Twinkie Town: An SB Nation Community

Navigation: Jump to content areas:


Pro Quality. Fan Perspective.
Login-facebook
Around SBN: NFL Week One: Previews and Predictions for all 15 games

Uh oh... Ramon Diogenes Ortiz?

From Rotoworld.com ...

"The Twins are offering Ramon Ortiz a one-year deal worth more than $3 million, according to FOXSports.com's Ken Rosenthal.

Incredible. Ortiz finished with ERAs of 5.36 and 5.57 in the National League the last two years. With the way his stuff has declined, it's hard to imagine him lasting as even a fifth starter in the American League."

I have to say, if this is true and we pick up Ortiz ... well then I have no idea why we weren't in the running for John Thomson.  Ramon Ortiz is one of those guys on teams other than the Twins that I've always irrationally 'liked' despite their subpar performances (I live in MO, and watched his near no-hitter last season against the Cardinals at a bar while watching the Twins play on another TV) but he's just not capable of outpitching Garza/Baker/Perkins.  Or Ponson, most likely.  He's had ERA+ of 83 and 79 the past couple seasons.  In the National League.  Ramon Ortiz for $3 mil seems like a tremendous, unimaginative waste.  Thoughts?

0 recs  |  Comment 47 comments |

Story-email Email Printer Print

Comments

Display:

I hope
I hope ther eis no truth to this.  That would suck ass.
"Baseball is great because you can't take a knee or kill the clock. You have to put the ball over the plate and give the other guy his damn chance." C Stengel

by AdamOnFirst on Jan 19, 2007 2:51 AM EST reply actions  

Agreed...
I also for some odd reason have always liked Ortiz (saw a lot of him with the Angels out here) but yes, he is terrible.  Especially if it's guaranteed, this makes zero sense.  As a ST invite to compete with Ponson, et al., sure, but not as that.

And not only was he in the NL, but he was in a GREAT pitchers' park.  

Please TR, NO!!!!

I still would have MUCH rather had Jerome Williams than either of Ortiz/Ponson.  He at least has talent, is young, and has upside.

by djskilbr on Jan 19, 2007 3:53 AM EST reply actions  

It's official
Twins sign Ortiz for $3.1 million, contingent on his passing a physical.

http://www.startribune.com/509/story/947073.html

by dwintheiser on Jan 19, 2007 2:41 PM EST reply actions  

I need a drink
I had hoped against hope that this was some kind of joke. But no such luck. This is undoubtedly the worst signing since Batista. I didn't have a huge problem with Ponson because he doesn't get guaranteed money or a roster spot. But guaranteeing Ortiz $3 million and a roster spot is insane.

Perhaps ubelmann can put a better spin on this with FIP stats. I know the Nationals didn't exactly have good defense behind him. But he has consistently been above 5 in the NL in ERA, which is not good. I would rather have Garza, Baker, Perkins or Slowey starting than this guy. The thought of Silva, Ponson and Ortiz starting the year in the rotation makes my head spin.

Where's the Tylenol?

Joe Mauer for MVP (for real).

by cmathewson on Jan 19, 2007 4:18 PM EST reply actions  

Agreed...
the bars by me should be VERY, VERY afraid tomorrow.  As should the other patrons on the golf course I will be going to in the morning.

by djskilbr on Jan 19, 2007 4:44 PM EST up reply actions  

from RW again...
"The Twins are set to sign Ramon Ortiz to a one-year, $3.1 million contract, assuming that he passes a physical.

We're sure like to know who they wre bidding against. Despite being a flyball (pitcher) in a friendly environment, Ortiz was terrible after taking a $2.5 million contract from the Nationals last season, going 11-16 with a 5.57 ERA. Also, he was almost as bad with the Reds in 2005, finishing with a 5.36 ERA. We don't see how he's a better bet for 2007 than either Matt Garza or Scott Baker."

This kind of sucks.  No... it kind of sucks A LOT.  

Funny thing is, he's the kind of guy that I'd love rooting for if he comes up with a flukey 4.00 ERA season.  He just won't do that.  At all.

Guesses on Diogenes' (as I will now always refer to him) a) ERA and b) starts made with the Twins?

a) 5.70 and b) 24

by gophersw on Jan 19, 2007 4:52 PM EST up reply actions  

Starts
I'm actually quite afraid that you are wrong on the number of starts.  He's probably going to get more than that.  He's had 30+ 5 of the last 6 years.  The one year he didn't he still had 34 appearances (20 out of the pen).

The ERA's probably spot on, but do we really need to pay 3.1 for 180 or so innings of 5.7?

I just puked in my mouth.

by GACTwinFan on Jan 19, 2007 6:20 PM EST up reply actions  

Sorry...
Perhaps ubelmann can put a better spin on this with FIP stats.

No dice.  Ortiz's FIP and xFIP have both basically been close to his actual ERA the last two years.  PECOTA thinks Ortiz is going to be awful next year, too.

I just don't see the allure.  He's not very good at anything.  Other than strikeouts, I guess he's not awful at any particular facet of pitching, but he's not good at anything, and that adds up to a pretty bad pitcher.  

At least with someone like Silva, he's always kept his walks low, so I can hope for him to start getting some ground balls again.  With Ortiz, you have to hope for multiple things to go well, and that just ain't happenin'.

I'm beginning to wonder at what point Rick Anderson trots out to the mound, takes the ball, and pitches himself.  It's gotta be easier than trying to work with Ramon Ortiz.

by ubelmann on Jan 19, 2007 6:43 PM EST up reply actions  

Sometimes ERA...
...is misleading, but it's not in this case.  Ramon Ortiz isn't very good at any particular aspect of pitching, and I find it highly unlikely that Rick Anderson can "fix" him.

