September 11, 2008
Following his recent campaigns, Cliff Lee's 2008 season looks to have come completely out of the blue. What began as a productive year that many thought would fade with more innings has instead withstood the test of time, at least within the limits of this single season. What many people have forgotten is that there was a time when Cliff Lee was considered a quality prospect in the Expos' and Indians' systems, before fatigue, an inability to translate his pure stuff into results, and injuries all derailed what seemed like a promising career. Today we'll have a go at what we might expect from this new-look Lee going forward.
Clifton Phifer Lee was selected three times before finally signing with the Montreal Expos in the 2000 amateur entry draft; the lefty had also been chosen by the Marlins (#246, 1997) and the Orioles (#609, 1998) before signing as an Expos fourth-rounder (#105) and heading to Cape Fear in the Sally League. While Lee's walk rate (7.4 per nine) was atrocious during his 44 innings there, he did strike out nearly 13 batters per nine innings. The 22-year-old would spend the entirety of his full-season debut in 2001 at High-A Jupiter in the Florida State League, where he would put together a much more impressive showing: Lee struck out 10.6 per nine, walked a much more acceptable 3.8 batters per nine, and proved difficult to hit, giving up just 6.4 per nine at a level of play where poor defensive support makes that difficult. If not for a stiff shoulder that forced him to miss a month, he most likely would have walked away with the league's ERA title.
Baseball America would rank Lee the 11th-best prospect in the then-stacked Expos organization-this is an organization that had Brandon Phillips, Brad Wilkerson, and Grady Sizemore as their top three-mostly based off of his pure stuff and his recent FSL performance. According to BA, the Expos believed that they had gotten a steal in the fourth round when selecting Lee, as they felt he was one of the top three college southpaws in that year's draft class. Lee was known for having four above-average offerings, but his problem to that point had been consistent command of all of them; here was a pitcher who could struggle due to his lack of command, or absolutely dominate you when he was on. Baseball Prospectus 2002 discussed that the Expos wanted their young hurlers pitching to contact, and that despite this, Lee led the FSL in strikeout rate, "showing how tough he is to hit."
As for the 2002 season itself, Lee moved up to Double-A Harrisburg and continued to show off his stuff. He struck out 11 per nine while lowering his walk rate to 2.4 per nine, and though an increase in home runs (1.3 HR/9) was worrisome, he once again used his strikeouts and the Expos' pitch-to-contact style to hold the opposition to 6.4 hits per nine. Those low hit rates probably didn't help Lee in the long term, though, as they made him appear to be a more accomplished and polished pitcher than really he was at that stage.
Lee was a piece in the blockbuster deal that Nationals fans might have nightmares about (who knows what Expos fans think these days), because that was the exchange where Grady Sizemore, Brandon Phillips, and this Cy Young candidate were all swapped for a few months of Bartolo Colon (and Tim Drew). The Indians plopped Lee onto their own Double-A affiliate for all of three starts, then moved him to Triple-A. He didn't set the world ablaze there, striking out just 6.3 per nine and walking 4.6, a big increase on his previous marks. Despite this, the Indians called him up to the big leagues at the end of the year, where he posted a 1.77 ERA in 10
Baseball America was excited be Lee's potential, ranking him the third-best prospect in the Indians' stacked farm system-they'd ended up stealing most of the best of the Expos' system, adding Phillips, Lee, and Sizemore to Victor Martinez, Jeremy Guthrie, Travis Hafner, Jhonny Peralta, Ryan Church, Coco Crisp, Ben Broussard, and Josh Bard among their top 30. Baseball Prospectus 2003 pegged Lee perfectly:
...Lee has a variety of brutal pitches ... and he can throw all four for strikes. Which is not to say that he does throw them for strikes ... but it happens. Lee is pure stuff at this point, with his control coming and going. He's a guy who could find consistent command and be a great pitcher, or he could be one of the majors' flakiest starters and would still be a guy I'd buy a ticket, and a OSHA-approved hardhat, to go see...
