Interesting post about Holliday's home/road splits at Bay Bridge Baseball. (Hat Tip: Faketeams.com).

Blogger Jeff Fletcher argues that Holliday is hurt on the road by not seeing pitches that break normally in the thin Coors Field air. But as Holliday got further into his road trips and adjusted to the normal action of breaking pitches, his numbers improved significantly. Here's the chart:

Matt Holliday's career road numbers

First 2 games of a trip: .277 BA/.472 SLG
3rd game on: .281/.446
4th game on: .281/.458
5th game on: .291/.483
6th game on: .305/.522
7th game on: .321/.567
8th game on: .356/.683

So you might want to go the extra buck on Holliday after all.

Update:

Jeff Erickson wrote a great, in-depth article about Holliday's prospects for success a couple weeks ago. Here's the bulk of it:

As should surprise no one, the average player declined in performance after leaving Planet Coors:

|STAR|All stats cited here are courtesy of Baseball-Reference unless noted otherwise. The source for OPS+ is noteworthy, because they normalize for park-effects.

Because there are half-seasons involved, I'm more concerned with the rate stats than the counting stats, though it's still instructive to note how homers declined.

There's one more filter that I wanted to add, however. Because of the nature of player we're discussing, I wanted to get rid of the below-average hitters. After all, lousy hitters at Coors are going to be likely to be lousy anywhere. It's not especially instructive to see what Mike Lansing did after leaving the Rockies (the Laser retired after his one full season in Boston), or what happened to a 37-year old Todd Zeile, split between the two New York teams. So I took out all samples where the player in question had an OPS+ below 95 in his final full season in Colorado, and the subsequent season elsewhere. Say goodbye to Royce Clayton and his ilk. I'm not sure I set it at the right level, to be honest, because it filters out Vinny Castilla's first walk year, when he somehow managed to have 33 homers, 102 RBI and an 83 OPS+. But I didn't want to cherry-pick seasons any further than I was already doing.

This result surprised me - I didn't expect the drop-off to be so stark. Yes, the elite players as a whole were a season older, making it more likely that they'd hit a cliff, and indeed, in many cases those players did hit a cliff. Looking at the players individually from this sample is somewhat interesting:

Before

After

Remember earlier where I alluded to certain post-Rockies seasons sticking out as counter-examples of the post-Coors Field malaise? It turns out, among those that were coming off of above-average seasons, only Andres Galarraga, Ellis Burks and Larry Walker were really able to overcome that park change. (Adding the first Vinny Castilla departure would have further skewed those numbers.) In particular, Jeff Cirillo and Jay Payton really tanked. It takes a really special hitter to overcome the change in parks. Certainly other extenuating circumstances existed (Preston Wilson shredding his knee comes to mind), but I think this paints a pretty clear picture.

So, does Matt Holliday fall into that special class of hitter as Galarraga, Burks and Larry Walker? Let's look at the other factors.

2. Home/Road Splits

Holliday's career home-road splits aren't that encouraging. Since his 2004 major league debut, he has hit .357/.423/.645 at home, with 84 homers. In that same span on the road, he's hit .280/.348/.455 with 44 homers. The road numbers do get a little better the last two years:

Home/Away

Keep in mind when looking at Holliday's 2008 numbers that he only played 139 games, missing games due to hamstring injury early in the year and back stiffness in September once the Rockies had been eliminated. So far there's been no report of any lingering back trouble, so his projection doesn't factor that in as a negative factor.

The improved road numbers the last two years are encouraging. I believe they're a better indicator of what he's capable of than his career road numbers, particularly after factoring in his improvement in other areas of his game - he significantly improved his walk rate (.07|PERCENT| in 2006, .09|PERCENT| in 2007, .12|PERCENT| in 2008) while maintaining his contact rate (.82/.80/.81). If you extrapolated from his three-year weighted road numbers, that might be a good start.

But the analysis doesn't quite stop there. He'll be playing half his games in McAfee Coliseum, and we noted what a pitcher's park that is. Unfortunately, that only scratches the surface. It's especially tough on right-handed power, indexing at 83 for RH homers in 2008, according to the 2009 Bill James Handbook. Holliday |STAR|will get hurt by his new home park, undoubtedly. But on the flip side his road numbers should improve a little. Keep in mind that with the Rockies, he's not getting Coors Field among his road numbers, but also the plurality of his road games were played in the NL West, home of three traditional pitcher's parks (Petco Park, Dodger Stadium, and Telecom-of-the-Month Park in San Francisco - though the latter didn't play as much that way in 2008). The fourth ballpark in the division was a hitter's park, but also featured a home team rotation that included Brandon Webb, Dan Haren and Randy Johnson last year.

For comparison's sake, here were the home/road splits for Galarraga/Burks/Walker in their respective final full seasons in Colorado:

Galarraga's '97 splits aren't as stark as Holliday's in '08, but the others are rather similar, and it's not just in their final year in Coors Field where they had that sort of disparity. Holliday's 2008 road numbers compare quite comfortably with the elite class of ex-Rockies. If anything, looking at his track record on the road is encouraging, not discouraging. I'm more concerned with the identity of his new park than I am that he's leaving Coors Field.

3. Age

The average age of the departing Rockies regular was 32, and the average age of the filtered elite was 33. In fact, of those in the latter group, only Todd Walker was younger than 30. The generally accepted peak age of a ballplayer is between 27-29. Of course, there are plenty exceptions, but I'm comfortable enough with the research done on the topic to suggest it's likely that Holliday is still in his prime, and hasn't yet hit his decline phase. It follows that he would be more likely to adjust to his new surroundings than others that have left Coors Field.

4. Contract

Because Holliday is in a walk-year and is a Scott Boras client, the A's aren't likely to sign him to a long-term deal during the year. This could have multiple implications. The obvious one is that he could become trade-bait around the deadline, if the A's don't contend this year. They're spending (or, in the case of Rafael Furcal, attempting to spend) as if they believe they have a chance this year in the AL West. That might be the case, particularly after the Angels failed to re-sign Mark Teixeira. But if they do fall out of it, chances are Holliday could get dealt to a contending team if the price is right. That would take some of the sting out of Holliday playing his home games in McAfee Coliseum (see above), and would come with the probable added benefit of inserting him into a better lineup. The other implication is what happens to Holliday from a motivational perspective. We've seen plenty of examples of how a player can raise his performance during a walk-year, even if there's no one-to-one correlation. Not every player reacts the same to that motivation, but on the balance it might work in Holliday's favor.

5. Stolen Bases

Holliday ran more often and more successfully than at any other point in his career, converting 28 of his 30 stolen base attempts. There's a really good chance that his stolen bases decline - the question is by how much? Consider that his previous season-high was 14 steals, and also consider that the A's are among the least likely franchises to encourage him to run. My initial projection of 19 steals might be a tad optimistic.

Initial Projection and Conclusion

I had thought about all of these factors when coming up with Matt Holliday's projection, though I hadn't quantified just how badly departing players declined in the first year away from Coors Field. Here's the initial projection for Holliday:

Before doing this exercise, I felt pretty comfortable with the end result and where that put him overall compared to other players - 11^th on the RotoWire 200. Now I'm not so sure that I didn't project too much power for him, though I'm still happy with everything else. My next projection for him will likely take away a handful of homers, to give him a slugging percentage somewhere in the .505-.515 neighborhood, and drop him down a few pegs in the overall rankings. He's still a top-24 player overall for me, but perhaps not quite as high as before.