Picking up replacement level pitchers should cost the minimum, not $2.5M over the minimum.  I was much happier with the Ponson signing.

by ubelmann on Jan 19, 2007 6:38 PM EST reply actions  

Okay...
There is one upside to this that I can think of: it's only a one-year commitment.  Yes, it's a bad commitment, and Ortiz might not even last the year, but it's only one year.  It could out to be a season-wrecking decision, but it's not an organization-wrecking decision.

So there's that.

by ubelmann on Jan 19, 2007 6:51 PM EST reply actions  

Service time rears its ugly head...or should
I don't like the stockpiling of these yahoos either, but I will maintain that Matt Garza should not see the inside of the big league clubhouse until June.  His late season call up practically begs the club to keep him stashed at Rochester until he can stay the remainder of the season and not cross the 1 full year threshhold.

 The Twins fouled up royally last year by leaving Liriano to pitch bullpen mop up in April and it will eventually cost them a year's service time there.  I know service time isn't the only issue but with Gil Meche at 11 MM per, the Twins HAVE GOT to be cognizant of that reality.  You cannot allow pitchers of Garza and Liriano's caliber to "waste" service time finding their way at the "big league level."  Baker is a different story.  By now it's fish or cut bait with him--lower ceiling.  I'm betting on Johan, Silva, Ortiz, Bonser and Baker to start the year.  If Ponson has a good spring, Baker may be peddled or returned to Rochester.  

I think Ryan agrees with this sentiment.  Garza will start the year at AAA.  

What interested me most about the Ponson/Ortiz signings is that they should boost payroll over 70 MM, which I didn't think would happen.  So, Pohlad continues to keep up with inflation or at least close to it, even as teams like the Royals and Indians outspent inflation this year.  

by HrbekIsMyHero on Jan 19, 2007 8:16 PM EST reply actions  

The payroll...
...I believe the current payroll is $65,505,000 including all the arbitration players at the midpoint of the Twins and Player numbers.  That also included every other unsigned player at the minimum (which is likely where they will be with this year's minimum at $380,000).  It includes 25 players including Liriano, thus another player needs to be added.  Should it be Ortiz they will be at $68,605,000.  If any player such as Harvey, LeCroy, Ponson, etc. makes the squad...they are pushing very close to the $70,000,000 number.  

Am out in the desert so I don't have the exact salary numbers with me.  Will publish another detailed update when I get back around February 1.

by roger on Jan 19, 2007 8:44 PM EST up reply actions  

arbitration clocks
I'll add to this a bit.  If Garza, Perkins, and Slowey (just for example) were given the remaining 3 spots in the rotation, you'll have a huge problem in 5 to 6 years.  While delaying their arb clocks is a thought, I think TR would probably feel more comfortable adding 1 rookie to the rotation each year, making it much easier to fill a spot when they don't pan out, and a lot easier to manage your depth in the long run.

that's just me.  

by diehardtwinsfan on Jan 24, 2007 7:50 AM EST up reply actions  

Secondary concerns
All this crap about arbitration clocks and breaking in one prospect at a time is secondary to the real criterion here: Use the best, most talented players available to you for your team. To the extent that Silva, Ponson, and Ortiz are less likely to be successful than Garza, Perkins, Slowey, Durbin, or Baker, you shouldn't invest money in them and block the progress of more talented pitchers in the process.

When the flood of Twins pitching prospects from the 2004 and 2005 drafts hits spring training in 2009, it simply will not be an option to take just one or two. Swarzak, Duensing, Rainville, Sosa and company will be just too good to repeat AAA while inferior retreads pitch five innings and give up four runs every time out. If you try to put artificial constraints on the primary criterion, you'll end up with a better staff in Rochester than Minnesota.

That could well be the case this year through the first two months of the season, taking Santana out of the equation.

Joe Mauer for MVP (for real).

by cmathewson on Jan 24, 2007 10:21 AM EST up reply actions  

I think
We're gonna have a repeat of last year again, only instead of the defense and offense getting the bump by ditching the worthless "experienced" veterans and other guys getting hot, we'll have the young pitchers coming in and Santana taking off around mid-May like he does.  Won't be soo ridiculously dramatic, but it will be similar in a way I think.
"Baseball is great because you can't take a knee or kill the clock. You have to put the ball over the plate and give the other guy his damn chance." C Stengel

by AdamOnFirst on Jan 24, 2007 10:55 AM EST up reply actions  

well
That should just about conclude our offseason.  This signing also squashes the hope that the Twins could make a trade in spring training.  With Ortiz, Ponson, and Silva the Twins have to hope that these guys are capable of 5-6 innings giving up 3-4 runs a start to give the bullpen a chance to win the game.  Also it gives the Twins a chance to keep service time down.  But don't we have a boatload of pitching propects down on the farm anyway?  So we could develop a pitcher and send them packing for propsects as they get established and the new pitchers step forward and so forth and so on.

I have also noticed that TR has paid attention to offense depth a little more this offseason than in the past.  With Cirillo, Harvey/LeCroy Rondell White re-signing, keeping both Ford and Tyner(wait that doesn't help the offense).  I think our offense should be able to score in the neighborhood of 850 runs next year to help offset some of the sucktitude of our rotation.  Plus we still have 6 or 7 solid relievers who can keep us in the game if it isn't lost in the first 5 innings.

by doofus04 on Jan 19, 2007 9:19 PM EST reply actions  

Should, but probably won't
I think our offense should be able to score in the neighborhood of 850 runs next year to help offset some of the sucktitude of our rotation.

