Simply put, the FLA system scares me. They do a horrific job protecting the young arms that go through there. I think Beckett has better stuff but he is in a bad pitching system that does not protect arms well.

So given that, I'd go with Harden...acknowledging that Beckett is, IMO, a better pitcher but also a greater risk due to the abusive team he pitches for...

Simply put, the FLA system scares me. They do a horrific job protecting the young arms that go through there. I think Beckett has better stuff but he is in a bad pitching system that does not protect arms well.

So given that, I'd go with Harden...acknowledging that Beckett is, IMO, a better pitcher but also a greater risk due to the abusive team he pitches for...

Harden is also 2 years younger, which while it increases injury risk, means its more likely he's still got room to grow. Just as Beckett's BB/9 have come down since he was 21 and 22, so will Harden's in all likelyhood. Maybe not to the same degree, but somewhat.

The two years, and especially the difference in health status is really what it comes down for me.

The main problem with Harden is the walks.  But when looking at his splits, he is improving in that aspect.  Before the All Star break last year, Harden had 48 walks in 91 innings.  After the All Star break he had 33 walks in 98 innings, which was actually better than Beckett (32 walks in 80 innings).  Harden also lowered his BAA by 25 points in the second half of last year.  This is the same kind of improvement Jake Peavy made in the 2nd half of 2003 to prime himself for his breakout 2004 campaign.

It seems that while Beckett has more upside, he has not showed the type of improvement that had Harden did last year.  Harden has gotten better while Beckett has not - except for that one postseason.

Beckett has stronger control and a better K rate. I like Beckett's pure stuff better as well.

Moving on to more rational discussion...

Harden's home/road splits from 2004:

Home: 95 IP, 75 H, 6 HR, 37 BB, 86 K, 3.02 ERA
Road: 94 IP, 96 H, 10 HR, 44 BB, 81 K, 4.96 ERA

The sample is small enough where it might just be noise, but we can't overlook the fact that his road ERA would have made him a below average pitcher.  He clearly benefited from pitching in Networks Associate Colliseum last year.  

Then, there's his abnormal home run rate.  Harden's groundball/flyball ratio was 1.21, basically even with the league average of 1.17.  He was a slight groundball pitcher, but nothing drastic.  However, his HR/9 rate was 0.75, significantly better than the 1.2 HR/9 league average.  So, Harden somehow managed to give up an average number of fly balls, but keep them in the park.  Those of you familiar with the DIPS theory know that it is generally accepted that this isn't a practical skill, and as studies have shown, its likely that Harden will give up more home runs than he did last year if he keeps putting balls in the air at the same rate.  

This makes a bigger difference than you'd think. If you adjust Harden's HR/9 rate to 1.2 instead of the 0.75 that he posted last year, he'd have given up 25 home runs, 9 more than the 16 he actually gave up.  Add 9 homeruns to his defense independant pitching stats formula, and you get an ERA around 4.6, not 3.9.  

Harden's performance last year was influenced by several factors which he likely had very little control over and are probably not going to be in his favor to the same degree for the rest of his career.  

Switching over to the scouting side of things, his command is also still a pretty clear problem.  There aren't many pitchers who are going to succeed while walking nearly 4 batters per 9 innings.  Harden's never shown the ability to throw his slider for strikes consistently, and his changeup is barely passable as a major league pitch.  The lack of quality secondary pitches is a problem, and not one that can be dismissed with "he's young".  

Back to John's comparison with Beckett, I think Josh has the edge over Rich in nearly every important category.  

Velocity: Slight edge Harden
Movement: Edge Beckett
Command: Big edge Beckett
Control: Big edge Beckett
Second Pitch: Big edge Beckett
Third Pitch: Edge Beckett

Harden's fastball is slightly faster and he hasn't had recurring blister problems.  Everything else, Beckett has the lead.  Makes it a pretty easy call for me.

the burden of proof is upon harden himself.

