Catching Up
Sorry about the lack of posts the last few days – Derek has been pimping his book down in Arizona and hosting pregame fun with Bill Bavasi and Kevin Towers, while I’ve been taking care of some photography business, Jeff’s hanging out with sunsets in Okinawa, and Jason was graduating from the CIA – the cooking one, not the espionage one.
Even with multi-author blogs, occassionally we’ll all get busy and you’ll get a day or two without a thread. Thanks for not burning the place down in our absence.
So, what have we missed;
The team is beginning to build faith in a resurgent year by… losing every spring training game they play. They’re now 0-5 in games that don’t count. No, this doesn’t matter at all. No, you shouldn’t care. No, this isn’t evidence of Mike Hargrove’s inability to motivate his players. It means nothing. Really.
The team has setup the rotation so that King Felix gets the opening day nod. To which I say huzzah. There’s a good sentiment behind protecting his young arm, but your best pitcher should be rewarded with the first start of the season, and there’s simply no way to not ackonwledge the fact that Felix Hernandez is the Mariners’ best pitcher. To give the honor to a mediocrity like Jarrod Washburn isn’t going to help Felix. Developing bitterness in your franchise players isn’t protecting them.
Jerry Brewer wrote a column about how Richie Sexson is misunderstood. He’s not though – we understand perfectly well that he’s a league average first baseman making $14 million dollars, and he’s a competitive disadvantage that the Mariners would have been wiser to have dumped and spent the money on better players. No misunderstanding – he’s just not that great of a player, and his salary his one of the reasons the Mariners continue to spend more money than other teams and finish 10 games behind them in the standings.
Geoff Baker continues his entertaining series of blog posts, including this little nugget about his article on Mat Olkin
What “Predicted ERA” does is essentially the same thing for pitchers. James did the bulk of the legwork, but Olkin finalized the product, which, by the way, is how roughly 99.9 per cent of all inventions in this world usually wind up happening. My goal is not to turn this blog into a stats-laden think-tank that will bore the heck out of the average reader. But there are certain stats in baseball that are very important to the outlook of a team and the Mariners did hire Olkin on a freelance contract basis in 2005 so I think even traditionalist, stats-phobic fans will learn something here.
It’s great to see that something like Predicted ERA can get a writeup in the local fishwrap, considering it’s far from a mainstream stat. It’s great to see Baker acknowleding that ERA can vary wildly from a player’s true talent level, and pointing out that this tool exposed Jarrod Washburn as a fraud from day one. However, and this isn’t a knock on Baker as much as it is a knock on the Mariners, Predicted ERA has been made obsolete by much more effective statistics that do the same thing, only better.
As I wroteup in my Evaluting Pitcher Talent post, looking at things on a macro level is going to lead to incorrect assumptions because of the difficulty of separating out influences of the pitcher, the defense, and the park. Predicted ERA (OPS/31, for those who haven’t read the article) doesn’t adjust for any of these things.
Mat Olkin’s a smart guy, so hopefully he was just throwing Baker a bone and talking about a stat that he helped create a while ago to give him good column fodder, because if the Mariners are really leaning on Predicted ERA to evaluate pitchers while the rest of baseball has moved on to far more accurate statistics, it’s just another sign of how far behind the times this organization is.
However, Baker’s willingness to interact with his readers and throw bones to the statistical community is a stunning reversal from the Pocket Lint era, and it’s great to even be able to have that kind of discussion with the Times beat writer.
Oh, and I know I’m behind on the Community Projection updates – the series is still going on, and if you’re on the list, you’re still getting the emails – I’ll post a recap of the results of the last few tonight.
I didn’t know you where a photographer, nice work Dave.
Anyways, thanks for the update!
My Father was a professional photographer/printer. He photographed the opening of Asahel Curtis park in the Cascades years ago. When he passed away in the mid ’60’s, my Mother had to sell all his stuff, which included about 5-6 incredible Speed Graphics. Good luck on your career, you have talent (too bad you can’t pitch…)
Speaking of evaluating pitchers, there was an interesting article posted recently on THT about evaluating pitchers based not necessarily on strike % or K/9 but rather by swinging strike, which has a correlation of over .6 to K/9 while called strike % basically has zero correlation (which can sometimes be used to explain flutuations in K/9). One of the examples it used was Putz, who’s swinging strike % went way up last year, a sign that his K rate is sustainable.
How does one get on this mysterious projections list?
Wow, the Mariners are already 0-5 in Spring Training.
Mariners baseball. Catch the Fever!