Given that the Twins had breakout seasons from both Morneau and Cuddyer and still barely broke 800 runs, I think a 50-run improvement is awfully optimistic. Given that the offense improved by over 100 runs from 2005 to 2006, the smart money would be on the club's offense regressing, not improving in 2007, especially if many of the same players are still playing.

by dwintheiser on Jan 22, 2007 2:09 PM EST up reply actions  

A statistician's viewpoint
I know what you're trying to say, but the often misunderstood concept of "regression to the mean" isn't applicable here, sorry.  

It applies to taking players at the top of the curve (say, take the Twins top three batters) and finding their expected 2007 production value relative to 2006.  If these players were all in their approximate prime and you guessed their expected production to be equal to 2006, you would be overestimating it, yes, though not by as much as you'd think.  

But Mauer and Morneau certainly weren't yet in their prime years (generally ages 27 and 28 for batters).  Cuddyer is, but he's probably benefiting pretty heavily from hitting between this guys.

Now, at the same time, regression to the mean applies also to the bottom part of the curve.  And there you have Kubel/White, Batista/Punto, and maybe Castro/Bartlett.  It seems pretty obvious that the bottom of the curve should outproduce last year.

Regression to the mean is not applicable in the way you try to apply it.  Applying it correctly should generally come up with the same overall mean as before, unless there is reason to believe improvement (positive or negative) is occurring, which should be the case (positive).  The Twins batters will not inherently get worse, and they are not too old.  The biggest risk would be Hunter, but he's shown he can produce in a contract year, so I think he'll be good again, numbers-wise.

To conclude, I'll bet a beer that the Twins score at least 25 more runs this year than last, relative to a 162 game season (apply the avg runs to any games not played).  Who wants in?

You can't dust for vomit.

by twinstalker on Jan 23, 2007 5:09 AM EST up reply actions  

Scoring
I think the Twins will score more runs too, and 25 sounds like as good a shot in the dark as anything. The only Twins hitters that I project for rgression are Castillo and Punto. The pthers will benefit from additional experience prior to their peak years, or, in the cases of Bartlett and Cuddyer, more regular playing time early in the year. And in the cases of Hunter, White and Kubel, better health overall. I like betting and everything, but I like beer too much to gamble against the odds on this one.

Still, I project a starting staff ERA of around 5, assuming Ortiz is in the rotation the whole year and Silva continues his struggles with consistency. That's pretty bad when your number 1 starter projects to a 2.75 ERA. If Gardy can use the bullpen effectively and minimize Silva's and Ortiz's damage when they get shelled early, the overall staff ERA could be around 4.

I don't know off the top of my head, but I recall that last year's staff finished with an ERA around 3.5. So the million dollar question is, will this line-up score a half a run more per game? The answer is, I doubt it. Until they make changes with Castillo and Punto, they will continue to struggle to get on base at the top of the order, which will put undue stress on Mauer, Cuddyer and Morneau. The rest of the line-up looks good.

But the team will be playing from behind an awful lot in 2007, and that puts a damper on the running game. All in all, I'd be surprised if they match last year's run differential, which translates out to a lower Pythagorean. A shot in the dark tells me 88 wins, unless they move quickly to fix the rotation and the top of the order.

Joe Mauer for MVP (for real).

by cmathewson on Jan 23, 2007 10:57 AM EST up reply actions  

Rotation
You've got too much doom and gloom on that rotation when you're realistic with the lineup.  Bonser will probably put up around  4.20 ERA, like last year, only over the entire seaason.  Our worriesome guys, like Silva, Ortiz, and Ponson, won't be allowed to coexist with terrible results, at least after the first couple months.  WE'll get  ahealthy dose of Garza, especially by midseason, and Perkins will figure into the mix as well.  Silva can't possibly really get any WORSE than last year, nor can the time we see Baker for, so only growth potential on those points.  On top of that, Slowey will probably see some action at some point down the road.

Last year, our total staff ERA was 3.95, but our rotation's ERA was 4.50.  I think a repeat of that number is quite likely considering we'll have a full year of Bonser, The same or better from the 250 innings Silva and Baker gave us last year at over 6 ERA's (especially Baker, I expect him to bounce back some), Vastly improved performance from Garza, positive help from the likes of Perkins and Slowey, and the deduction of Kyle Lohse's time.  I think that all should overall be enough to make up for Liriano's relatively small number of innings (under 100 starting).  Also, remember, while we lost Radke, he was awful for the first two months and finished with  only a 4.32 ERA, so despite his amazing courage down the stretch, from the perspective of average, he barely was on the helpful side anyway.

Basically, Silva and Baker or whoever else take their innings can't POSSIBLY be worse, right?  That's what I'm hoping...

"Baseball is great because you can't take a knee or kill the clock. You have to put the ball over the plate and give the other guy his damn chance." C Stengel

by AdamOnFirst on Jan 23, 2007 12:12 PM EST up reply actions  

I hope you're right
I think we agree that Silva, Ortiz and Ponson don't project to be even replacement level. The question, then, is how long are they allowed to suck? I assumed that the investment in them is an indication of their rlative leashes: Ponson might get a month, Ortiz two months, and Silva half a year or so.

In retrospect, that's not much worse than the bottom of the rotation last year. Baker was sent down in mid May in lieu of Liriano. Lohse was delat in June. Radke didn't come around until about that time. Bonsor had two ineffective stints with the big club before finally breaking through. Silva was bounced to the bullpen for parts of two months. In a sense, we can't have more chaos in the rotation than we did last year.

The big difference is, we have nothing like Liriano either on the big club or in Rochester to start the year. That alone gives us less margin for error, which makes this Ortiz move all the more baffling. Anyway, I would expect at least a .25 increase in team ERA with the absence of Liriano. It might have been only a couple of ticks if Ryan had been able to trust Garza and others. How that translates out into wins and losses is more than I can give at this time.