One more point I'd like to make, from a fan's perspective; Harden is, by all accounts, an extremely cool, nice guy, while I have read quite a few things that indicate Beckett is a prick. All other things being somewhat equal, I'd rather have the guy that I really want to root for on my team.

I'm not sure its realistic to ask Beckett to show consistent improvement and hold his 2003 standard as the baseline from where he should go up from.  His '03 season was awesome; his era was 32 percent above average.  If you want an A's-centric comparison, Tim Hudson's career ERA is 38 percent above average.  So, in '03, Beckett pitched like one of the best pitchers in baseball.  To show consistent improvement, he'd have had to pitch at hall of fame levels.  

Pitcher's develop differently.  Maybe Harden's second half improvement is for real and he'll keep improving.  We don't know, and with his command problems, I wouldn't bet money on it.  But as of today, it's fairly clear that Josh Beckett is the better major league pitcher.  

Also, I remember reading a quote from Mulder to the extent of--"Harden has more talent than all three of us combined (referring to the big three)."

Harden, on the other hand, walked 4.1 batters per 9 in 334 minor league innings.  Then he walked 4.8 per 9 in 2003 in the majors and managed to cut it all the way down to 3.8 last year.  

Command and control, right now, aren't even close.  It's Beckett by a mile, and Harden has no professional track record of consistently throwing strikes. If John had asked "would you rather have an age 22 Rich Harden or age 22 Josh Beckett", well, that's a different question.  But today, Josh Beckett is a significantly better pitcher than he was two years ago.  Comparing Harden to the 2002 Josh Beckett doesn't really do anything for this conversation.  

(http://www.futilityinfielder.com/dips04.html).

the assertion that Harden got lucky with homeruns is absurd.  especially because there has never been any proven correlation between GB/FB ratio and homeruns allowed.  it's all over the map every year. the stats suggest that instead of allowing 16 homeruns like he did, he actually pitched like a pitcher who only gave up 15 homers.  the conclusion here is that there was nothing out of the ordinary about Harden's home run rate, any way you slice it.  you simply just can't add 9 homeruns to his stat totals because you feel like it.  the data doesn't support that conclusion, which is completely unscientific in every way.  if you think you need a new biased estimator for Harden's peripherals, that's fine.  but at least know what you're talking about statistically before you do so.

also, the home/road thing doesn't hold much weight, or at least less weight than you think, because Harden actually was a little bit unlucky on the road this year.  His 9 inning ratios of 9.16H, 4.19BB, and .95HR for road peripherals translates into a 4.55 ERA (based on the .53H + .36BB + 1.01HR - 2.77 expected ERA calculation) as opposed to the 4.96 he actually put up.  

his dERA was 3.74, while his BABIP was .289.  there certainly isn't anything out of the ordinary with his BABIP (although it's a bit high). basically, his ERA should have been a quarter of a run BETTER, but he was actually hurt by both luck, and, believe it or not, his defense behind him.

Harden's control is something he needs to address.  I don't think anybody disputes that.  He made very good strides in the second half last year, lowering his BB rate from 4.71 to 3.03.  that is something that simply can't be ignored, especially when it happens at the age 22 season.

And his slider is already a plus major league pitch.  from the scouting side, Harden has projected as one of the top pitchers in baseball for a while now.  Stop pretending that you're a real scout.  You're simply making yourself look even dumber.

To call Harden overrated, or to suggest that he doesn't project well is a completely bogus, and, let's face it, just plain stupid argument.

If you're going to try to use stats to prove your point, at least use the right stats.  Don't just cherry pick the ones you want, and then completely misrepresent others.

all that aside, i think beckett WILL figure it out, learn how to tune the expectations and baserunners and whatever else out and just pitch. overall, beckett is better than harden. beckett to me can be a baseball legend. harden is more of "just" an allstar. but all this could change. its still really early for both of them.

