Pitching staff xFIP

[Ennder]

xFIP is a stat that tries to pull luck, park and fielding out of a pitchers stat line and show how good he really was. Personally I'd value it more than ERA for a single season but it is flawed in its own way and ERA is probably a better long term stat when you have 3+ years worth of data. keep in mind RP's tend to beat their xFIP's for whatever reason.

 

 

Shouse - 3.66

Gallardo - 4.00 (that start in colorado really killed his stats, I expect better)

Gagne - 4.19 Tex, 4.21 Bos (yes basically he was lucky in Texas and unlucky in Boston)

Torres - 4.21 (yes he wasn't near as bad as that ERA suggests)

Turnbow - 4.23

McClung - 4.24 (small sample)

Mota - 4.32

Capauno - 4.39 (there was a lot of bad luck in his season)

Bush - 4.43

Sheets - 4.49 (yes his stats were ugly last year which worries me way more than the injury issues)

Riske - 4.56 (he has a long history of beating his xFIP by a lot)

Parra - 4.61

Villanueva - 4.84 (way too many walks last season, this is 2nd year in a row he beat his xFIP by a lot though)

Wise - 4.84

Suppan - 4.85

Vargas - 4.89

 

Note to self, this post scares me. When I started this my idea was to show that we pitched much better than ERA suggests but I'm not so sure that is true. I think a lot of our guys have upsides better than the xFIP. Sheets, Gallardo and Villanueva for sure do. What I do like is that most of our guys are at least in that 'average' range, there are no really ugly guys in that list.

[mothershipconnection]

Not pretty. I can't believe we have all of one pitcher under 4.00.

[logan82]

So the conclusion could be that our defense didn't effect our pitching as much as we had at first thought?

[And That]

The flip side to this is that improvement by Gallardo (just by progression) and Villanueva (he walked a lot early in the season... remember when we were nervous about him in April?), as well as a healthy season by Sheets (if it ever happens) would improve 3/5 of the starting innings. I also have to think one or two of Gagne/Torres/Mota will improve, and Riske keeps outperforming the more detailed stats (he has a lot of weird peripherals year after year, but it works).

[rluzinski]

The defense certainly hurt Cappy and Suppan, two guys who need decent defense to do well. Also, I'm not sure how relevant a year's worth of FIP is for a reliever (although, it's certainly better than ERA).

[sbrylski]

But xFIP generally regresses everyone's preformance a bit to the mean. So I think its fair to say that at half of those guys will beat their FIP's, and some by more than a half point of ERA (and, of course, the other half the other way). So I think we'll see four or five guys solidly in the 3's, a bunch in the low to mid fours, and then a couple blowbags in the 5's. Because we're so deep, we hopefully can limit the innings of those 5's and use our top guys in the critical situations, we could have a nice low 4's team ERA. 4.10 to 4.40 would be an improvement from last season (4.41), and place us solidly in the 2nd tier.

 

I think that list looks fine, doesn't have me worried. Someone who has more time could compare it to a list of the Cubs pitchers.

[Ennder]

Oh I have time baby!

 

Prior - 3.56 (That was in 2005 when he was last healthy)

Marmol - 3.63

Wuertz - 3.82

Howry - 4.15

Dempster - 4.15

Hill - 4.17

Marshall - 4.56

Zambrano - 4.65

Eyre - 4.92

Marquis - 4.98

 

We had as good of a pitching staff as the Cubs last season, I had that discussion with them on their forum. They had a much much better defense and Wrigley played as a pitchers park last year but our pitching as a whole was as good as theirs. Unfortunately that is hard to prove to the "common fan" since they just look at ERA and decide it is cut and dry based on one of the weakest stats out there.

 

 

Lilly - 4.39

Wood - 4.52

[Tbadder]

Is there a stupid person's explanation for this stat somewhere? Thanks in advance.

[TerraceReserved]

FIP itself is short for Fielding Independent Pitching, and it uses a pitchers walk, strikeout, and HR rates and applies an average hit rate and creates a number on the scale of ERA that shows how well they pitched with no luck involved. xFIP I believe adjusts for the pitchers HR/FB ratio.

[logan82]

Looks like pitchers who relly on ground balls would always have a high FIP.

[Fatter than Joey]

So the conclusion could be that our defense didn't effect our pitching as much as we had at first thought?

 

No -- Our defense hurts our pitching compared to the Cubs.

 

Zambrano - 4.65

 

Zambrano seemed to either suck or was brilliant.

 

Basically FIP is a stat where you try to only incorporate stats the pitchers control, which would be HR, BB, HBP, K -- basically the 8 guys behind the pitcher have nothing to do with HR, BB, HBP or K. You then take HR+HBPK multiply each by a factor and then divide it by IP, to give you an ERA type stat in format.