Joe Mauer for MVP (for real).

by cmathewson on Jan 23, 2007 2:05 PM EST up reply actions  

Isn't
Isn't it amazing that with all that bouncing arouns, Silva still managed 31 starts?  I think it is.
"Baseball is great because you can't take a knee or kill the clock. You have to put the ball over the plate and give the other guy his damn chance." C Stengel

by AdamOnFirst on Jan 23, 2007 2:27 PM EST up reply actions  

Obviously I disagree
Regression to the mean is not applicable in the way you try to apply it.

Well, I'll admit I'm not a statistician, but I can read the words of folks who are. Specifically Tom Tango's, who has a projection system he calls 'Marcel' that simply regresses a player back toward the major league average for all players. No park or league adjustments, no adjustments for age, just simple statistical regression to the mean in its purest form. (If this wording looks familiar to you, I 'borrowed' it from Dave Studeman's essay on the Hardball Times about regression: http://www.hardballtimes.com/main/article/but-i-regress)

The most sophisticated projection system out there, PECOTA, which adjusts by similar players and takes every bell and whistle one can imagine into account, had an r-value of .736 for batters with at least 500 AB in 2006 (according to Sean Smith: http://lanaheimangelfan.blogspot.com/2006/12/pecota.html). Marcel had an r-value for the same players of .664. Simple regression to the mean is really that powerful.

Marcel gets even closer to PECOTA when considering pitcher projections: PECOTA's r-value for pitchers with at least 100 innings in 2006 was .451; Marcel came in at .442 (http://lanaheimangelfan.blogspot.com/2006/12/pitcher-projections.html).

Also keep in mind that we're not talking about individual players in a vacuum here, but how players will perform in conglomerate to produce an aggregate team result. Aggregating individal results into a larger conglomerate result is exactly where you're going to see these statistical effects, right? One player might exceed his projection, but another will fall below his, so overall the result should come close to the expected mean.

With all that said, I downloaded Tango's 2007 Marcel projections (http://www.insidethebook.com/ee/index.php/site/comments/marcel_player_forecasts_2007/), split out all the totals for players on the 2007 Twins roster according to ESPN.com (http://sports.espn.go.com/mlb/teams/roster?team=min), and ran the numbers through the 2002 Runs Created formula (http://en.wikipedia.org/wiki/Runs_created). I actually got an astonishing total of 871 runs created, which surprised me no end. Looking closer, it's not quite so surprising - some players did so much better than the mean in 2006 that even regressing them to the league mean keeps them very high (specifically Morneau, who projects to .292/867, Cuddyer, who projects to .276/811, and Punto, who projects to .270/699).

Of course, your own argument that players should be regressed to their own personal means rather than to the league mean argues that this total should be lower, if you presume that Cuddyer's, Morneau's, and Punto's performances don't represent new means, but simply performance significantly above their 'real' talent level, as I do.

One last point:

Now, at the same time, regression to the mean applies also to the bottom part of the curve.  And there you have Kubel/White, Batista/Punto, and maybe Castro/Bartlett.  It seems pretty obvious that the bottom of the curve should outproduce last year.

This is an extremely generous interpretation of 'bottom of the curve'; Batista and Castro accounted for only about a third to a quarter of the team's PAs at their respective positions (Batista had 193 PAs before being released, while Castro had 162 before being traded), while Punto's and Bartlett's numbers pushed the conglomerate up to if not over the league mean, which means their individual totals were significantly above the mean. Punto in particular is a tremendous candidate for a powerful regression effect; not only was his .290 batting average significantly above the league mean (and it's not like he adds much offense over and above his batting average, so it it drops, his value drops), but his overall numbers in 2006 (.290/725) look like an outright aberration compared to his career numbers to that point (.238/623). I'd compare Punto's 2006 season to Zoilo Versailles's 1965 season, except that wouldn't be fair to Versailles, whose performance in 1965 was actually much closer to his career norms to that point than Punto's was last year (http://www.baseball-reference.com/v/versazo01.shtml).

I'll stick by my assertion that the 2007 Twins will score fewer runs than the 2006 Twins, though I also think that the 2007 Twins will have much better starting pitching than folks think right now. Radke, Bonser, Guerrier, and Baker all had FIP ERAs of 4.6-5, and I don't see any reason why either Ponson or Ortiz can't match that given their career numbers. In addition, I see Silva's 2006 season as just as much a fluke as Punto's; he'll be better than he was last year. Granted, these are guesses, but I think they're reasonable guesses.

Unfortunately, fewer runs scored plus about the same or slightly more runs allowed (Nathan has to prove himself mortal someday, right?) translates to a frustrating just-out-of-the-running season in the ridiculously competitive AL Central.

by dwintheiser on Jan 23, 2007 12:31 PM EST up reply actions  

LOL
Even if you don't believe me, which you should (it's like believing a lawyer when he tells you the law), you still have to apply your "regression" to both ends of the curve.

r-value?  Are you talking about correlation?  That number has to be between -1 and +1.  It seems you might be interpreting things a little differently than intended by the stat.  My guess is that you're looking at it as a linear coefficient, which can actually be greater than 1.

Very briefly, regression to the mean is this concept: If you choose a unit based on where it falls on the curve (let's say high in this example), and you predict the next value for that unit to be the same next time round, on average you are going to overestimate that value.  That's regression to the mean.  On average, because of the way you chose the unit, you will overestimate it (on average).

What regression to the mean is not is choosing your unit based on nothing whatsoever to do with its relative value, and then because its value happens to be high, expecting it (on average) to decline the next time round.  And that's what you're doing.

