasn't called clutch, but luck. However: there is no way to get around the correlation for hitting or pitching with RISP and it's very strong correlation to win probability. The team that gets the most hits with RISP wins roughly 70% of the time and to circumvent that fact is the solo homer or runner on 1st base only homer contributing only 30% to win probability. If a team leads in both RISP hits and solo homer or runner on 1st base only homers that increases their win probability from 70 to 80%. The other 20% in loss probability is when the opponent can lead only in the solo or runner on 1st base homers circumventing the RISP leading in RISP hits. The other 10% difference is when a team has less hits with RISP but drove in more runs with their RISP hits. But again, the most hits with RISP has the larger impact on winning 70%.

[@sirql2246]

On D1 stats, I need to look FIP or xFIP ?

[@royrowland5763]

I would argue that Wins is a stat that has some merit. Yes, it is useless over a short span, like a few seasons. However, if a pitcher goes 18 seasons, and he has had a lot of different names in the lineup to back him up over the years, then accumulating a lot of wins is probably not due to luck. I would be curious to see how many pitchers with 250+ wins had terrible FIP. Probably not many.

[@willianlopez787]

How do we get the constant numbers for HR, BB+HBP and K (13, 3 and 2) on the FIP formula?

[@johnnysmoke612]

Another thing missing in baseball board game design is the winning is associated more with RISP hitting or pitching than platooning, home and away, ballpark effect, and defense. Of course, the argument always comes up that RISP is a small sample with larger variance swings. Ok, there is no way around a team that hits .250 overall and .300 with runners on third or scoring position (SF counted as at bats) isn't at a disadvantage against a team that hit .275 overall and .300 with runners on 3rd or scoring position. The accuracy of the game is compromised. It's like a race being started with the pop of a gun, but one rider has to wait 5 to 15 seconds before taking off after the pop. LOL! This is why when analyzing pitching and hitting it is very hard to do unless viewing the outcome as the batter/pitcher both contributing to the result. If the pitcher is only responsible for the HR in the sequence of hits that would mean the batter or the pitcher defense is controlling the two singles of the sequence beyond the pitcher's control. That is like saying the pitcher in the way of control isn't responsible for any batting order sequences for allowing his homers. Reminds me of gopher ball pitching Robin Roberts giving up a ton of homers, but just giving up an inordinate amount with the bases empty. So, FIP is telling me Robin Roberts is controlling giving up his homers before his two walks are allowed instead of the opposite way of giving up two walks then a home run.

[@bobconrad578]

Well done.

[@SimpleSabermetrics]

Thanks for the comment, Bob!

[@Writeous0ne]

How can we use FIP to figure out if a pitcher is unlucky/lucky? If a guys FIP is lower than his ERA would you conclude he's been unlucky? and vice versa? Like, if a pitcher induces a groundball and the infield gets it, throws it but the 1b makes an error, how would this be represented in FIP?

[@DEEPGNOSTIC]

Would have been helpful if you were to discuss those weights in the FIP formula....they seem rather arbitrary

[@alkukayen4150]

2:30 wait but if a pitcher allows a hr after 2 singles, he did a worse than a pitcher who gave up the hr first. you pitch different with runners on base.

[@SimpleSabermetrics]

Absolutely! Which is why this stay alone isn’t the only way to analyze pitchers performances. FIP is better than ERA, that’s for sure - but there is definitely still room for improvement!

[@dukedematteo1995]

I love FIP....great stat. The problem is some pitchers just consistently outperform their FIP. Mariano, Glavine, Jim Palmer, Jimmy Key, the Neikro brothers to name a few. Verlander in recent years.

[@thomash.l.9382]

So a guy giving up doubles all day is equal to a guy giving up singles? Wouldn't it be better to use DRS to see how good the fielding is or the expected fielding %? So every pitcher playing with the same fielders behind them

[@MrHOTPEANUTZ]

You still have to factor in the mental makeup of the pitcher. You can't say 2 hits and a homerun are equal despite the sequence. If a pitcher doesn't respond well under pressure they are more likely to give up a home run in key innings or with men on base, whereas a pitcher like Randy Johnson will strike out the next hitter stranding the two men on base. I would assume someone like Bob Gibson would motivate his team to play better defense behind him versus the effect Carlos Zambrano has on his team. No different than hitting a home run when you are up by 12 runs vs. being down by 1 run. Both home runs, but the ability to perform during high-pressure situations has to be accounted for in sports.

[@soysauske5768]

FIP is good but doesn’t it favors K’s a bit too heavily. What about a pitcher who gets a lot weak contact but less K’s?

[@SimpleSabermetrics]

This is a good point - as the stat only takes into account a pitcher’s walks, strikeouts, innings pitched, and home runs. As for your argument on pitchers who pitch to soft contact, the fact that HRs are weighted at a 13 times multiplier really benefits pitchers who limit the long ball. Every stat will have its flaws, but at the end of the day this is one of my favorite stats for evaluating a pitchers performance separate of the positional players behind him! Thanks for the comment!

[@soysauske5768]

Simple Sabermetrics that’s a great point if u let up a lot of home runs then u r not getting weak contact

[@Il_Exile_lI]

@@SimpleSabermetrics When looking at a guy like Mariano Rivera, his FIP is a lot higher than ERA. Rather than attributing this to team's defense outperforming his FIP, it stands to reason that FIP is undervaluing a guy who excelled in breaking bats rather than striking guys out. FIP has value, for sure, but in terms of finesse pitchers, even accounting for the home run weighting, it can undervalue them.