After removing sallary considerations from teh equation... I can see Florida trading for Harden, I cannot see Oakland trading for Beckett, straight up.

Given the expected durability of Harden and his age, coupled with the benefit of coming through a system that focuses (admittedly more prior to Rick Petersons departure) more on long term health of their pitchers.  I would trade Beckett for Harden, not likely the other way around. So, long term, I'd rather have Harden.

For the record, right now, for one game, I would start Beckett.  If we are simply trying to decide who is better.

Josh Beckett, age 22.
108 innings, 7.8 H/9IP, 3.8 BB/9IP, 1.08 HR/9IP, 9.5 k/9ip

Harden, age 22.
190 innings, 8.1 H/9IP. 3.8BB/9IP, .75 HR/9IP, 7.9 k/9ip

so Harden was clearly better in both innings pitched and homeruns allowed, while Beckett had a better K rate, and the two were roughly even in hits and walks allowed.

Beckett sort of broke through in his age 23 season, but he didn't pitch many innings, and then he regressed last year to the point where his VORP (28.3) was well below Harden's (41.3).

Beckett is quickly gaining the reputation as that of an injured pitcher (if he hasn't already).  If you can only count on the guy for at best, 150 innings per year, that's asking for trouble.  I don't know how you can quantify the health/potential factor, but Beckett clearly hasn't met his.  And as an added bonus, his K rate is decreasing year by year (of course, his BB rate is, too).

to suggest that the two are not even close is just plain asinine.  

had a GB/FB ratio of 1.33.  His HR 9/IP ratio was a paltry .57

what does this mean?  Well, as Lord Helmet once said, "absolutely nothing," since the GB/FB rate has little to do with the expected number of homeruns allowed.

for the record, Beckett was at .92 last year, while his GB/FB ratio was right at league average.

seriously, the GB/FB stat is about as relevant as batting average.  

wasnt it you who said:

"If you're going to try to use stats to prove your point, at least use the right stats.  Don't just cherry pick the ones you want, and then completely misrepresent others."

basically thats what youre doing. for you to say gb/fb ratio is useless is absurd. not only to pitchers with higher gb/fb ratios usually give up less homers (which always is a good thing), a pitcher who knows how to keep it in the infield is, at the very least, a smart pitcher.

i like how you somehow vaguely connect beckett's gb/fb ratios with the statement "gb/fb is as relevant as batting average". good job pulling stats to help your argument and ignoring others.

by the way, i still would rather have a guy how can hit 300 than a guy who his 220. wouldnt you? then obviously batting average tells you SOMETHING.

I tend to agree with ya, David. The onus of proof is upon harden himself. I wouldnt bet my money on it, either. I'd hope that Harden can convince me otherwise with a strong season.

unfortunately, ESPN doesn't distinguish between line drives and fly balls because ESPN is a worthless pile of steaming dung.

but hitters don't the ball hard off harden.  that's why he doesn't give up many line drives.  it's the same thing with josh beckett, too.  both are good at missing the sweet spot of the bat.

If you took one of these young pitchers and created not just one list, but a series of SIMS projections (one after every season in the majors) would a critical mass of similar pitchers appear on every list (suggesting a similar development curve) or would they be just random lists of changing names?

PS Good to see some love for Fresno's own Jim Maloney. If either Beckett or Harden has a career as good as his they should consider themselves fortunate.

"jdelavalle ((Pembroke Pines, FL)): Hey Rany, I don't know how well you can answer this question, but how does PECOTA like in the long run pitching prospects that rely heavily on groundballs? I am talking specifically about pitchers like Edgar Gonzalez, or Sean Burnett to a lesser extent. Are these guys that get better as they age? Thanks.