In principle, I agree with you on Sexson, Dave. His production from 1B is nothing special.
However…
A surprising number of AL teams are taking the “throw mediocre players out there and see what sticks” approach to first base this year. If he doesn’t experience significant decline, Sexson is a better than league average first baseman this year.
How many true slugging first basemen are there in the AL this year? I count:
Morneau
Konerko
Teixiera
Giambi
Sexson
Am I missing anyone?
Offensively, there isn’t another first baseman out there I can think of who’s any kind of serious threat to come close to Sexson’s production. Overbay, Blake/Garko, Casey/Shelton, Youkilis, Millar/Gibbons/Huff, Minky/Phelps, Wigginton/Cantu, Kotchman/Morales, and Shealy……Only a couple of those are reasonable bets to even provide 80-85% of Sexson’s value.
I’m not justifying Sexson’s roster spot – I think the team could have traded him for scraps last winter, rotated Broussard in, and played with the money to much better effect. I do think he’s better than league average given this year’s 1B crop though. I really don’t see him being worse than the 5th most valuable 1B in the AL offensively, and he should hold that spot down by a pretty wide margin. When you adjust for park effects, he’s pretty close to that elite group. He gives some of that back on defense for sure, but not everybody behind him who’ll have a starting job this year has a plus glove.
I think Sexson can be upgraded to “above average but not spectacular” relative to this year’s AL first basemen.
Given the number of mediocre pitchers getting big contracts, I’d be shocked to learn that many clubs
Oops… got cut off there. Anyway, I’d be surprised to learn that many clubs are using anything other than ERA and wins to evaluate pitching.
http://www.forbes.com/business/2007/03/02/sports-greatest-gms-biz-cz_jg_0302gms_2.html
According to Forbes we don’t have the worst GM in all of sports, only the 87th worst GM in all of sports. Being behind Isiah Thomas on such a list is something you could put on a resume.
#6- and you can also take Giambi off the list…he’s gonna be the DH in New York…so that puts Richie even higher on the production list…
A surprising number of AL teams are taking the “throw mediocre players out there and see what sticks†approach to first base this year. If he doesn’t experience significant decline, Sexson is a better than league average first baseman this year.
I’m not a big fan of adjusting to a league specific positional baseline. The population of major league first baseman isn’t limited strictly to the American League, and I think its a little more instructive to look at an MLB-wide baseline instead.
That gives us the following players who are clearly ahead of Richie Sexson in terms of value: Pujols, Howard, Berkman, Morneau, Teixeira, Hafner, Thome, Giambi, and Ortiz. The M’s would clearly be a better team with any of those nine players manning first base, and it’s not even particularly close. Sexson is in with the group of guys that include Konerko, Delgado, Johnson, Thomas, Helton, Overbay, Garciaparra, Swisher, and Gonzalez. You could argue over the placement of those next nine guys, but you’d essentially be splitting hairs.
That makes Sexson something like the 10th-18th best option for the M’s at first base currently in the majors. That’s basically league average.
Offensively, there isn’t another first baseman out there I can think of who’s any kind of serious threat to come close to Sexson’s production. Overbay, Blake/Garko, Casey/Shelton, Youkilis, Millar/Gibbons/Huff, Minky/Phelps, Wigginton/Cantu, Kotchman/Morales, and Shealy……Only a couple of those are reasonable bets to even provide 80-85% of Sexson’s value.
How many of those guys are making $14M this year? You can’t really speak about value to a team without considering contracts. By paying Sexson that much, the Mariners had to cut corners elsewhere, and that undoubtedly hurts the team.
#6- and you can also take Giambi off the list…he’s gonna be the DH in New York…so that puts Richie even higher on the production list.
This is a logical fallacy that a lot of people fall into. The Mariners would be a better team with Jason Giambi playing first base than Richie Sexson. The fact that NY is DH’ing him (or that Bostin is DH’ing Ortiz, or Chicago DH’ing Thome) doesn’t change that fact.
He’s part of the population of players that could replace Richie Sexson. Therefore, he’s part of the comparison.
#9: (Sportswhiz)
Bill is again ranked dead last in all of baseball on the Forbes list. Oh well, at least he edged out Matt Millen for overall “honors”. Nice work.
there are hundreds of players you could put at first base that aren’t first basemen…if you are going to compare “First base” production, you should only include first basemen…not guys who play DH or that are going to just be thrown into the job because they are too old, too slow, or too bad to play any other position…
Overbay outhit Sexson last year (.880 OPS to .842) and cost $12.5 million less (and Overbay will still be $10 million cheaper this year). Nick Swisher (.895) and the immortal Greg Norton (.864) outhit Sexson too for $14 million less. Blake (.835), Garko (.829), Huff (.813), Youkilis (.811), and Shelton (.806) were within 95% of Sexson, yet all were at least $10 million cheaper (except for Huff at $7 million cheaper). Sexson suffers even more when you do not limit the sample size only to AL, as the NL is stocked with quality first baseman.
