Three observations from Phil Hughes’s first half


Phil Hughes is no No. 5 starter. He might have won the competition for that spot in spring training, but his results have been more in line with a No. 2 or No. 3 guy. It was what the Yankees had in mind when they placed him in the 2010 rotation instead of their 2009 No. 5 starter, Joba Chamberlain. Given their first half performances, that decision appears justified.

AP Photo/Bill Kostroun

While Hughes has shown flashes of his ceiling at times, he has also hit a few rough patches. In four of his 16 starts he has allowed five or more runs, and in another he allowed four runs in 5.2 innings while using 117 pitches. The seven starts with seven or more innings and two or fewer runs helps offset those, but they don’t make them go away. There are still issues he needs to work on in the second half if he’s going to reach his ceiling.

Overall Hughes’s numbers look great. He not only sports a 3.65 ERA, but he has the peripheral stats to go with it, a 3.59 FIP, 3.96 xFIP, and 3.66 tERA. He has struck out 22.2 percent of the batters he’s faced, and has walked just 7.1 percent. As a testament to his control, with a little nod to luck, he has avoided hitting a single batter this year. His HR/FB ratio and BABIP are right in line with his performance last year. Yet despite these numbers Hughes has shown a few concerning trends in his 16 starts.

Performance in 3-2 counts

Of the 410 hitters Hughes has faced this year, 49 have run the count full. This is a bit below the league average rate, but I’m guessing above average for a top of the rotation starter. In those counts Hughes has fared terribly, allowing opponents to hit .258/.531/.419. Of the 29 batters he has walked this season, 18 have come in full counts. It certainly hasn’t been his best situation, though it doesn’t appear he’s doing anything differently.

In full counts Hughes throws 55 percent four-seamers, 28 percent cutters, 15 percent curves, and 1 percent changeups. He throws pitches with nearly the same frequency in 2-2 counts. Why the difference in performance, then? Clearly, throwing the ball outside the zone on a 3-2 count has been a problem. When he does throw a strike opponents hit for a better AVG and SLG than league average in full counts. That’s something he’s going to have to improve on in the second half. Thankfully it’s not something that comes up too often.

Facing the No. 9 hitter

Starting pitcher will obviously face the No. 9 hitter at a lesser rate than hitters higher in the order. In the first half Hughes faced the No. 1 hitter 50 times, while facing the No. 9 hitter just 38 times. Against the leadoff men he’s been quite excellent, allowing them to hit just .213/.240/.340. Opponent batting numbers rise through there, peaking at No. 5 hitters, who have a .879 OPS against Hughes. Then it starts to decline as the hitters get worse. Yet there’s another peak at the bottom of the order.

No. 9 hitters have knocked Phil around, hitting .344/.432/.500 against him. It’s a small sample, sure, but those are still terrible results against the nominal worst batters on each team. The AL average against No. 9 hitters is .239/.299/.342, which actually looks even worse for Hughes because he doesn’t have to face the best No. 9 hitter in the league, Brett Gardner. It seems like a fluke that Hughes would fall apart against these guys. I doubt it’s anything more than that, but if I find some time this week I might examine this a bit more deeply.

Bad when behind, but not behind often

Opposing hitters go nuts against Hughes when they’re ahead in the count. In 108 plate appearances they’re hitting .321/.500/.526 against a league average of .308/.477/.507. But while Hughes manages a worse-than-average line while behind in the count, he doesn’t face those situations often.

In the American League the hitter has been ahead in the count 35.9 percent of the time — that is, 35.9 percent of plate appearances have ended with the batter ahead in the count. For Hughes that number is just 26.3 percent. He has faced an even count 33 percent of the time and has been ahead in 40.7 percent of plate appearances. The league as a whole faces an even count more often than having the pitcher ahead.

This trend might be related to the 3-2 count as well. Again, it will take a bit deeper of a look to make a determination, but intuitively the connection makes sense. Maybe he’s trying a bit too hard and ends up not throwing his best stuff. Maybe he throws fatter pitches when behind, as to avoid walking batters. Again, I’m not sure. I’m just glad that Hughes doesn’t face those situations often.