Rany Jazayerli: You'll need to ask Nate about the inner workings of PECOTA, but one of the best parts of doing the Prospect list is seeing how PECOTA reacts to different sets of players, so I can answer this to some extent. Groundball/flyball ratio is definitely a significant factor when evaluating pitchers; generally speaking, the higher the better. As Nate found last year, G/F ratio actually predicts next year's HR rate better than this year's HR rate. So this becomes important when a prospect gives up few homers but has a fairly low G/F ratio; his homer rate is likely to increase the next season."

I don't just make this stuff up.

http://www.baseballprospectus.com/article.php?articleid=2885

A few quick highlights:

As I discussed in the PECOTA essay in this year's book, groundball induction--or, more specifically, groundball-to-flyball ratio--is remarkably consistent from year-to-year, more so than virtually any other measure of pitching performance.

Where groundball rate is valuable in serving as a proxy for predicting home runs. The number of home runs allowed has a very large influence on a pitcher's ERA. At the same time, home run rate varies quite sharply from season to season--I won't run another graph, but the correlation coefficient on HR% for the pitchers in the sample above is just .25. This phenomenon can be thought of a corollary of sorts to DIPS theory, which suggests that pitchers have little influence on what happens to a batted ball once it is in play. Though DIPS theory explicitly excludes home runs from consideration, it is proper to say that a pitcher can influence (through pitch location and pitch type) whether a ball is hit in the air or on the ground--but there's considerable luck involved in determining whether a ball hit in the air will turn into a home run, a warning-track shot, or just a routine putout for the right fielder.

I don't want to step on Nate's work anymore so than that, so if you want to read the groundwork of why he reached those conclusions, a BP subscription would be a great investment.  

1. I have a hard time comparing the minor league stats of these two players.  Beckett's stats are better, but he did them when he was a year older than Harden... Harden was more advanced at the same age.
2. I am concerned about Harden's walks.  They were high in the minors and they are still high.  A high walk rate + a high K rate makes it hard to go deep in to games (otherwise known as "Kerry Wood-itis"), especially when your organization cares about pitch counts.  This, in turn, makes it harder to win games.

1. I admit it, I am probably affected by the hype surrounding Harden, which is fresher in my mind than the hype surrounding Beckett which has partially subsided.
2. I don't have confidence in the Florida staff to take a good young SP and develop him further (see: Livan Hernandez, who may never have been a "great" prospect, but who had to leave to improve).  On the flip side, I trust the Oakland staff, even without Peterson, to bring Harden along.  
3. I prefer the Oakland bullpen - a great safety net.  
4. Beckett has shown himself to be more injury prone.

 The correlation between home run rate in year N and groundball rate in year N-1 is -.345. Though the sign proceeding the correlation figure is negative (since a higher groundball ratio tends to predict a lower home run rate), the magnitude of the correlation is a bit higher."

---------------------------

like I said before, it's a weak correlation.  in statistics, any correlation between -.5 and +.5 is considered weak, and not very much should be inferred from the results.  

Nate Silver is speaking in relative terms.  yes, .34 is higher than .32, so he's correct in saying that GB rate may be a better predictor than the actual home run rate itself, but not by much, and both are very weak correlations.  I also wonder what the variance, standard deviation, and mean squared error of his study were.  I would be willing to bet money that the tiny differences in the two numbers can be explained completely by random chance.

"A simple regression model that uses home run rate in year N as the dependent variable, and home run rate in year N-1 as the independent variable, is capable of explaining only about 11% of the variance"

that's another direct quote from that same article.  if that's explaining only 11% of the variance, it tells you that you can pretty much throw out those results.  I'm not sure why this article was even published.  If I had attempted to turn this in to any of my stat teachers, I'd have been laughed out of the classroom.

the correlation between extra base hits on balls put in play and GB/FB ratio is -.01 (courtesty USS Mariner 4/12/2004.)  that means there is no correlation whatsoever.  giving up ground balls or fly balls has no impact on whether a pitcher is giving up extra base hits.  