 

Where defense comes into play, is in the IP variable -- If Braun/Fielder make less plays or more errors than Ramirez/Lee a pitcher is going to pitch less innings. Basically a bad defense is going to get less IP out of 100 pitches than a good defense will.

[Ennder]

Yeah I prefer qERA a lot more which includes things like GB% but I couldn't find a listing for it so couldn't use it.

[rluzinski]

Looks like pitchers who relly on ground balls would always have a high FIP.

 

When you say, "Rely on groundballs", you mean, "strikes no one out", he's going to be in trouble, sure. A strikeout is an out virtually 100% of the time. Even a weak groundball is converted into an out less often. In actuality, there are many good groundball pitchers out there who have excellant FIPs. Maybe, they have a mediocre K rates but they supress HRs and have good command and low walk rates.

[Patrick425]

Yeah, I don't really get Fip at all. Lets bacically just look at Walks, HRs, and K's...huh? Hate to sound like an old curmudgeon (by the way, I do understand and agree that OBP is a better measure than average, so I'm not completely stuck in the past.). However, whats wrong with ERA? If a pitcher gets a guy to hit an easy ground ball to the shortstop, isn't it expected that a major league SS is supposed to make the play? How does that have anything to do with luck? Gee, he was lucky that the professional ballplayer didn't botch the play that he is paid handsomely to make. Isn't that what the E in ERA is all about?

[logan82]

I am just trying to understand FIP a little better even though i doubt I will use it much. I guess I should have said higher FIP, but then again like you said a pitcher with low K's and GB tendancies would benefit from fewer HRs in general.

[TooLiveBrew]

However, whats wrong with ERA? If a pitcher gets a guy to hit an easy ground ball to the shortstop, isn't it expected that a major league SS is supposed to make the play? How does that have anything to do with luck? Gee, he was lucky that the professional ballplayer didn't botch the play that he is paid handsomely to make. Isn't that what the E in ERA is all about?

 

Yea, except that many (a majority?) of misplays by fielders aren't registered as errors. Capuano is a prime example of this in 2007. Since Weeks & Braun missed so many 'routine' balls, yet weren't charged with errors, the burden then falls to the pitcher - even though an average fielder would turn most of those balls into outs. You've made light of the fielding situation, but in actuality, what you've described is precisely why FIP/xFIP has value.

 

In other words, why punish a pitcher for something that's totally out of his control? That's what ERA does to pitchers like Cappy with the 2007 Crew defense behind him. And it's also why all the 'OMGz he totalee lozt 17 gmz in a row!' talk from 2007 was comprised of plenty of hot air.

[mothershipconnection]

The issue is that defense can't be broken down cleanly into "error" and "non-error". There are so many other plays that don't show up, as we can't categorize phenomenal play, pretty darn good, solid, fair, weak, pretty crappy, and error. Ozzie Smith made a lot of plays that other guys never would, but it didn't really show up in their numbers since Ozzie didn't get extra credit and the other player didn't get an error for not making a play only a freak could make.

[Patrick425]

Yea, except that many (a majority?) of misplays by fielders aren't registered as errors.

I think this is way overused excuse for Fip. In reality, what percentage of misplayed balls are not classified as errors? People make it sound like fielders should make plays on any ball hit within 30 feet of them. 99.9% of the time the routine plays outnumber the bad and spectacular plays in a game and Fip ignores all of them.

 

Pitcher A: 7 IP, 10K, 5 ER, 0 HR, 3 BB

Pitcher B: 7 IP, 3K, 1 ER, 1 HR, 3 BB

 

Maybe I'm understanding Fip wrong, but I get the impression that Fip would say pitcher A is better than pitcher B. I don't get that.

 

Baseball is played with 9 players, not 1. I think it's reasonable that pitchers should be expected to be able to rely on his fielders and not be expected to K everyone. If a pitcher makes a good pitch on a hitter in a tough situation and gets him to ground into a DP, he should definitly get some credit for that. Sounds like Fip would just ignore that. Sounds like people are trying to reinvent the wheel with this stat.

[Ennder]

ERA just has way too much luck involved. How many balls fall for a hit, how many guys are on base when you give up a hit, how many runners your bullpen let in, mistakes made by your fielder that don't go as an error, which field you are playing in etc. ERA shows a very weak correlation from year to year so it is not a predictive stat. FIP and xFIP show stronger year to year correlations so they are more predictive. As I said in the original post though, they are flawed in their own ways and you pointed out one of them, GB%. Another flaw is that some pitchers have a sustained ability to strand more runners than normal and that doesn't show up in FIP.