The first half has been mostly good for Phil Hughes, despite a handful of pitiful starts. He’s had his share of excellent ones, though. His biggest problems seem like either flukes or correctable aspects of his game. If he can make those adjustments in the second half we could see a better Phil Hughes. That’s a scary proposition for the American League contenders.

Categories : Pitching


  1. king of fruitless hypotheticals says:

    what rotation will we see to start the second half? will the pitchers actually be lined up in there ‘presumed worth’ order? and if so, where will we see phil?

  2. Overall Hughes’s numbers look great. He not only sports a 3.65 ERA, but he has the peripheral stats to go with it, a 3.59 FIP, 3.96 xFIP, and 3.66 tERA.

    Where do you go for readily available tERA stats? Neither FG.com or BR.com have incorporated it yet, and they’re by far the most convenient stat sites out there.

  3. Of the 29 batters he has walked this season, 18 have come in full counts.

    My initial reaction: that seems low, not high. I’d want more of his walks to have come in full-count situations. That stat means that 37% of his walks issued this year have been on 3-0 or 3-1 counts, meaning he’s exhibiting worse control than if the walks had been on 3-2 counts.

    • Chris says:

      In the AL as a whole there have been 4024 walks so far this season. Here’s how they break down by count:

      3-0: 955 (24%), also 955/1005 PA or 95% (only 230 were classified as IBB)
      3-1: 1184 (29%), also 1184/2303 PA or 51%
      3-2: 1885 (47%), also 1885/6153 PA or 31%

      For Hughes, here are the same numbers (29 total BB):

      3-0: 7 (24%), also 7/7 PA or 100% (1 IBB)
      3-1: 4 (14%), also 4/7 PA or 57%
      3-2: 18 (62%), also 18/49 PA or 37%

      So the issue with Hughes seems to be more the amount of times he get’s into full counts more than the outcome of plate appearances when he gets in that count.

  4. phughesisgod says:

    Hughes has undoubtedly been great for the Yankees this season, and he did hit a rough patch there for a little. But he pitched much better in his last start. I do think that some of his faults can be easily corrected by using his changeup alot more. The development of his changeup during the spring was the reason he won the job. He needs to trust his changeup, because that is the only way it will develop into a good pitch. If he trusts it, he will use it more and it will continue to develop. When hitters are forced to have to think about 4 pitches, it will be alot easier for Hughes to get guys out. The development of that offspeed pitch will be crucial if Hughes is to reach his full potential, which is that of an ace.

    • Mike HC says:

      He has already started to go to the curve a bit more the past couple of starts. And I’m sure they are going to whip out the change again at some point.

      The best part is, Hughes has proven he can get guys out with just the two fastballs, so these extra pitches are only going to make him that much better. He has a really good chance of turning into one of the best pitchers in baseball in the next couple of years. I am excited.

      • phughesisgod says:

        The knuckle curve that he has is by far his best pitch, IMO. And thats scary considering that his fastball and his cutter are 2 plus to plus plus pitches. His knuckle curve really reminds me of Moose. Moose’s was just downright disgusting and always made hitters look like idiots, and Hughes’ has been like that this season. That pitch is just nasty.

  5. Kiersten says:

    Sorry if this is off-topic (I don’t think it is), but can someone explain what xFIP and tERA are?

    • Tom Zig says:

      Yeah I could use a good explanation myself.

    • phughesisgod says:

      The only thing I know is that they are sabermetric statistics, which became famous because of the “Moneyball” philosophy of Billy Beane. Sort of like the next level statistics.

      I rarely ever use these types of stats, and really only pay attention to AVG, OBP, Slugging, HR, RBI, ERA, WHIP and K/9, and the gb-fb ratio.

    • Jose the Satirist says:

      xFIP is like FIP, except HR are normalized toward 10.6% of flyballs are HR.

      tERA uses linear weights to compare how a pitcher did versus expected number of runs in a game situation. Basically you compare what was expected to happen and what did happen.

      Just a quick and easy explanation.

