Will Ubaldo Jimenez Bounce Back?
Through seven starts last season, Ubaldo Jimenez was the best pitcher in baseball. In 48.1 innings pitched, Jimenez allowed 28 hits while striking out 49 batters en route to a miniscule 0.93 ERA. This season, things have changed for the 27-year-old hurler. In his first seven starts this season, Jimenez hardly looks like the same pitcher. Over that same period, Jimenez carries an unsightly 6.14 ERA. Expected to be entrenched in a fight for the division crown all season, the Colorado Rockies need to figure out what’s wrong with their ace before it’s too late.
There are a few reasons to be optimistic about Jimenez’s current performance. Despite the high ERA, Jimenez is currently striking out batters at a career rate. His BABIP is slightly higher than normal, and his LOB% is a paltry 61.8%. All told, Jimenez has performed much better than his ERA has indicated this season, posting a 4.45 FIP and a 4.24 xFIP. Still, that’s hardly the type of performance the Rockies expect from their ace.
There currently seems to be two main reasons for Jimenez’s struggles- the first of which is control. Never his strength, Ubaldo’s control has completely deserted him this season. In 36.2 innings, Jimenez has walked 23 batters, good for a 5.65 BB/9. Oddly enough, Ubaldo’s first-strike percentage is the highest since his rookie season, making his struggles all the more confusing. Ubaldo has been able to get ahead in the count early, but appears to lose hitters as he gets deeper into at-bats.
The second reason for Ubaldo’s struggles is his declining velocity. According to PitchFx, Ubaldo’s velocity is down on nearly all of his pitches this season. The most drastic change has occurred with his slider, which has dropped from 86.4 to 82.6 mph. On top of that, his fastball is down nearly two miles per hour. The dropoff seems to have had a clear effect on the effectiveness of Jimenez’s pitches. Both his fastball and slider currently carry a negative pitch type value this season. Over his career, those have been Ubaldo’s best pitches.
There seems to be one unified theory that helps explain Ubaldo’s poor start to the season: his mechanics are out of whack. This would go a long way toward explaining both his issues with walks, and his lack of velocity. If Jimenez cannot find his proper release point, he’s going to have less control over his pitches. If something is out of whack with his mechanics, that would easily explain why his velocity is down. Jim Tracy seemed to agree with that assessment, noting the following after Ubaldo’s start on Tuesday.
“There is no doubt in my mind that what we’re dealing with here is mechanical in nature, and once he gets the complete feel for that and the consistency of it to repeat it pitch after pitch after pitch, then I think you’re going to see the Ubaldo Jimenez that you guys have been waiting for, and all of us have been waiting for, actually, to show up.
If his start on Tuesday was any indication, Jimenez may already be working through his issues. In his longest outing of the season, Jimenez walked only one batter – the first time that has happened since his season debut. For a pitcher like Jimenez, who doesn’t have the “ideal” delivery, even the slightest change to mechanics can have a devastating effect on performance. The good thing is that Jimenez, his coach and his trainer seem to realize the issue at hand is not physical, and are working to correct it. So long as there’s no physical issue with Jimenez – and we have no reason to believe there are – we should start seeing the old Jimenez again. The one that finished third in the Cy Young voting last season.
Chris is a blogger for CBSSports.com. He has also contributed to Sports on Earth, the 2013 Hard Ball Times Baseball Annual, ESPN, FanGraphs and RotoGraphs. He tries to be funny on twitter @Chris_Cwik.
” Despite the high ERA, Jimenez is currently striking out batters at a career rate”
This is why using K/9 is dumb.
He’s K’ing 22.1% of batters he faces this year. He was K’ing 24% of batters last year. Thats not an increase.
He’s getting hit harder. Its that simple.
Any stat that thinks a hit is a better outcome for a pitcher than a flyout/groundout is a stat that doesn’t make much sense. Stop pretending /IP is defense independant. Its not.
The tone of that post did not make me want to hear what you had to say. Especially when you called the writer dumb right off the bat.
When writers consistenly use statistics that are seriously flawed, I’d call that dumb.
K/9 doesn’t tell you how well a guy strikes people out. Its essentially a ratio of K% and the inverse of babip. (IE, K/IP where most of the IP is made up of BABIP)
Which leads into why FIP is such a flawed stat (IE, it claims to be defense-independant, but its denominator is heavily dependant on BABIP)
In both of these stats, giving up a hit is better than having a fielder catch the ball.
And fjrobinson’s comment goes right over RC’s head.