 

With 1 year of data I find ERA almost meaningless, especially for a RP.

[TooLiveBrew]

I think it's reasonable that pitchers should be expected to be able to rely on his fielders and not be expected to K everyone.

 

Right, but patrick, why are you desiring - essentially - to punish the pitcher when his fielders aren't reliable? That doesn't make sense. In a vaccuum, ERA is the perfect stat. But since baseball isn't played in one, it has serious drawbacks.

 

Your stat line example is too unspecific to determine whose FIP would be better, iirc. You just have some random ER, but it's a good way to understand how ERA punished Cappy or Soup - if all those runs are earned, there's the chance (and with the 2007 Brewers, a very very good one) that some of the hits were plays that should have been converted into outs, but weren't, and weren't scored as errors.

99.9% of the time the routine plays outnumber the bad and spectacular plays in a game and Fip ignores all of them.

 

You have to ask yourself which is preferred though - punishing pitchers for easy plays that weren't turned into outs & weren't scored as errors (which is highly subjective anyway), or ignoring the impact of defense entirely and only focusing on what a pitcher can control...

Fielding Independent Pitching.

[Ennder]

Ok here is another look at the bullpen from a different angle... ZiPS projections!

 

Gagne - 3.60 ERA

Gallardo - 3.60 ERA

Sheets - 3.72 ERA

Riske - 3.80 ERA

Torres - 3.94 ERA (pirates projection)

Wise - 4.00 ERA

Turnbow - 4.08 ERA

---- League Average RP ---- 4.15 ERA

Shouse - 4.21 ERA

Parra - 4.31 ERA

Villanueva - 4.46 ERA

Capuano - 4.48 ERA

---- League Average SP --- 4.55 ERA

Bush - 4.64 ERA

Suppan - 4.76 ERA

McClung - 4.97 ERA

Vargas - 4.99 ERA

Mota - 5.06 ERA

 

ZiPS likes us to have 4 above average SP and 5 above average RP's. That would be a pretty big improvement over last season and in general ZiPS is generally harsh on pitching.

[sbrylski]

Also note that xFIP takes away park effects.

[Patrick425]

punishing pitchers for easy plays that weren't turned into outs & weren't scored as errors (which is highly subjective anyway),

This is what I just don't get. This happens very very rarely. A routine play in my opinion is an easy ground ball hit directly at a fielder. If a fielder misplays it, it goes down as an error. Unless the official scorer is blind or something. People are acting like this happens all the time. It hardly ever happens.

 

I understand that everything else being equal a pitcher that has an overall better defense behind him will more likely have a better ERA than a pitcher that doesn't because his players make more "outstanding" plays. However, I don't think the solution to understanding the effect that has on ERA is to eliminate ALL plays that involve a fielder, even though a majority are routine. I'm going to have a hard time ever understanding that logic.

[sbrylski]

This is what I just don't get. This happens very very rarely. A routine play in my opinion is an easy ground ball hit directly at a fielder. If a fielder misplays it, it goes down as an error. Unless the official scorer is blind or something. People are acting like this happens all the time. It hardly ever happens.

 

You're saying the difference between a good fielder and a bad one is a rare play here or there? Are you taking into account range, or even arm strength?

[Ennder]

This is what I just don't get. This happens very very rarely. A routine play in my opinion is an easy ground ball hit directly at a fielder. If a fielder misplays it, it goes down as an error. Unless the official scorer is blind or something. People are acting like this happens all the time. It hardly ever happens

 

Not true at all though. Taking the most extreme example from last season Florida's defense let in 102 more runs than you would expect from an average defense. Colorado let in 58 fewer runs than an average defense. Some of those are from unearned runs but the difference between those two numbers leads to a staggering difference in team ERA.

 

When you look at a RP's ERA the differences can be amazing. Riske threw 69 2/3 IP last season and had a 2.45 ERA giving up 19 ER's. Lets imagine he pitched for the Brewers instead and Braun and Hall each miss one big play on him resulting in just a paltry 3 extra ER's, then the bullpen comes in and allows 2 more inherited runner than the cleveland bullpen did. I mean 5 ER's is nothing right, it is really easy to see how things could be 5 ER different over 65 games pitched without Riske being at fault at all. He now has a 3.10 ERA.

 

So how good was Riske? All it takes is 2 missed plays during the year and one RP giving up a hit after he is taken out to make his ERA jump by 0.55. How many times did the Brewers miss a play and end up extending things into a big inning, how many times did our bullpen come in and give up inherited runs. Those things add up very fast on the ERA stat and no it doesn't even out over a single season, especially not for a RP.