What To Do About Roy Oswalt
We’re deep in the heart of Roy Oswalt Watch 2012, which is not quite a hullabaloo but is slowly ratcheting up the volume and frequency. Now the rumor is that he’ll go to the Rangers. Should he picked up in all leagues right now?
First, the obvious answer is that the deeper your league, the more likely it will make sense to pick him up. Before we go negative on Oswalt, he’s only 34 years old and he’s been very useful the last two years in the league. He’ll still throw over 90 and he’ll still have his slider and curveball, and he’ll still show great control in all likelihood. That takes a pitcher pretty far in deep leagues.
But how about your typical twelve-teamer? Then it gets a little murky, especially if he ends up in Texas.
Pick a stat, any stat, and Oswalt is in decline there. Velocity — down to 91.4 MPH from 93.7 in 2002. Strikeout rate — down to 6.02 per nine last year from an average just over seven. Swinging strikes — down to 8.0% and significantly below average for the second time in his career. Ground-ball rate — down to 45.1% from an average of 47.3%. FanGraphs provides a fan-tastic graph for the one stat whose decline is the most obvious:
One thing you’ll notice is that Oswalt’s xFIP (3.95) and SIERA (4.04) were worse than his 3.44 FIP in 2011. Both normalize home run rates to some degree, and Oswalt’s 6.3% home runs per fly ball last year was the lowest of his career. That number normalizes to 9-10% every year across baseball, and there’s no research that indicates that a pitcher can control it. His career number is 8.8%. He could expect an ERA much closer to four based on that statistic alone.
If the latest rumor is true, and Oswalt is headed for Texas, the damage could be worse. Our park factors have the Rangers’ park augmenting home runs by 9% (11% for lefties, 7% for righties). That should help push the home-run-per-fly-ball needle back to league average.
Not only will the park be harder on the pitcher, but so will the league. Various studies have found that the move from the National League to the American League can cost a pitcher half a strikeout per nine, or as much as 10% in ERA+. Oswalt had a 96 ERA+ last year and was exactly at league average in 2009. His strikeout rate was a full strikeout per nine below league average last year. Neither of those numbers can afford a healthy ding.
Back to our original caveat, those desperate for pitching in leagues of depth can reach for the free agent. He is certainly a skilled pitcher and will have some great matchups along the way. But for those of you in standard mixed leagues, Oswalt will have to ride your bench with the Yankees and Red Sox come to Texas — don’t stash him now if you can’t sit him much, or if your league is stream-friendly.
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.
(1) While Arlington has a park factor of 109, Philadelphia has a park factor of 104, so the gap isn’t that big.
(2) Texas plays NY or Boston 10 times from June 28 to the end of the season. Texas has 42 games against the worst offenses in the league during that span: Oakland, Seattle, Kansas City and Minnesota and another 34 against Los Angeles, Toronto, Detroit, Baltimore, Tampa Bay and Cleveland, only two of which play in hitters’ parks. The need to spot-start Oswalt may be less than you suggest.
(3) Oswalt wouldn’t have to pitch against the best offense in the league.
(5) Oswalt would be backed by a very-good fielding infield and a deep bullpen not prone to blow his wins.
1) The full range of PFs is only 22, so a difference of five seems like a big deal to me.
2) KC isn’t a patsy and the AL is, generally tougher. I’d be nervous to start him against NYY, BOS, TAM, DET, TOR at home, too.
3) Good point but tough to quantify exactly.
5) Sure, but mostly only one cat – wins – since unearned runs don’t go into ERA/WHIP. Some issues with longer innings, I guess.
hey rotofan, how’d that bitch slap feel?
hey! he had some good points. Overall, though, it’s not a good sign when a guy sits out the beginning of the season and can’t seem to get the contract he wants, plus the degenerative back problem, and the switch to the other league…
(1) A difference of 5 means the park factor in Texas is 5% greater than in Philadelphia. Since half of games are on the road, one would expect Oswalt home run rate to go up 2.5%. And 102.5% of 6.3% is 6.4575%. Assuming 100 fly balls (which is about right given the number of innings pitched and his historic FB rate) we’d be looking at 1 additional home run for every 1,000 fly balls. The other factor you hint at, regression to his career norm HR rate, appears likely to be a bigger factor, though even there, we’re talking about 2 or 3 more home runs over 100 fly balls.
(2) Kansas City is next to last in runs scored, edging out Oakland and averaging fewer than 4 runs a game.
(3) Certainly the AL is generally tougher on rate pitching stats because of the DH.
(4) Tampa and Detroit are pitchers’ parks. Both have had league average offenses (just worse than average, actually). Tampa has struggled with the loss of Longoria and Detroit has quite a few lineup holes notwithstanding Cabrera/Prince. I’m in a deep A.L. league and certainly haven’t kept my pitchers on the bench when they start in those two parks.
(5) The point about infield defense also goes to range – the Rangers will get to more balls than most teams, balls that would be deemed hits elsewhere, so it affects ERA too.
I wouldn’t used runs scored to judge offenses so far. by wRC+ the offenses I named are all above-average or better. The bad offenses he’d face in the NL are worse than the bad offenses he’d face in the AL, which is why there’s the 10% conversion factor.
I wouldn’t use last year’s home run rate as an indicator, because of sampling issues and the fact that the stat doesn’t stabilize in any one given year, but does so over a career. I’d probably take those numbers and apply them to his career rate and predict a 9+% HR/FB rate, which is a big difference from 6.3%.