The slight advantage Sexson may provide over some of these guys is simply not worth a $10 million premium. Sexson is league average, but if you feel better calling him one of the better league average first basemen, go right ahead.
that Forbes report isn’t about winning…it’s about money…McHale was #1 and the Timberwolves have never won a title and are terrible…
Could Ortiz really play 1B on a daily basis? If he could, don’t you think the Red Sox might try to get Manny off the grass?
Oh, and Overbay outperformed Richie by quite a bit last year. Those 35 points of OBP count for a lot.
But there’s no point in looking at where Richie ranks on a largely arbitrary list (who plays 1B right now). By playing Richie, not only are you paying him $14 million, but you’re also paying the opportunity cost associated with not playing any of those hundreds of other guys who could play 1B for you.
You’re not measuring Richie against other first basemen; you’re measuring Richie against other possible first basemen (because NONE of the other guys on the list are currently first baseman for the M’s, which really is the scope of our investigation).
you have to choose one though…you have to compare him to just guys that are locked into 1B or against the entire major leagues…you can’t just pick and choose who “might” be a first baseman…he makes a lot of money, most would say too much…but he gives us above average offensive numbers, against everyday first basemen, and should do well with the supporting cast this year…watching him strike out more than a geek at the prom is heartbreaking, but watching those grand slams last year was pretty sweet…i like the kid…lets stop bashing the guys and get ready for a better season than the last couple…it WILL happen…
you have to choose one though…you have to compare him to just guys that are locked into 1B or against the entire major leagues…you can’t just pick and choose who “might†be a first baseman…
No, you don’t. Not enough if you want to do real analysis, anyways.
Players provide value by being better than what other teams have. If another team has two guys who are better than Richie Sexson, they can only play one of them at first base – that doesn’t make the other guy non-existant.
This whole line of thinking is why I don’t like positional baselines anyways. Eventually, we’ll get good enough defensive analysis that positional adjustments will be a thing of the past, and we’ll just evaluate players on total runs added with the bat and the glove.
[Giambi]’s part of the population of players that could replace Richie Sexson. Therefore, he’s part of the comparison.
there are hundreds of players you could put at first base that aren’t first basemen…if you are going to compare “First base†production, you should only include first basemen
Well, there’s some question whether Giambi, Thomas, and Ortiz could really play 1B. And there aren’t hundreds of guys you could reasonably put there; perhaps 50 if you stretch the list to include players like Ibanez, Vidro, or Jeff Kent.
But if you’re combining the DH/1B pool of players, isn’t Sexson 10-18th of 44 hitters rather than 30? That would make him above average as a hitter (I’d say 15 or 16), but his defense pulls him down to average.
Jason Giambi played 68 games at first base last year. This isn’t some kind of hypothetical position switch, fellas. If you consider Richie Sexson a first baseman, then you have to consider Jason Giambi a first baseman.
Wow, the Mariners are already 0-5 in Spring Training.
Mariners baseball. Catch the Fever!
No, really, a spring training win-loss record is irrelevant. The Royals have won the Grapefruit League in years they finished last. The Yankees were notorious for not winning in spring training and then winning pennants the years they DO win. As I recall, the M’s were something like 12-17 during the 2001 Cactus League.
Basically, what I care about in Spring Training is a) do players come out healthy and b) do players look better or worse than you expected. Win/Loss is useless when you’ve got guys who should be in A ball pitching to guys who should be in AA in the 8th inning of a 3-3 game.
This whole line of thinking is why I don’t like positional baselines anyways. Eventually, we’ll get good enough defensive analysis that positional adjustments will be a thing of the past, and we’ll just evaluate players on total runs added with the bat and the glove.
On a related note, I’m increasingly dissatisfied with the notion of a replacement-level hitter. Except at DH, hitters don’t get replaced–players get replaced. There are certainly replacement-level players, but no one gets replaced just for hitting poorly if they’ve got a good enough glove. At some point, it kind of seems silly to call an great glove, no-hit SS a replacement level hitter if he’s not going to be replaced because his glove good enough that overall, he’s more valuable than a typical journeyman player.