LAST     G   GS       IP     H    HR    BB  K
HARDEN    31  31    189    168    14    81  172
BECKETT    26  26    161    142    14    54  158

Harden 8.0 H/9 IP, .67 HR/9 IP, 3.9 BB/9 IP, 8.2 K/9 IP
Beckett 7.9 H/9 IP, .78 HR/9 IP, 3.0 BB/9 IP, 8.8 K/9 IP

I would take Harden over Beckett. I just think he will lower BB rate. This would allow him to increase his value. It's all a crap shoot anyway with pitchers.

Age 22 VORP:

Harden: 41.3
Beckett: 9.8

2004 VORP:

Harden: 41.3
Beckett: 28.3

Harden was over 4 times more valuable than Beckett when both pitched their age 22 seasons.  Harden was about 50% more valuable than Beckett in Harden's only full season.

And Beckett's so-called "breakthrough year" in 2003?  His VORP was 35.8, still not as good as Harden's first full year.

Harden was better at age 22 than Beckett has ever been in his career.  You simply can't ignore that.

You can take Beckett.  But I'm taking Harden.  I'll see you at the Hall of Fame.

From statistics such as BB/9, HR/9, and K/9 you can come up with a more accurate predictor of future ERA than by looking at past ERA. Given Harden's numbers, you would expect a pitcher to have an ERA a quarter run lower.

Except, like with every other point he made, he was wrong.  

The league average DER for the American League was 69 percent.  Harden's was 71 percent.  Harden got better than league average support from his surroundings (since DER is affected by fielding, ballpark, and luck).  

He also talked about Harden not giving up home runs because he gives up a less than average amount of line drives.  Harden's line drive percentage was 17.7 percent.  The league average was... 17.7 percent.  

Facts are facts, whether Mr. Waffle wants to believe them or not.  Harden benefited significantly from factors beyond his control in 2004.  Unless he actually pitches better than he did, it is unrealistic to expect him to be able to sustain his ERA going forward.

I also came across a statement in this debate about blisters, and how they had never been a long-term problem. This is not exactly correct, as Ismael Valdez's career was pretty much torched by blister problems.

I see Harden as a bit of a safer bet, and that's probably why I would lean that way in a general sense. Ironicly, if I were the A's, I would probably prefer the performance Beckett is likely to give next year, as their bullpen should be both deep and excellent, and therefore easily able to make up the 50 inning difference between the two with acceptable performance levels.

Younger
Smoother mechanics - and still throws harder
Durability - No injury history
Good breaking ball - just doesn't have to use it as much because he spots his fastball in/out so well.  Hitters know about it though.  Beckett does have a better one though.

Beckett's post season accomplishments are impressive.  But his mechanics seem a little less stable, although he has a tremendous arm - I think the force he throws that curve ball with is probably what is hurting his elbow.  It seems like he really leads with that elbow when he throws it, or any pitch.  Didn't he have problems either before or after he was drafted?? Or am I thinking of Robbie Beckett?  I got them confused for some time.  They're both from Texas I think.  

They're both (Harden/Beckett) excellent as long as they stay healthy, but Harden has a much better chance of doing that.  Partly because he isn't put through the Marlins meat-grinder pitch counts, and his mechanics/arm motion is smoother.  

I'm not a pitching coach or anything, but I did pitch for a couple years in college.  Also if you're into pitching mechanics, 'The Pitching Edge' by Tom House is very good.  I think of it as 'The Science of Hitting' except more modern and for pitchers of course.

2001.  74 IP, 3 HR,  .36 HR/9IP
      
2002. 153 IP, 6 HR,  .35 HR/9IP
      
2003. 177 IP, 11 HR, .56 HR/9IP
      
2004. 195 IP, 16 HR, .73 HR/9IP

yeah, I guess I take it all back.  Harden is historically a guy who gives up a ton of Home Runs.  I mean, look at those rates.  Could those numbers be any higher?  What an overrated schlep that guy is.