It all comes down to why you chose the unit you're looking at, a concept that most non-stats people have a really hard time with.  It makes no sense to most that there should be a difference in the prediction.  After all, it's the same unit no matter how you picked it, right?  Wrong.  And I'll save the why for now.  But if you accepted it, then it would start to become intuitive to you, I think.

All these analyses you're talking about?  It seems reasonable to conclude that the "top performers" were chosen because they were top performers, not coincidentally for any other reason.  And yes, there will be a regression there.

And by the way, my "bottom" of the curve statement with Batista/Punto and Castro/Bartlett was looking at positions.  Batista/Punto was a pretty putrid position...definitely low end of the curve for the Twins.

And finally do the trend analysis on guys like Morneau and Mauer and even Cuddyer (apologies, I'm making this subjective).  I don't really see Morneau hitting less than 40 HRs, and I do see Mauer starting to really break out the power, regardless of whether he bats .350 again.  Cuddyer I see improving (inherently) and then benefitting from being between these guys who are getting better.

Can one injury to a Morneau or Mauer shatter this thing...probably.  And if the Twins don't have last year's production, it's going to be an injury to one of the three (or four, with Hunter) that's going to be the cause.

You can't dust for vomit.

by twinstalker on Jan 25, 2007 3:58 AM EST up reply actions  

Mean
The biggest hole in the regression to the mean ship is the M & M boys are still so young, especially Mauer.
"Baseball is great because you can't take a knee or kill the clock. You have to put the ball over the plate and give the other guy his damn chance." C Stengel

by AdamOnFirst on Jan 25, 2007 5:16 AM EST up reply actions  

Right
Regression to the mean might not be the best analogue to what happens in baseball. As twinstalker demonstrated, it's a concept in statistics about units of measurment--the result of a simple fact that the units we choose are somewhat arbitrary and likely to not fit the case perfectly. There are similarities between this phenomenon in statistics and a similar phenomenon in baseball. But it's not entirely the same concept.

In baseball, you set an arbitrary mean at between 27 and 29 years old (depending on your theory) and you estimate the likelyhood that players are on one side of the bell curve or another. If a player is on the young side of 28, you expect some improvement as he learns the league before his body starts wearing out. If he's on the old side of the curve, you expect him to regress as he just can't get his body to do what it could in its prime. The further away from 28 a player is, the steeper the curve.

Players who had career years but are still on the young side of the curve are still expected to make improvements as they learn opposing pitchers and new ways of reading opponents. On the Twins, that includes Mauer, Morneau, Cuddyer, Bartlett, and Kubel.

Players who are on the old side of the curve are expected to regress. That includes Hunter, White, Castillo and Punto. In some cases, you make exceptions: Hunter's in a contract year and coming off a year in which he was only healthy for half of it. White also had a half a healthy season. Castillo wasn't healthy, but his lack of good health is a symptom of age --more chronic injuries than the one-time events of Hunter and White.

Taken for all in all, Punto and Castillo are the only ones who should experience serious regression. That will affect the top of the order, until Gardy can make adjustments, say, by moving Bartlett into leadoff and Cirillo into the second spot. What that translates into as far as run differnetial is more than I can do right now. But I will say that the top of the order is a key component of scoring, especially early in the game. Combined with poor starting pitching to start the year, it could mean tough sledding in April and May.

Joe Mauer for MVP (for real).

by cmathewson on Jan 25, 2007 10:04 AM EST up reply actions  

Confusion
In baseball, you set an arbitrary mean at between 27 and 29 years old (depending on your theory) and you estimate the likelyhood that players are on one side of the bell curve or another.

Cmath, you're confusing regression to the mean with the expected development curve.

The expected development curve says that a player enters the league young, improves until a peak season at about age 27, then declines steadily until retirement. Not every player follows this curve, but enough do so that it's the expectation.

Regression to the mean is different: a team that finishes above .500 is expected to 'regress' back toward .500 in the following season, while a team that finishes below .500 is expected to 'regress' upward toward .500. A player who hits for a higher-than-mean batting average is expected to hit for a lower, closer-to-mean average in the following season, while a player who hits below the mean is expected to hit for a higher, closer-to-mean average in the following season. Again, not every team and player follows this expectation, but enough do so that it's the expectation.

I'd love to see a study that investigates whether young players are more or less resistant to regression-to-the-mean than older players or vice versa, but as far as I know, nobody's done such a survey.

by dwintheiser on Jan 25, 2007 11:55 AM EST up reply actions  

Not that different conceptually...
For any given team, the mean expectation is .500--if we didn't know anything about that team, we would expect them to be a .500 team.  So when you do regression to the mean, you regress to .500.

Now say that we have some more information about that team, like say, it's not actually a MLB team, but a high school team that got a weird exemption to play one year while the Yankees sat one season out.  Surely, as a high school team, we expect them to do worse than .500.  Thus, we would regress their performance to some value less than .500.

With players, if we don't know anything about them, we can regress their performance to league average.  However, we have extra information about these players.  For instance, we know that Pedro Martinez is a pitcher.  We also know that in the NL, pitchers are a much worse group of hitters than position players.  Thus, it makes more sense to regress Pedro to the average hitting performance of pitchers than the overall league average.

In fact, we have quite a lot of extra information about these players.  If we're really good at interpreting the information then we can pick increasingly more intelligent means with which to regress each player.

I don't know if this is exactly what cmathewson is saying, but my main point is just that choosing the mean is an awfully important step in regression to the mean.  Considering what we know about each player, we can clearly do better than regressing each player to the same mean.

by ubelmann on Jan 25, 2007 1:41 PM EST up reply actions  

L...O...L...?
Even if you don't believe me, which you should (it's like believing a lawyer when he tells you the law), you still have to apply your "regression" to both ends of the curve.