Even if you do get rid of positional adjustments by accurately accounting for defense, it seems like it would still be useful to have an overall replacement level.
# NAME 2006 VORP
1 Albert Pujols 85.4
2 Ryan Howard 81.5
3 Lance Berkman 70.1
4 Justin Morneau 52
5 Nick Johnson 51
6 Paul Konerko 47.7
7 Mark Teixeira 37.4
8 Lyle Overbay 36.3
9 Carlos Delgado 35.2
10 Adam LaRoche 33.3
11 Adrian Gonzalez 32.8
12 Todd Helton 31.1
13 Nick Swisher 28.1
14 Nomar Garciaparra 27.6
15 Wes Helms 26.3
16 Richie Sexson 24.9
17 Ty Wigginton 20.5
18 Prince Fielder 20
19 Kevin Youkilis 19.6
20 Ben Broussard 18.3
#24– Seattle Times, March 05, 2001
MARINERS SPRING TRAINING AT A GLANCE
THE SCORE: Seattle 18, San Diego 3 at Peoria, Ariz.
THE RECORD: Seattle is 1-4 in the Cactus League.
THE ARMS: Starter Aaron Sele allowed just one hit and faced the minimum six batters over two innings. Rookie Jeff Heaverlo from Ephrata also pitched two shutout innings. For pitchers who threw in five or more games last year, the Mariners are 1-0 with a 0.82 ERA in five games this spring.THE BATS: The M’s put up 14 runs after having scored just 11 in the previous four games. Rookie outfielder Anthony Sanders went 3-for-5 with a homer, one of five homers in the game on the day for Seattle. David Bell drove in four runs, three of them with a second-inning homer.
THE GLOVES: When the game was still competitive, Mike Cameron made an over-the-shoulder catch in right center with a man on base in the second inning.
THE INJURY: 2B Bret Boone left the game after being hit on the head by an Adam Eaton pitch. Boone is not expected to miss much time as the injury is thought to be slight.
THE QUOTE: “It was nice to have everyone make a contribution and finally get a win.” – David Bell.
Dave, speaking of first basemen, any news on where or when Broussard may be dealt?
The team is beginning to build faith in a resurgent year by… losing every spring training game they play. They’re now 0-5…
[Do see # 24 and # 27] That inverse relationship between ST and the Regular Season – 0-5 ? World Series, here we come!
BTW, yesterday’s Triple Play brings back memories of last year’s Triple Play–a type that had never been seen (2-6-2)–which featured Ben Zobrist and the usual Mariner incompetence. On that same day (9/1/06) *, in another game, Kevin Kouzmanoff hit a grand slam on the first MLB pitch he’d ever seen. (Not often nowadays in MLB that two firsts happen the same day.)
__________
*http://tinyurl.com/2rpefl (See especially paragraphs 4-8.)
That’s the fundamental problem with WARP. While there are replacement-level hitters and replacement-level fielders, no one who ever plays in the majors is both. A freely available player (the sort replacement-level is supposed to describe) may well be a replacement-lever hitter, but then he’s likely to be an league-average defender.
#24: but what if most of the guys who are supposed to be in the M’s starting lineup are going oh-for? if that continues, would that be saying something about how likely hitters like Sexson are to get off to a slow start again?
#30: but by definition a replacement level defender IS essentially a league average defender….
I don’t think replacement level means what you think it means.
Replacement level offense is strictly defined…it varies by position but basically it’s around .85 of league average for 1b/DH; .80 of league average for everyone else but catcher and .75% of average for catchers.
Since the minor leagues are full of guys who are good enough for the majors defensively but not good enough offensively. replacement level for defense can be assumed to be league average for most positions and probably a little above league average for firstbase.
What part of the above don’t you agree with?
Concerning WARP, obviously it’s screwy in large part to how it handles defense.
Does anyone have any reports about how the feed went? Were there any juicy responses by Bavasi, or did he get away with the typical canned stuff?
The logistically-challenged East-coasters in the crowd would love a recap!
Speaking of defenses and stats, apparently THT has some defensive metrics up now. Dave, any thoughts?
Sure the minors are full of average defenders — but they can’t hit, which is why they’re in the minors.
I assume the intent of setting the bar at replacement level hitters at 85% of league average or whatever is that those are guys who play passable defense — or better — as well. But if you’re going to look at replacement level defense you ought to make the same assumption — that they can hit a little bit.
memmmmmmrrrrieeeeees
Drayer on Buher as coach, Baker on http://blog.seattletimes.nwsource.com/mariners/2007/03/when_losing_matters.html>when losing matters
crap.