      • Do either xFIP or tERA take into account the fact that the ball arrives at the plate at a slower velocity than it leaves the pitcher’s hand?

        Because that actually happens; the ball slows down during flight. I didn’t know if you were aware.

        Tim McCarver

        (I thought about going for a Joe Morgan joke, but I wanted to stand out from the crowd.)

      • Kiersten says:

        So are they both like ERA in that lower is better?

        and FIP is kind of like adjusted ERA right?

        • So are they both like ERA in that lower is better?


          and FIP is kind of like adjusted ERA right?

          FIP is Fielding Independent Pitching. It’s a conceptualization of what a pitcher’s ERA should be based only on things he can directly control without the help (or hindrance) of the skill of the 8 defenders around him.

          • Kiersten says:

            Perfect. Thanks.

            Still not sure I completely understand tERA though.

              • Disco says:

                That is my blog and that’s a post from a writer on the staff.

                That post came out before fangraphs added tERA. It’s just a coincidence the same name was given.

                In our post, tERA was the name given to the stat of the future. It was based off the video technology that MLB has implemented recently and that won’t be available for information for a couple seasons. So the tERA in our post is not real, but is a stat we might see in the future with the new wave of information we will get with the video technology.

                tERA on fangraphs is a real stat obviously.

            • Jose the Satirist says:

              I’ll try to make it simple. If this still doesn’t make sense read tsjc’s link.

              tERA realizes that a groundball, flyball, and line drive all have different outcomes. It sets weights for all different outcomes and creates a stat scaled to ERA. Example: A pitcher who gives up more line drives than ground balls is not going to do well.

              • Or think of it this way:

                Saturday night, Joba came in in the 8th inning on the road with a 1-0 lead and
                1.) with the bases empty, gave up a short fliner to shallow CF to the first batter
                2.) with a man on first, forced a soft groundball back to himself for the second batter
                3.) with a man on first, gave up a short fliner to shallow LF to the third batter
                4.) with men on first and second (later, second and third), walked the fourth batter
                5.) with the bases loaded, gave up a grand slam homer to the fifth batter.

                He was charged with 4 runs, all earned.

                tERA would look at those exact 4 batted balls hit to those exact places at those exact velocities over several decades worth of historical data and say what would have happened with a totally average defense behind him. Maybe different fielders get to those first and third liners of the inning and more outs are made, so the GS homer would have only been a 2 run or a 3 run homer instead. And so while Joba gave up 4 runs in real life, if you replay that exact scenario of what happened thousands and thousands of times in baseball history, the average pitcher with the average defense who gives up those exact same hits in that exact same sequence gave up, say. 3.1425 runs–on average. (Or maybe they gave up 4.3256 runs, had Joba been lucky as opposed to unlucky. Either scenario is possible.)

                The probabilities of all the possible outcomes of those batted balls hit to those spots is used to determine how many eventual runs each of those initial hits (including the homer) would have scored under normal conditions, and the values are all added up, so you have a probabalistic view of how many runs Joba would have given up in a theoretical large sample size of that specific small sample size of what actually happened.

        • Ross in Jersey says:

          and FIP is kind of like adjusted ERA right?

          Not exactly. Adjusted ERA takes park factors into account, and compares the ERA to the league average.

          FIP tries to measure how well a pitcher does at things he has direct control over. Strikeouts, HRs, walks, etc. If a pitcher’s FIP is much lower than his ERA, it can give you an idea that he has simply been unlucky in areas he has no control over.

  6. NYYROC says:

    During yesterday’s pregame PH was talking to Jack Curry. Hughes said using his changeup will be one of his keys for the 2nd half. He said teams have seen him now and he needs to use his secondary pitches more.

  7. Mike HC says:

    Hughes is only getting better. That first half was just the beginning. I expect a solid second half, but when it could get really scary for the rest of the league, is next year.

  8. ADam says:

    im curious to see how innings 140 – 170 go… Uncharted waters for young master Hughes

    • phughesisgod says:

      If Hughes is dominating in those starts, I find it hard to believe that Girardi will take him out to preserve his innings limit. I think it will get really interesting if one game Hughes is pitching a no hitter or a perfect game and during the game he reaches 175 IP or something, then Girardi will really have a decision to make. Shut him down, or let him continue until he either a) achieves the feat or b)gives up a walk and/or a hit. Boy wouold I love to hear those discussions in the dugout with Eiland, Pena, Girardi and Hughes.