The Rangers BABIP allowed this year is .268. The Phillies’ was .286 last year. So there’s some merit to that argument.
I think we’re use the same metric for HR/FB rate. Butthe difference between a 9% and 6.3% Hr/FB over 100 fly balls is 2.7 more home runs — which is why I wrote 2 or 3 more home tuns.
Oswalt no doubt carries some risk: His age suggests regression, his health isn’t entirely clear, he is moving to a tougher league for pitchers and Arlington is a tough park to suppress runs. I already have him on my roster because I’m playing in a 12-team AL only league with 17-man reserves plus a 23-man active roster — I picked him up for my reserve in March thinking most of the deeper-pocket teams who would need pitching are in the A.L. For a 12-team mixed league, he’s a much closer call.
In any case, I enjoyed our exchange.
Slow…clap…for two intelligent people having an reasonable discussion while disagreeing with each other. I think you might have broken the internet.
One question is how to weight Roy Oswalt’s degenerative discs? I would think an injury like that would affect performance projections.
What was not mentioned in this conversation was the distinction between the months of the year and Oswalt’s performance. Here are his FIP/WHIP splits per month throughout his career.
Apr – 3.72/1.22
May – 3.71/1.26
June – 3.28/1.28
July – 3.35/1.16
Aug – 3.35/1.16
Sept – 3.03/1.12
There are a number of possible conclusions one (and the Rangers) could draw from these (from least favorable to most favorable to the Rangers):
1. He takes a while to get going (which may be OK if he can get enough innings in him before the pennant push and/or playoffs) -AND/OR-
2. He pitches better based on the season of the year (Rangers would benefit in that they miss the first-half Oswalt altogether) -AND/OR-
3. He is a competitor and turns it up a notch down the stretch (Rangers could only hope this is true and seems to be some what the anti-thesis of traded-away CJ Wilson –> see CJs playoff performances and his monthly splits)
Interesting that chart showing a downward trend in average season velocity would show up on this site, when back in January Dave Cameron posted this chart showing Roy’s velocity on (I think) a weekly basis. http://www.fangraphs.com/blogs/wp-content/uploads/2012/01/Oswalt.png
In his article – http://www.fangraphs.com/blogs/index.php/roy-oswalt-potential-steal/ – he discussed Oswalt playing through injury from May through June and how after he returned from the DL in August not only were his results better (ERA, xFIP) but his velocity was also back up. If this explanation is correct then his sub-3.7 xFIP during healthy 2011 months gives him a lot of potential no matter where he pitches. There’s a ton of risk in that assumption, but that’s why he won’t cost you hardly anything to get.
Lastly, a prior comment mentioned the Rangers defense and that can’t be ignored. Roy could be going from a -10 team UZR last year with Philly to the team with a league leading UZR in the Rangers. The difference breaks down to a half a run per game average. That’s pretty huge especially for a guy like Roy who’s GB% is higher than league average. That won’t completely offset the park factors, but Roy should be a great start @AL West opponents and a good start depending on other matchups.
The potential is high and at this point in the season there’s not many free agents in most 12-team leagues that offer this kind of potential. In roto leagues I’d still stay away since he’ll likely need a while even after he signs (if he does) to get the velocity back up and get into mid-season form, but in HTH leagues where playoffs are where championships are won I think he’s a solid add.
That chart shows his K/BB, not his velocity.
My main point is that by the ERA estimators he was much closer to league average in the NL than most think, and that the move to the AL and the Rangers’ park will be bad for him, even if it’s mitigated by defense and bullpen.
His injury is also not one that just goes away.
It says it’s velocity: “Here’s Oswalt’s Pitch F/x velocity chart for the last few seasons: [chart] After spending his career sitting in the 90-95 range, he was more regularly 88-92 in the first part of last season. His diminished velocity corresponds directly to his two stints on the disabled list, and it’s pretty obvious that they are directly related…”
I do understand the injury concerns and that’s definitely not something to understate. However the point is a healthy Roy still seems to pitch very well, so if he’s healthy you could potentially get a #3-4 for your playoff run for almost nothing. We aren’t talking about spending a 10th round pick or trading away Andre Ethier to get Roy. In most leagues he’s a FA add that takes up your last bench spots for the few weeks – that’s all he costs.
Oh I thought you were talking about my chart. Yeah, as FA pickup, sure it’s okay, but the point of this was to point out his downside. So would I drop James McDonald for him? Not sure, he’s still in NL in great home park, way younger, trending better. Erik Bedard? sure I’d drop him. Question of degrees.
So you’d drop Billingsley for him?
I did not when I had the chance. But that was probably because I had Mike Minor to drop. I dropped Mike Minor, John Danks and Jarrod Parker (non keeper) for Oswalt. I did not drop Dickey or Billingsley for him.
I traded Carl Crawford for him… But I have depth at OF and my pitching is turr-bull. Plus my league values wins more than any other stat. 10 points per W, 0 points per QS, 1 per K, 1.75 per IP…
Edwin Jackson droppable for him?
I just did, but my league stupidly over-values wins.
Dropped Niese this morning for him.
Knowing the risks involved, I’m still considering the add since Weaver will be on the shelf for a little while (I hope). Any guesses on when Oswalt will be added to the 25-man?
Would I be nuts to drop Samardzija for him?
huh. i find it interesting you dropped parker (pretty good park, pretty good K rate, not the hardest of schedules all things considered). i get the other drops (personally, i dropped jake westbrook for him in a deep-ish 14 team league but i’m sitting on the likes of dillon gee in my 12 team league until i see actually see oswalt pitch).