More photos » Jeff Roberson - AP

8 months ago: Colorado Rockies starting pitcher Ubaldo Jimenez throws during the first inning of a baseball game against the St. Louis Cardinals Sunday, June 7, 2009, in St. Louis. Jimenez threw eight innings of four-hit ball in the Rockies' 7-2 victory. (AP Photo/Jeff Roberson)

Browse more photos »

Over at Beyond the Boxscore Purple Row, RockiesMagicNumber (who surprisingly is a Rockies fan) has an excellent FanPost Story, in which he looks at Ubaldo Jimenez's season so far.  The most interesting part of his analysis was that he found that Jimenez started off slow, but then significantly improved as the game went along.  That may not seem that suprising, but it turns out that it is the other way around for most pitchers.  Here is a chart showing how Jimenez pitches from inning to inning in comparison to the average NL starter:

As you can see, Jimenez significantly improves as the game progresses, while the league average starter gets worse.  Because I am a sucker for anomalies, I wanted to look more in depth at why Jimenez has done this.      

First, let's start off by getting a picture of what kind of a pitcher Jimenez is.  Most people likely only saw Jimenez extensively in the 07 playoffs, and they probably know the stuff is clearly there.  Here are all of the pitches he has thrown this year, based off their horizontal and vertical movement:

I am not quite good enough at PTICHfx to be able to accurately create my own pitch classifications, so the ones here are provided by the Gameday algorithm, and unfortunately, those aren't perfect.  Some of those curves are probably sliders and vice versa.  Ditto with fastballs and changeups.  However, this should be good enough to give you a rough representation of the kind of stuff he has and the frequency at which he throws those pitches.  

Jimenez has a really, really nasty fastball, averaging over 95 MPH, and topping out at around 100 on several occasions.  He is currently tied with Justin Verlander for the leader in fastball velocity in the majors.  In addition to that sick velocity, he'll put some good movement on them now an again, with a small cluster of them having as much as 12 inches of horizontal break.  Overall, his fastball has been worth roughly 3.7 runs above average so far this year and it was a plus pitch for him last year as well.  He'll suplement that fastball with a changeup, that has similar movement but is usally about 10 MPH slower.  That speed differential has worked for him this year as the changeup has be worth 1.1 runs above average.

His next most used pitch is a hard slider.  He'll throw it anywhere from 80-90 MPH with a lot of break.  His slider may well be his best pitch this year.  Fangraphs has that pitch saving a ridiculous 6.8 runs already this year when he throws it.  Thanks to there awesome leaderboards (seriously, someone should buy David Appelman a car), we can see that it has been the 8th "best" slider in the majors this year.

He also possess as big breaking, slow curve.  You would think that going from a 95 MPH fastball to a 75 MPH curve would have great results; however, his curve has been his worst pitch so far this year, actually costing him 2.3 wRAA this season.  Interestingly, it was a plus pitch for him last year and the year before, so it might just be a small sample size fluke or some bad BABIP luck that is causing it to be a negative pitch this year.       

However, while it's clear that Jimenez has great stuff, his command still has a long way to go.  Despite shaving nearly 1 walk per nine this year compared to last season, his current 3.86 BB/9 still ranks him in the bottom 5th of all qualified pitchers.  Still, Jimenez has the stuff to avoid getting hit hard even when he doesn't hit his spots well, as evidenced by his amazing .30 HR/9 mark.  If you combine that with the fact that he strikes out hitters at an above average rate, his high walk totals usually don't bite him in the ass: overall, he has been one of the best pitchers in baseball this year with a 3.28 FIP, in Coors no less.

So now that we know a little bit about Jimenez as a pitcher, I thought it would be interestinting to look at his inning by inning splits to try and see why he improves so much as the game goes along.

First off, let's look at his pitch distribution by inning (I disgregarded the 8th and 9th inning as he has only thrown a total of 44 pitches in those two innings this year):

As you can see, he throws signficantly less fastballs and more offspeed pitches as the game progresses.  Interestingly, in the first inning, when he throws his fastballs it over 85% of the time, hitters have clobbered him for a .861 OPS.  In contrast, in the 5th inning, when he throws the lowest percentage of fastballs, opposing hitters are managing just a .605 OPS against him.  In fact, despite the fact the the fastballs has been a plus pitch for him overall this year, it appears to have a somewhat inverse relationship with performance; meaning when he throws his fastball less, he actually does better:

It isn't a great correlation, but there appears to be something there.  If in fact Jimenez does pitch better when he throws his fastball less, what could be the possible causes?  His velocity remains virtually unchanged from inning to inning:

1st)     95.68037383
2nd)    95.94805195
3rd)     95.67777778
4th)     95.09922481
5th)     95.40526316
6th)     95.14375
7th)     95.2967033

It's likely that throwing less fastballs and more offspeed pitches improves the effectiveness of both of them.  When you are mixing a 95 MPH fastball with slow stuff, the fastball looks a lot faster and hitters will have less time to react to it.  Similarly they will be out in front of the slower pitches.  Conversely, hitters can hit a fastball, no matter how hard it is, if they know it is coming.  This certainly seems like something worth looking at more in depth, in relation to Jimenez and the league as a whole.