I'd be more inclined to believe you, regardless of your professional status, if you actually demonstrated that you read and understood what I was saying. For starters...

r-value?  Are you talking about correlation?  That number has to be between -1 and +1.

Notice that both numbers noted in the essay are between -1 and +1.

Also, keep in mind that the r-value is for different prediction systems; what's being correlated is the prediction for players' numbers and those players' actual numbers. If you're doing a correlation study on your prediction system, and you end up with a correlation coefficient near zero, then your prediction system sucks. If you end up with a high negative correlation coefficient, you've got a backwards prediction system; your system predicts as poor those players likely to do well and vice-versa. Given those points, it's not surprising that a number of well-formed prediction algorithms will have high positive correlation coefficients; that's the whole point of having a prediction algorithm, that the predictions correlate with actual results.

What regression to the mean is not is choosing your unit based on nothing whatsoever to do with its relative value, and then because its value happens to be high, expecting it (on average) to decline the next time round.  And that's what you're doing.

Well, for starters, I'm not doing the regression - I'm just reporting a secondary result based on the regression performed by Tango (or Marcel, if you prefer).

Secondly, there is a relative value associated with the regression - the values of the individual players are compared relative to the mean performance of all major league players in 2006. I'd have thought that would be obvious, given, y'know, that I said that Marcel "regresses a player back toward the major league average for all players." You can also look at Tango's methodology for Marcel (http://www.tangotiger.net/archives/stud0346.shtml) and see exactly how he came up with the system he did. (Note: I owe Tango an apology, as he makes clear in that methodology that he doesn't want his name associated with those predictions. So from here on out they're the Marcel predictions, period.)

And by the way, my "bottom" of the curve statement with Batista/Punto and Castro/Bartlett was looking at positions.  Batista/Punto was a pretty putrid position...definitely low end of the curve for the Twins.

And here's another problem.

You combine Punto and Batista at third base and call the position 'putrid' and expect that it'll improve in 2007. But that performance was composed of two factors; Batista's, which was below mean, and Punto's which was significantly above mean in some ways, specifically batting average. It's like that classic problem about the three men who check into a hotel and end up missing a dollar - you're making a fallacious move in comparison. You've eliminated Batista's contribution, which not coincidentally had the best potential for improvement, at least based on regression to the mean. What you're left with is Punto, who no matter how you slice it is setting up for a fall next year. Punto may well end up being better in a full season than he and Batista were in their shared season last year, but it won't be regression to the mean that's responsible, since the part that would be regressing upward no longer exists in the sample. If Punto is better than Punto/Batista last year, it'll be because a full season of Punto is better than two-thirds of a season of Punto and a third of a season of Batista; regression will have nothing to do with it.

And finally do the trend analysis on guys like Morneau and Mauer and even Cuddyer (apologies, I'm making this subjective).  I don't really see Morneau hitting less than 40 HRs, and I do see Mauer starting to really break out the power, regardless of whether he bats .350 again.

Sorry, this kind of 'trend analysis' reminds me of those charts of Internet usage growth that came out in the late 1990s, and that the Onion used to predict that there would be twice as many Internet users as people alive on the planet by 2012. There are so many ways in which Morneau could fail to even reach 30 HRs (regression to the mean, injury, inconsistency, increased scrutiny from scouts and pitchers, etc.) that to predict he'll break 40 HR is, in my eyes, pretty much just wishful thinking. It'll be great if it happens (and keep in mind I didn't think he'd break 30 last year, either, so my predictions aren't anywhere near airtight), but expecting it is only setting yourself up for heartbreak, IMO.

As for Mauer developing power, I have more faith in the comparitive studies of players like Mauer that suggest he'll never consistently hit as many as 20 HR in a season. Again, I could easily be wrong, but that's where the evidence appears to lie; the folks who say Mauer is going to develop home-run power seem to be relying on an anecdotal, 'players develop power as they get older, so he will to' argument which might not apply in Mauer's case.

You're certainly welcome to your opinion, but please at least make an attempt to understand what I'm saying before you simply dismiss me as ignorant or uninformed.

by dwintheiser on Jan 25, 2007 11:46 AM EST up reply actions  

Whether
Whether or not Mauer develops home run power in droves, he's only 23, and the odds are overwhelmingly in the favor of him continuing to improve for several more years, and Mauer might even be the type of player who improves beyond the time his body grows because of his talent and drive.

I think marcel is extremely poorly conceived.  As a baseline, the players should be regressed to average for their POSITION, which would make a big difference.  Also, they should be regressed to the mean relative to the positional average for their age, which would change from year to year.  So say, if Morbeau was, say, 130% (of whatever you're measureing) of the average mark for 25 year old first baseman, then regressing to the mean would put him at, say, 120% of the average for first baseman his age.  But that age would be 26 now, and hte avrage would be higher, so the regression wouldn't be as great, and it could even be higher based on the baselines.

Marcel is currently far to rough.  It may generally average out okay, but you simply cannot talk about individual players with it, and I have difficulty endorsing a system that averages out it's inaccuraies mostly by chance.  For a team like the Marlins, a young team bound to improve a lot with age, it would horribly underestimate, but a team like the Yankees, who it would regress to the mean but not aaccount for their aging it would overestimate.  It's logic is only true when you look at the biggest possible pictures, and htat doesn't cut it at all with me.