[@JayTemple]

My quibble with FIP is how they normalize it. Here's an analogy in tests. If the top score is 86, you can treat 86 as the basis (that is, a student who got 71 on the same test would have a 72/86 or about 83%), or you can add 14 points to everyone's score (so the second student above would get 71 + 14 = 85%). By subtracting a constant instead of multiplying by one, you get the occasion FIP that's negative!

[@alexandera2586]

I didn't think about the pitchers lacking control in giving up home runs, that is smart. Let me double down, play devil advocate sort of. Could the walks portion be included, since the pitcher does not have any control in calling balls or strikes? Like the strike zone is different for every umpire and for different batters (due to their height). Plus I read that umpires favor the home team more, which makes matters worse in regards to this. Let me know what you think.

[@SimpleSabermetrics]

All of those points are very valid! Breaking down these statistics to avoid the human error involved with evaluating players is crucial! Especially for players entering the league for the first time with a smaller sample size... I love these types of content! Question everything, you’ll always get better answers a couple layers deeper.

[@rico993]

I think your point about the umps would make FIP poor for evaluating a single game, but over the course of a season it will balance out. It may not technically be a truly accurate number, but it works for the purposes of comparing one pitcher to another.

[@steve_etzel]

The problem with the simple forms of FIP is that it assumes the pitcher has no influence on what happens to the ball once it is in play (excluding home runs). The corollary to this is that BAbip should be around .290 or .300 for all batters, and it isn’t.

[@APBAChatter]

What pitching stat has the best correlation to winning?

[@hkgcgsdhjgd]

You’re asking the right question. At the end of the day, the only thing that matters is your team’s win-loss record.

[@seanyboyclutch]

does WHIP count an errors made? in other words if a baserunner gets on with an error made by a fielder would that number be in WHIP? I think not right because error is technically not a hit.

[@SimpleSabermetrics]

WHIP does not take into account errors, it is simply the measure of Walks plus Hits per Innings Pitched. Thanks for the comment, Sean!

[@elianyrodriguez2432]

thanks from cuba

[@SimpleSabermetrics]

Thanks for watching, Jorge! Always great to hear from my international viewers!

[@wowihaveachannel4862]

Sounds like this FIP basically tells me how often balls are hard hit against a pitcher. It's good to see analytics being used in baseball but one thing I will say for anyone who's played. Every at bat is unique and independent . I could hit 4 lazers to the outfielders for outs and I am 0 for 4. Meanwhile the pitchers era is 0.0 and batting average against is 0 too. But he got lucky that's all. I prefer stats that tell me how a player impacts his team , either hitting or pitching . Wins isn't perfect but no great pitcher has ever had a losing record (staters). Relievers maybe a few less decisions.

[@darkninja___]

Why don’t they just make this simpler by using ERA but then multiplying it by a coefficient. The coefficient would be calculated by measuring the pitcher’s team’s defensive abilities on average compared to an average team’s defensive capabilities. So, for example a bad defensive team has a value of 0.8 for example and a good team one of 1.2. Idk exactly how they would quantify this value but I imagine it wouldn’t be that hard since errors are tracked. Might not be super precise but seems simpler.

[@darkninja___]

The main problem I see with this is that not all errors are equal and some hits would be ruled a hit and not an error even though a great fielder would catch it.

[@edwardwood3622]

Why doesn’t FIP include; wild pitches, balks, and runners picked off?

[@MarshallOwens]

How about SIERA?

[@SimpleSabermetrics]

SIERA is another stat that should be brought on the channel in the near future. Thanks for the comment!

[@joespinosa3272]

Hits are based more on the pitcher than the fielders. While I believe that scoring for errors is too lenient, I don’t like that FIP completely focused on the 3 true outcomes. Foolish Baseball created new stat based on hard hit %

[@JADiaz10]

I absolutely hate FIP. That stat tells you a guy like Alex Colome is lucky. Which is bullshit. You can’t get “lucky” for 5+ years. It provides context to ERA but it should never be the main stat to evaluate pitching.

[@SimpleSabermetrics]

You’re not entirely wrong on that, JADiaz10. In my opinion, this stat does a better job than statistics like ERA or BAA - but it isn’t perfect. There are a ton of different performance indicators that do a better job highlighting the best of the best like CSW%. I’ll be sure to do a video on some of those stats in the future. I appreciate your feedback!

[@rwwilson21]

Okay, If wins and loses don't matter, let's just take way the Cy young award. you know the award that was named after the pitcher who WON over 500 games.

[@tubbyidk1474]

😂😂😂. good for him. you do know that Wins and Losses mattered when the award was created in 1956. theres a foolish baseball video on why wins/losses do not matter, check it out. or this comment could be bait that i fell for

[@walterbison]

Wins and Losses mattered when pitchers were throwing complete games. It's only irrelevant today because today's pitchers have a hard time lasting 5 innings.

[@hkgcgsdhjgd]

Even then, a pitcher’s wins and losses are heavily dependent upon the amount of run support they get. One pitcher can lose a game 2-0, and another one can win a game 8-7. Is the second guy better even though he gave up 5 more runs?