      • Mike HC says:

        I don’t think it is going to be that rigid of an innings limit. I’m sure if they put him out for a start, it is with the intent that he could all nine if he keeps his pitches around 100.

        And I think it is safe to assume the Yanks are not solely using “innings” as the the marker for limiting his use. They chart every pitch, even the practice ones.

  9. ZZ says:

    Even stranger than his struggles with 3-2 are his struggles with 1-2 counts.

    His line still looks decent at 1-2: .268/.268/.329, but in the context of the count that is pretty poor.

    His sOPS+ is 193 on 1-2 counts.

    A pitcher like Hughes should be blowing people away on 1-2, like he does on 0-2 with a .047(!) OPS.

    Just some strange with some of Phil’s statistics so far. This 1-2 thing really does not make much sense in the context of his other statistics and his stuff.

    • Not Tank the Frank says:

      I’m too lazy right now but I’d like to see his numbers in 2-2 counts. I think his numbers on two-strike counts are related to his (perceived/real) problem of foul balls and putting hitters away.

      I think the kickass 0-2 numbers are the combo of a SSS and a testament to all the weapons Hughes can go to when he gets to 0-2. Of course, he’s at his best when pounding the zone and can go upstairs with a fastball, get you with the curve, or induce weak contact with the cutter.

      What this all seems to say is that hitters fare much better against Hughes when they work the count/foul pitches off. Not exactly a revelation but it confirms what I’ve seen.

  10. Rich M says:

    Cesar Izturis is 4 for 11 against Hughes batting in the 9 spot this year.

  11. ZZ says:

    Also, Phil has to get better with 2 outs.

    With 0 outs he has an sOPS+ of 74.

    With 1 out a 55.

    But with 2 outs, 115.

    There is no reason he should be that much worse with 2 outs. Has to stay focused for the whole inning.

    • whozat says:

      I kind of hate how we project onto players. Couldn’t it just as well be that he amps up and tries too hard to get the 1-2-3 inning? Why does it have to be that he “loses focus”?

    • Ross in Jersey says:

      On the bright side, his OPS+ diminishes as the leverage increases:

      Low Leverage : 95 OPS+
      Medium Leverage: 79
      High Leverage: 44

    • Jose the Satirist says:

      I don’t know that you can say this is a focus thing. I mean Ubaldo Jimenez who is having an insane year has a 40 sOPS+ with 0 outs and then a 92 sOPS+ with 2 outs. That is a bigger gap between the two than Phil Hughes. Are you going to claim Ubaldo has a focus problem this year? A lot of it could be just a freak occurrence. That is pretty presumptuous to immediately play the focus card.

      • Rivera Venue Blues says:

        How about maybe the pitcher faces the 3 or 4 hitter slightly more often with 2 outs over the course of a season because that’s how the initial set-up of the lineup makes it happen. Wouldn’t it make sense that most teams have better hitters at 3 or 4 than 1 or 2 make up the gap a decent number of pitchers seem to have in that situation?

    • Mike HC says:

      I think both of your comments come down to mixing his pitches. After a batter has seen the 4 seamer and cutter a couple of times in their at bat, they get the timing down, so it is harder to put the hitter away. And same applies to the last out of an inning. After Hughes has been throwing the same two pitches for an entire inning, batters start to get the timing, so by the time the final out comes, nothing comes as a surprise.

      The key is to mix an more off speed pitches and that will solve both problems.

    • Am I says:

      I am just guessing, but I would expect that those numbers are impacted by the fact that he is more likely to have a runner on with 1 out than 0 out, and even more likely to have a runner on with 2 outs (there being more chances to get a runner on). It is fairly commonplace for a pitcher to have worse numbers with runners on, no?