"Baseball is great because you can't take a knee or kill the clock. You have to put the ball over the plate and give the other guy his damn chance." C Stengel

by AdamOnFirst on Jan 25, 2007 2:33 PM EST up reply actions  

Stuff
Does anyone know what his stuff is like? What he throws and such?  Did he loose some velocity after leaving the Angels?
break the 30 HR curse!... okay the 40 HR curse!!!!

by caseintheface on Jan 20, 2007 1:17 AM EST reply actions  

From
what I read, he has a 4-seam and a 2-seam, a slider and a change.  This was in a scouting report from two years ago, though...I didn't see anything from '06.

by Jesse on Jan 20, 2007 2:46 AM EST up reply actions  

Roger's payroll numbers look right
I get a shade over 69 MM with R. Ortiz so we're 400 K apart...I put Cirillo at the midpoint of his incentives...in other words...it's very close.  So, I would also presume we are done.  

From 63.5 to somewhere between 69 and 70 MM, that's approximately an 8 - 10 % bump in payroll, which is consistent with league average increase of around 9% in the last several years, so I'll shut up about Pohlad for this year.  If they keep raising at the rate of inflation (ceteris parabis and all) we'd be at around 75-76 MM for 08, then 82-83 MM in 09, and if we get that 20 MM or so bump from the new ballpark plus "normal" inflation in 2010, we may live to see a 110 MM payroll to usher in the new era.  

So...Santana's extension at 4/80 isn't totally out of the question, is it?  Sign Mauer at 4/34, Morneau at 4/38 and I'll call it an off season.

Question for you guys?  Why didn't Tyner file for arbitration?  Cot's had his service time as something around 2.140 before the Twins called him up last year, meaning he should be over 3 full years ST and thus eligible.  Yet he didn't file and I haven't seen that there's a contract out there.  Is he on the 40 man?  If so, apparently Cot's screwed up or something, because he should be arb eligible, shouldn't he?  

by HrbekIsMyHero on Jan 20, 2007 10:15 AM EST reply actions  

Tyner and stuff
I think the service time issue changes when you've been released and resigned. I'm not sure about that, but I put his service time just under 3 years. I think you only qualify for Super 2 status if it's continuous without passing through waivers.

As for Pohlad's budgets, I wouldn't presume that he'll keep up with inflation. The budget is set according to the previous year's revenue and sometimes adjusted up slightly for attendence projections. Before the offseason, I estimated that the Twins payroll would be around $70 million going into 2007 based on last year's attendance plus the extra revenue from radio. The fact that it's keeping up with inflation is a coincidence.

If the attendence hits 3 million in 2007, which I wouldn't be surprised by if the Twins get off to a better start, the 08 budget should outpace inflation. If the Twins get off to a slow start again (as the state of the rotation suggests), they could be flat on attendence, which would leave their budget flat for 08.

Joe Mauer for MVP (for real).

by cmathewson on Jan 20, 2007 10:49 AM EST up reply actions  

M&M Deals just went up
Did you see the Chase Utley deal today?  7 yrs / $85M - something like $4.5M, $7.5M, $11M, $15M, $15M, $15M, $15M.  His numbers and service time are very comparable to Mauer and Morneau.   This is what M&M's will want for a long term deal.  Very risky for the Twins to do that.  We will probably go through arbitration with them for a few years or settle on one year contracts.

The way I see it we have about a two year window until TR will have to make some major moves - moving some combo of santana / nathan / morneau / cuddy.  

Time to enjoy it while we can.  Welcome the the economics of baseball even with a new stadium.

by bobio on Jan 21, 2007 6:47 PM EST up reply actions  

Ya...
not too happy about that contract.  To me, Utley and Mauer compare favorably.  Mauer is the best C in the game at a young age, and Utley is the best 2b in the game at a young age.  Morneau you could say is a little different because he is a 1b, and while good, there are a number of good young 1b, though he is one of the better ones and it is a more highly paid position.

Will be interesting indeed, but I certainly didn't like that news either.

by djskilbr on Jan 21, 2007 8:28 PM EST up reply actions  

One
The one thing with Utley is he has almost the same numbers as Morneau , but a second baseman, and he has about 50% more experiance in the league as does Mauer.  I think Utley has already been through one year of arbitration also.

food for though.

"Baseball is great because you can't take a knee or kill the clock. You have to put the ball over the plate and give the other guy his damn chance." C Stengel

by AdamOnFirst on Jan 21, 2007 9:16 PM EST up reply actions  

I didn't use...
...any incentives.  Therefore, my numbers are probably at yours or even a touch higher.  Thus, we all should be amazed if Mr. Ryan does anything more this offseason.

by roger on Jan 23, 2007 9:37 PM EST up reply actions  

Tyner and more stuff....
Yeah, ubelmann, I would imagine something was affected in Tyner's service time by his release but Cot's (which is still one year "outdated" as we type today) showed his ST at 2.123 before the Twins added him to the MLB roster in July?  He had to have been on the MLB roster for at least August and September (60+ days and then some) so I still get him up over 3 full years (isn't 162 the magic number for a "full year?").  What's your calculation that leaves him just short of 3 years?  I can see why he wouldn't be a Super Two, if that's correct, and you probably know something I do not, but I still don't see why he's not a 3.  Laughs.  Teach the professor something here, please.

Did you see Alex Romero was waived and claimed by the D'Backs, apparently to add Ramon to the 40 man and keep Chris Heintz.  Rotoworld's having a fit about it being an idiotic move.  I can't imagine where Romero would have been even a 4th OF, but why Heintz?  Is this why LeCroy was selected for Rochester...he certainly isn't going to be a backup C for the big club under any injury scenario, but maybe he starts for the Wings if Heintz has to come up for injury?  

by HrbekIsMyHero on Jan 20, 2007 10:58 AM EST reply actions  

Romero
This just adds a layer of maggots to the rotten fish Ryan served us. I had Romero making the transition to AAA and putting up good numbers this year, good enough for a September call-up. He's the only outfielder in the upper minors with decent power and good plate disciplne, meaning he's the only guy I saw as a potential replacement for Tyner ro an injury replacement for Kubel. Excuse me while I go throw up.
Joe Mauer for MVP (for real).

by cmathewson on Jan 20, 2007 11:07 AM EST reply actions  

Ouch!!
I likewise feel terrible with Romero gone.  Question I have is why?  Why not Simonitsch?  Have the Twins gotten off the Romero bandwagon?