Lee would begin the season in the minors after straining his abdomen during camp, and in spite of pitching in pain-he required hernia surgery after the season ended-he did pretty well for himself. He was able to improve on his strikeout rates from his last time at the level (8.7 per nine), but his walk rates (4.4 per nine) still left something to be desired. Lee was called up to the majors and slotted into the rotation on August 16, and he would pitch well, except for two starts where he was forced out early. Baseball Prospectus 2004 said he could end up, "somewhere between Steve Trout and [Sid] Fernandez, either merely succeeding or flat-out torching the league, but it will be fun to see which way he goes, starting this year."
Sadly for Lee, he was the one getting torched during his first full year in the majors. He threw 179 innings, and though the front half of those went well (3.77 ERA, 7.3 K/9, 4.3 BB/9, and 1.0 HR/9 in 107
Essentially, we had a young power lefty coming into his own slowly during the first 153
Lee should have been better than he was that year, at least based on his quality walk and strikeout rates as well as his excellent stuff, but the homers were a problem that weren't going to go away unless his high fly-ball rate came down. That appeared easier to say than do; Lee was an extreme fly-ball pitcher, and the 2006-2007 seasons, the ones where his homer rates exploded, were his most fly ball-happy yet.
The first step in recognizing Lee's genuine improvement as a pitcher in 2008 comes from taking a look at his batted-ball data, as Lee has given the Indians something closer to the average G/F ratio, coming in at 1.3 rather than an extreme like 0.7. Unsurprisingly, dropping his fly-ball rate by nearly 15 percent also means that his homer rate has cratered, as he's given up just 0.4 per nine on the year, and just 3.8 of his fly balls have gone for homers. While that's so low as to be unsustainable, chances are good that this new approach that's earning him more grounders will keep him from the ridiculous rates of homers allowed he was suffering through before.
Lee also improved further on the command that he had developed during his time in the majors, and looking at his strikeout and walk rates, leaves you with the sense that he puts the ball wherever he wants to against the opposition. He's striking out nearly seven per nine, standard fare for him, but he's walking just 1.3 per nine. If we were to adjust his BB/9 to his career rate of 2.7, via QERA we would still have a pitcher deserving of an ERA in the 3.47 range; since QERA is based off of components that aren't as susceptible to luck as others-more on that later-it's something we can trust more than his current (or past) ERA figures. It would be nice for the Indians if he could post an ERA of 2.28 every year, but they should be satisfied with a legitimate ace in the 3.50 range as well. Amazingly enough, that 3.47 QERA is not all that far off from his 90th-percentile PECOTA forecast this year: that had him down for an ERA of 3.81, 6.1 K/9, and 2.9 BB/9 with 1.0 HR/9.
A power lefty who took some years to put things together-that's not exactly an oddity, given the tendencies of many power lefties from the past. Think of Floyd Bannister, Mark Langston, Randy Johnson, Frank Viola: all of those pitchers had to get over some hiccups before turning into the quality pitchers we remember them as. The career arc for a power lefty is something we may forget, as there aren't that many of them for us to analyze; without a constant stream of them to remind us of situations like Lee's, we may forget that they very well may turn into the pitcher we always thought they could become. Lee always had the stuff, and now he has the numbers working in his favor as well.-Marc Normandin
If last week we looked at Ryan Ludwick, a player whose success has been deemed a great story, then Cliff Lee's incredible performance this year is nothing short of unbelievable. For the record, I mean unbelievable in the very literal sense of the word, as the large majority of baseball fans simply cannot believe that Lee has been this good. In fact, one of my favorite things to do following a Lee start is to check for blog posts wondering whether or not he is "for real." What does that even mean?
My inclination is that being "for real" would involve a performance line more talent-driven than the result of luck. Is this the case for the Indians lefty? Well, I am forced to give an ambiguous "yes and no" answer, since there are many aspects of his success directly related to talent or a change in approach, but there is at least one red flag that should regress either over this next month or next season.