This seems like a valid point (if presented in an overly belligerent manner). However, I wonder if this is actually very significant for most pitchers, or in this specific case. K/9 shows that his strikeout rate has climbed a little, but not much (8.69 last year, 9.08 so far this year). K% shows that his strikeout rate has fallen a bit, but again, not much. I think the assumption that batters faced is fairly constant per inning pitched should be critically evaluated, but I wouldn’t be surprised if the reality is that that assumption turns out to be accurate for the vast majority of pitchers.
He didn’t say the writer was dumb, he called the stat dumb. Big difference. And I think it’s a very good point he was making.
Doing some quick calculations, BF/IP varies quite a bit from pitcher to pitcher, but actually seems to be VERY consistent year to year for individual pitchers.
Seems to range from 3.9 to about 5.
For instance, Jon Lester the last 4 years:
4.155
4.146
4.139
4.127
4.25 or so seems to be the norm. Ubaldo is up over 4.5 right now, which is whats making it look like his “strikeout rate is up”
K/9 is a fine stat if you simply acknowledge what it is, by definition. It’s the % of his outs that the pitcher achieved via strikeout (x27). Which is a useful thing to know. Is the pitcher getting outs via strikeout, or via balls-in-play?
But it is not what % of batters a pitcher strikes out.
Picking one data point does not make for a successful argument. I’m just going to “arbitrarily” look at Dan Haren’s BF/IP for the last three full years…
2010: 4.23
2009: 3.96
2008: 4.08
Looks consistent to me!
Also, BABIP and BF/IP had a .643 correlation among pitchers with at least 50 IP, so while we’re picking on stats that aren’t “defense independent”…
K/9 is simply a measure of how pitchers get batters out. It’s not necessarily “defense independent”. Its correlation with BABIP is .002 (for 50+ IP last year), just as a side note, so it looks pretty defense independent to me anyway! Meanwhile, K%/BABIP had a correlation of -.104, still small, but a bigger value than K/9 nonetheless. Probably because the denominator of K% takes into account players who reach base because the defense doesn’t get them out…
Also, FIP has almost zero correlation with BABIP (-.001 for 50+ IP last year), so I would check your numbers on that.
Yes, Jimenez is definitely getting hit harder than last year (as shown by his LD%), which is a big part of why he’s been struggling. I definitely and wholeheartedly agree with that part of your argument. But to say that K/9 is “stupid” and K% is more defense independent is just bunk, considering they’re functionally the same thing.
” It’s the % of his outs that the pitcher achieved via strikeout (x27). Which is a useful thing to know.”
If the assumption that FIP makes, IE, that BABIP is Random/Uncontrollable, then K/9 is essentially this:
27* (OutsViaK)/(OutsViaK + OutsViaBIP)
where OutsViaBIP is an essentially random number that is signfiicantly larger than OutsViaK. I’m not sure how thats useful.
“Also, FIP has almost zero correlation with BABIP (-.001 for 50+ IP last year), so I would check your numbers on that.”
Take the normalizing factor out of FIP and recheck your numbers (IE, the +3.2 or whatever it is)
Taking out a constant will change the correlation by… zero. It’s a constant. Correlation compares the relative value of one thing to the relative value of another. If you’re adding the same constant to everything, the relationship between the values stay the same.
his gb% also has dipped below 40, way down from his usual 45 – 50 rate
In his last start the 2-seam FB that he used to to such good effect last year looked to be miving well again. Jimenez needs that combination of 4 and 2 seam fastballs to be effective.
That thumb injury may have cause a slight change in his grip, which in turn may have effected his control. The Rockies are in surprisingly good shape considering the poor results from Jimenez to date.
I’m not a statistician, but isn’t the fact his K% is down, but K/IP is up a direct reflection of the increased walks? Which goes to the authors point. By walking more batters, he is facing more batters/9 than last year. So despite striking out more batters/9, the overall K% drops. If the walks come down (which is a point of the article), the K% will increase and be above last year’s performance if K/IP doesn’t change. So I think he is correct in saying he is striking people out at a career high and if he can get his BB/9 down to last year’s level, he will be a stud. I would also highlight that the statement striking out at a career high can mean lots of things. Since he is at his highest K/9 and if he pitches as many innings as his career high, he will strike out more batters in a season than ever before. Therefore, he will in fact be striking people out at a career high. Nothing dumb or wrong about that statement.
No you are not a statistician.
He is striking out 22.1% of the batters he faces this year so if his walk rate decreases and his k/batter remains constant, his k/9 also decreases and he strikes out less players than last year (24% of batters).
If his k/9 were to increase than he would be allowing more hits which is what the original poster was getting at.
I may not be a statistician, but I can read something, interpret it and craft an argument. I said if his BBs drop, but K/IP (not K/batter) stays the same, then he faces fewer batters and his K/batter increases. If you disagree with that, I’ll be happy to illustrate it with actual figures.