      • Mike HC says:

        But Hughes pitches his best in high leverage situations, which many times implies, but not always, that there are men on base. So I’m not sure he has a problem with runners on, although there are probably splits for him on that too, so no reason to speculate. I just don’t know where to find them.

  12. B-Rando says:

    Watching Saint Phil pitch this year has been the highlight of the season for me. Watching him develop as a pitcher, dominate (and struggle against) other lineups and start to become the pitcher we thought he could be has been simply awesome.

    I agree with other sentiments here that his changeup is key. If he can continue developing that pitch, he can be a completely different monster. I also think another key to his 2nd half will be ensuring he doesnt lose the feel for that curve. Being able to throw that pitch in any count will make him a stronger pitcher, because teams won’t necessarily sit fastball/cutter in a situation where he must throw a strike (ie: 3-2 count)

  13. CS Yankee says:

    The grade should be based upon a curve whereas the high leverage situations are more of the grade than the 9th guy or the 2-out batter.

    I would rather have Hughes placed as the 5th starter and skipped a few times in August (heat, etc) and go full bore come September. They may plan on using him as the 8th inning guy after they clinch (Sept 18th) as long as the other four keep up their recent performances & Hughes reaches his innings limit. I’m good either way.

  14. nsalem says:

    Is that 1st place or playoff spot clinched on 9/18. At that point 6 games remain vs Boston and 4 vs Tampa Bay. Rays close out against
    Mariners, Orioles and Royals.

  15. A-Rod's Hip says:

    real good article.. great information

    quick side note and questions..

    can someone in their best ability distinguish between FIP, xFIP, and tERA?? i always get confused whenever i try to venture out on my own to try and figure these things out.. i’d appreciate it

  16. A-Rod's Hip says:

    on another note as well.. in reference to the stats from the AL’s #9 hitters and their .239/.299/.342 line.. i’ve been trying FOREVER to find these numbers to use as aamo against one of my moronic NL fan friends who keeps trying to tell me that the pitcher hitting makes no difference on the game and that “AL pitchers just can’t pitch” and that’s why their ERA’s are “so much” higher

    where did you find these numbers?? or does anyone else know where i can find these stats as far as an average around each league at a particular spot in the batting order?

    thanks again for anyone’s help!

    • A-Rod's Hip says:

      whooops.. nevermind! just found this on baseball-reference.com

      that has to be the second greatest site in mankind’s history (right after riveraveblues.com of course!)

    • phughesisgod says:

      Wait? Tell me if I read your post incorrectly…

      Your fans try to tell you that the pitcher hitting is like having a DH, and that it makes no difference if you have on or the other? You must have a lot of will power and patience then, because if I were you, they would have been hit a long time ago. I’m sorry, but anyone who says that they make no difference needs to be hit. The DH makes a huge difference! If you were a pitcher, would you rather face say David Ortiz or Roy Oswalt? I’d definitely rather face Roy Oswalt, but that could just be me.

  17. Ivan says:

    Hughes has been pretty good considering this actually his 1st full season of him starting.

    In 07 he was hurt for about 3 months.

    In 08 got off to a horrible start and then got hurt and didn’t pitch again til september.

    In 09, replace Wang, then was put in the bullpen for the rest of the season.

    I mean this is actually the 1st time since 06, that he has been able to make his starts consistenly without injury or anything else.

    So while he has develop very well, the scary thing for the rest of the league is that he has a chance to better, and I mean BETTER!!!

  18. MikeD says:

    His ERA has come in line with FIP line. Early this year there was a pretty large gap, suggesting his 1.40 ERA was going up. Then again, even without FIP we knew that!

    Overall I view the stats you posted as a positive because they are all very correctable, which means Hughes can improve. Then again, we also knew that! This is his first full year as a starter in the Majors, so he’s still going through a maturing process. Being knocked around a few times after getting off to a great start, or having difficulty putting away some hitters on 3-2 will force Hughes to continue to elevate his game.

    The best is yet to come. If not this year, in coming ones, for Hughes.

Leave a Reply

You may use <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong> in your comment.

If this is your first time commenting on River Ave. Blues, please review the RAB Commenter Guidelines. Login for commenting features. Register for RAB.