In April, Lee was performing better than Pedro Martinez circa 1999-2000, with numbers largely built upon a ridiculously low-unsustainably low-sub-.200 BABIP. He had always struck batters out at a respectable rate, but his rates of walks and home runs were high enough to prevent him from being a very successful pitcher. That inaugural month saw Lee walk next to nobody and vastly limit his home runs, allowing all of... well, one. The extremely low BABIP was begging for subsequent regression, so he naturally had a significant number of doubters.
As more starts passed and more statistics accrued, it became clear that Lee had turned himself into a different pitcher. He'd sustained high strikeout rates in the past, but now he was doing so while limiting his walks to an extent that his K/BB stayed above 4.5, steadily rising to its current 5.5. So, he was striking batters out, avoiding free passes, and keeping the ball in the yard, all controllable skills for a hurler, and elements of performance that are that much more impervious to luck than success based on balls in play (ie, based on how well his teammates field). His BABIP soon regressed to .298, right around the league average, but lo and behold, he still kept delivering tremendous performances. Even his rate of stranding runners normalized to a still above-average but sustainable 78 percent.
These signs pointed towards this being a legitimate improvement, but there's still the question of why he's improved. For starters, one of the first numbers to stabilize are rates of balls in play. In every year prior, Lee had produced ground-ball rates between 33 and 36 percent, and fly-ball rates around 45 or 46 percent. This year, however, his ground balls are at a career-high 46 percent, while his fly balls allowed are at a career-low 35 percent. On top of that, his rate of home runs per fly ball is at a (likely unsustainable) four percent. Given that the league average hovers around 11 percent, this is one luck-based indicator that is sure to regress over this final month, not to mention next season. However, along with this rate-based improvement, in terms of what he's doing on the mound, his fastball has gained velocity, but he's also mixing in more curveballs.
With all of that in mind, I feel confident that Lee can continue being successful, but it is not very likely that he will post numbers like this next year. The common misconception is that this slight decline would mean he is not a good pitcher, but nothing could be further from the truth. If he truly has become a different pitcher with a different approach, based on his location, sequencing, and an improved ability to keep the ball on the ground, then he would have the ability to post solid numbers in the future. For all we know, his luck-based indicators could lead to some unlucky runs next year that, when coupled with a higher home-run rate might inflate his ERA, but Lee's performance this year has definitely been more due to a change in approach than luck.-Eric Seidman
When people ask me to explain the difference between Cliff Lee circa 2007 and the one we're seeing now, I have two answers. The first one is based on pure scouting, and is therefore scientific and easier to understand, and perhaps even to agree with. Beyond Lee's much-improved physical conditioning, the Indians simplified Lee's mechanics by focusing on getting the left-hander to use the same release point on all of his pitches, and achieving what he does with his secondary pitches with only grip and arm action. That's the explanation for improved command-one delivery to commit to muscle memory, one delivery to perfect and to gain consistency with. That aspect of Lee's transformation has been a rousing success.
The second aspect is a bit more abstract, and not something that can be explained with any kind of metric or spreadsheet. I still remember a conversation that I had with a scout, and while I don't even remember who we were talking about, the point stuck with me. A significant percentage of pitching has little to do with radar guns, but rather involves fearlessness and confidence. Lee had very little of either last year, and this year he started out well, built them back up, and they have soared from there. Believe it or not, sometimes that can mean all the difference in the world. He doesn't have monster stuff, but depending on how you count pitch variations, he has up to five offerings, as well as the confidence to use any of them (four-seam, cutter, curve, slider, changeup) at any point in the count, and in any situation. He's not aiming for the perfect spot any more; he's simply aware of where he wants each pitch to be and letting it fly. He recognizes what's working and what's not in any single start, and he now makes adjustments to accentuate the positive and eliminate the negative on a start-to-start basis instead of pitching to a template. He's pitching smarter, and he's now attacking hitters as opposed to defending against them. It's one of the hardest things in the world to teach a player, and it may sound like a bit too much of that old-time religion, but that's been the real key to Lee's remarkable season. It's much deeper than Crash Davis' advice-"don't think, just throw"-that works when you have closer stuff in your arsenal. This is more, "Think, but not too much, and just throw."-Kevin Goldstein
Eric Seidman is an author of Baseball Prospectus.