Here is a guy whose k/9 dropped while his k % increased due to a large decrease in walks.
Minor leaguer, but where else is someone going to keep getting starts with 8 BB/9.
http://www.baseball-reference.com/minors/player.cgi?id=may—001tre
K/9 and hits are mutually exclusive by definition, as are K/9 and walks. Compare the two lines below (Ubaldo’s first and fifth starts of last season):
6.0 IP, 8 H, 1 BB, 1 ER, 6 K (4/5/2010)
6.0 IP, 2 H, 2 BB, 0 ER, 6 K (4/27/2010)
K/9 is the same (9.0), despite the fact that he has 8 H in one start and 2 H in the other. An increased K/9 simply means he’s getting more of his outs via the strikeout and has absolutely nothing to do with the number of baserunners the defense allows via hits or the pitcher allows via walks.
As a side note, here is another argument for K/9 being more defense independent as compared to K%. The first start has a K% of 23.1% with a BABIP of .421 as compared to the second start which has a K% of 27.3% with a BABIP of .143. So, because the defense got to 3 times more balls in play in the second outing, his strikeout stuff was better?
The idea of K/9 is to express how well the pitcher is doing regardless of the defense behind him. The more outs the pitcher gets via strikeout, the less he has to worry about how the quality of his defense.
Why don’t we just look at swinging strike rates? Take it down another level to something more predictable.
Yes but you are using a strawman argument. His K/batter is his actual K rate. That stat tells you the percentage of batters he is striking out and therefore his strikeout performance. Last year he was striking out 24% of the batters he faced. This year he is striking out 22.1%. He is not striking out at many, therefore his strikeout numbers are down, not up like the author leads you to believe.
If he walks fewer batters, there is no indication he will begin striking out batters at a higher rate, therefore you would expect his K/9 to decrease. If the K/9 stays the same those walks are either turning into hits, outs or strikeouts. If they are turning into strikeouts than his K/Batter goes up and he is pitching differently than he was earlier in the season… i.e. becoming a better pitcher.
You shouldn’t be using K/9 to justify that his strikeout stuff is better than last year and that he is giving up too many walks / hits and when those start becoming outs he will be equal or better than before. His K/Batter rate is down, therefore he is striking out batters at a worse rate than last year and therefore his strikeout “stuff” is not as good.
fangraph writers just look at the numbers, dont watch the games, dont know the players, etc etc. the problem is his health and training methods, he throws long toss on days he should’nt and runs 2.5 miles in between starts. whoever told him to do this has paid the price whether it be now or down the career. he even throws long toss in the rain for crying out loud. hire some people who know their stuff. if you want a start go to cbs.com and look at mainman24, he predict cargos 330 season as well as a 35/100 season from bautista(bautista shattered it by 54/120+), he even predicted j werth would get matt holliday money as well as jose bautistas current salary of 8 million dollars. this guy knows his stuff.
It’s tough to watch the games when you can’t get good reception.
That’s what happens when you live in your mother’s basement.
Wasn’t much of Ubaldo’s unbelievable 2010 start due to very low BABIP and HR/9 rates?
If guys are getting a hit every other inning, not getting many walks, and not hitting homers, it’s difficult for them to score many runs right?
The early 2010 UJ should not be the expectation for any pitcher. It was basically a “perfect storm”. Not unlike Liriano 2010 v. 2011.
Reduced velocity and control for a power pitcher are issues that may or may not get corrected during the season.
Curious as to how his pitches per inning have changed.
I believe that after the point where Ubaldo’s ERA bottomed out around 0.77 last season, he put up an ERA above 4.00. This season his xFIP and FIP suggest something in that range, so maybe he’s really just a 4.00 pitcher…
Well yeah, sort of. He had a .271 BABIP and 0.41 HR/9, which are both much better than average. But his career numbers are .279 and 0.58 for his career, so he has predecent of being in that area.
So it makes sense to me that a player’s K/9 rate could artificially increase a little bit if he’s facing more batters per inning due to an unlucky babip or increase in walks on 3-2 counts (for instance). And it would seem like K/PA is a more accurate measure of strikeout ability.
Using a hypothetical event, there are two outs in the inning and Jimenez gives up a single through the hole at short. In a parallel universe, Tulowitzki makes the play and the inning’s over.
Back to the real world, Ubaldo strikes out the next batter to end the inning. He got an extra strikeout that inning in part because you could say he “got unlucky.”
Now back to the parallel universe where Tulo made the play. The next inning Ubaldo faces the next batter and strikes him out (as he did in the real world the previous inning). He now has the same number of total strikeouts, but in an additional third of an inning. So Tulo making the play hurt him a tiny bit in the K/9 department, but “artificially” since his real strikeout ability didn’t change.