Replacement level offense is strictly defined…it varies by position but basically it’s around .85 of league average for 1b/DH; .80 of league average for everyone else but catcher and .75% of average for catchers.
Just because it is strictly defined doesn’t mean that it makes perfect sense. Juan Uribe’s VORP last year was -0.6 over 495 PA. That makes him, according to standard terminology, a “replacement level hitter.” But you can’t just go replacing his bat without taking his plus glove away. That is, you can’t replace hitters, you can only replace whole players.
The way it would make sense to me is to take each player’s offensive value and defensive value and add them together. Then, after you have a sense of each player’s total value, look at the talent distribution, and see where the freely available talent lies.
By and large, it might not make a huge difference in player evaluation, but tweaking the terminology could get rid of some silly arguments where the two sides don’t really disagree. As it stands, it’s still a very useful concept, but it just rubs me the wrong way a little bit.
#41: I don’t think anyone is arguing that offense and defense shouldn’t be summed when considering a players value. The two get summed in discussions here all of the time.
VORP doesn’t consider defense. It’s unfortunate that systems like winshares and WARP that attempt to sum a player’s value relative to a similar standard (i.e. average or replacement level etc.) drop the ball pretty badly on the defensive side of things. As has been pointed out in the evaluating defense thread, there are much superior ways to measure defense than offered by BP or Bill James.
The point of replacement level is that such a player represents freely available talent. Players who can hit but can’t field that well tend to make the majors despite their defensive deficiencies. They become DHs or first baseman or position players for the Reds. However, players who can play competent defense but have trouble hitting don’t get the promotion unless they are truly exceptional with the glove. Hence, it’s much easier to find a competent defender who can’t hit in the freely available pool than it is to find a quality hitter who can’t defend well. Given the dynamics of the talent pool, I don’t think it necessarily follows that a replacement level defender should be assumed to have the ability to hit.
Pretty much a replacement level player is defined as a league average or slightly better defender who hits somewhere between 75% to 85% of league average depending upon the position they play. Probably the offensive level is somewhat arbitrary and more arguable but the defensive side to me seems pretty intuitive.
You guys are looking the wrong way at limiting who can play 1st. You look at DH type players and say “they couldn’t play 1b so we need to limit the pool of candidates”, instead you should be looking at all the MLB players that could play 1B for the team and base your comparisons on that. For example, 2 years a go Berkman wouldn’t be in the comparison because he played outfield. The M’s could have looked for an outfielder and shifted him to 1st instead of limiting their search for a firstbaseman to players currently playing the position. If you expanded your search to all the players currently in MLB “capable” of playing 1st base Richie drops even further down on the list.
That’s one of the mindset thinking problems w/ MLB GM’s. They limit their searches to the current set of players instead of expanding it to all potential players.
I bet Richie wouldn’t drop that far using that standard…he might even improve a bit percentile wise…. A strict application of the defensive spectrum suggests just about any major leaguer could play first base. Richie would be compared with shortstops/catchers etc by that standard.
I think you have to draw the line somewhere if you’re going use a positional yardstick.
Dave,
I agree with you completely from a roster construction standpoint. The team could have been much more creative with Sexson’s roster spot, and almost certainly would have come out better for it.
But, it’s about time for the season to start, and the way I was looking at it was how many runs the Ms will get out of their 1B position relative to how many runs every other AL team will get out of theirs. From a general baseball standpoint, I agree with you 100%, but I think Sexson will help the Ms more this year by being significantly better than the average AL first baseman. With Giambi moving to DH (not sure how I forgot about that but included the Minky/Phelps platoon), a healthy, non-declining Sexson is at least the 4th best first baseman in the AL this year with the bat.
I don’t think anyone is arguing that offense and defense shouldn’t be summed when considering a players value. The two get summed in discussions here all of the time.
Right, but not all discussions take place here. It’s overly complicated to explain replacement level the way it’s defined to someone who isn’t aware. I’m not saying that the wheel is broken and won’t turn, I’m just saying we can/should do better. A replacement level player shouldn’t be defined as someone who hits at the level of freely available talent and is an average fielder, but rather as a player whose overall value is equivalent to that of freely available talent. It’s a simpler, shorter definition that conveys more precisely what we mean intuitively by replacement level.
If we had accurate defensive evaluations all along, replacement level would’ve been defined w/r/t overall value in the first place. The main reason it’s defined the way it is now is convenience: we’re better at measuring offense than we are at measuring defense. So when Dave mentioned potential advances in defensive metrics, I thought of this.