He started last year at Rochester and hit terrible.  After only a few outings in AAA he went back to AA where he hit well but not quite as well as in 2005.  He returned to Rochester at mid-season when Kubel came up.  He didn't play regularly and hit better than April but not real well.  Then he was one of their better hitters during the playoff run with several key rbi's.  He tore up the VWL until the playoffs when he was in a big slump.  With the power the Twins were looking for not developing, maybe they are no longer as high on this kid as we are.

Still would rather have seen them waive Simonitsch...or maybe even Durbin if they don't think he will make the 25-man roster out of spring training.

by roger on Jan 20, 2007 12:01 PM EST up reply actions  

Did I hear from you...
...at some point that the Twins thought Romero had attitude problems or something?  They are really big on "makeup" or whatever you want to call it, so that's probably the motivation here.

I agree that this hurts, though.  Our depth in corner bats above A-ball is awful.  We're really going to be stuck with Tyner or Ford in an OF corner if Cuddyer or White/Kubel goes down.  I like those guys and all, but they're going to be hurting the team if they're in an OF corner on an everyday basis.

by ubelmann on Jan 20, 2007 2:37 PM EST up reply actions  

Deeds and Rabe
I guess it's Deeds and Rabe for depth at the corners. Both profile as roster fillers rather than prospects, though Deeds could be a decent bench bat at some point. Guzman might be a Romero replacement at AA, but he's really inexperienced and small. I don't see any outfielders in this system who are really prospects above low A ball. Perhaps Romero was just a fringe prospect, but he at least has shown flashes of greatness in his minor league career.

I would much rather Ryan signed Nixon with the $3 million if we were going to lose Romero. A White/Kubel/Nixon rotation through left and DH would be fine by me.

Here's hoping Andy figures out what Ortiz was doing wrong with Washington and can take his ERA down a point. That's asking a lot because the move to the AL is worth a half a point of ERA at least. Maybe get him to throw his two seemer more and keep the ball down. He wouldn't be bad if he didn't give up so many fly balls.

Joe Mauer for MVP (for real).

by cmathewson on Jan 20, 2007 3:00 PM EST up reply actions  

Span
Span???  I thought he was a prospect.

by doofus04 on Jan 21, 2007 10:40 PM EST reply actions  

Span = fringe at best
My opinions are somewhat controversial, but widely held. Sickels, for example, does not even list Span in the top 30 Twins prospects for the reasons I state.

Span has some tools, but he has yet to put it all together into a package that screams prospect. He needs to be a Kenny Lofton or Juan Pierre type player if he is to make the jump to the majors. He has no power and his plate dicipline is somewhat suspect. So he needs to be a stellar defensive center fielder who gets on base and can steal a base when needed. If he projects as a high OBP guy with good base stealing skills and stellar defense, I'll rate him a prosepct. But he's never been a plus defender. He's usually caught stealing as often as he is successful. And his ave/OBP ratio is marginal for a top-of-the-order player.

We'll see if he can improve this year. But he's been at it for five years and he's only made marginal strides. If he does not put it together this year, my take is he'll be passed up in the system. I rate Roberts higher. The Twins also have Oeltjen, who at least has shown flashes of greatness. But none of these guys projects as a corner outfielder, where you need at least gap power.

Deeds has good power and is a fringe  corner outfield prospect, but his strikeouts are worrisome. I project his ceiling to be Jacque Jones, which wouldn't be bad at all in a platoon-type role. But he's at least a year away, and rated behind Romero at every level where they played together. And he's two years older than Romero.

The closest guy right now is probably Danny Santiesteban, but he's had an up-and-down career hampered by injuries, as brief as it is. I'd like to see him put up .300/.380/.500 numbers at Beloit this year, similar to what he did in the GCL. If he does that, he's my top outfield prospect in the system because of his amazing defensive skills. Still, we're talking about Beloit here. Best case is he skips a level and makes it up by 2009.

Then there's Joe Benson and Chris Parmalee. I rate them both as prospects, but they will probably start the year in extended spring and go to Elizabethton this year with an outside chance at Beloit. They might be in postion to win a spot in four or five years.

The rest of the pickings consist of guys like Jeremy Pickrel and Deacon Burns, who seem to struggle more the higher they get in the system. Romero was the only guy since Lew Ford who had decent doubles power and good plate dicipline at AA or above.

Joe Mauer for MVP (for real).

by cmathewson on Jan 22, 2007 8:50 AM EST up reply actions  

Comments For This Post Are Closed


User Tools

TT is an SB Nation blog of, by and for the fans. We strive to be the best Minnesota Twins blog by providing quality content and analysis, as well as daily news and notes on the team. We hope you'll make Twinkie Town your home for all things Twins!
Start posting about the Twins »

Join SB Nation and dive into communities focused on all your favorite teams.

Connect_with_facebook

FanPosts

Community blog posts and discussion.

Recommended FanPosts

Small
Minor League Report.....September 4, 2010
Summer_09_wedding_and_frank_084_small
Yet another overanalysis of our bullpen (extra innings rant)
Small
Minor League Report...August 28, 2010

Recent FanPosts

Kjhgfdfghjg_small
A Target Field Experience and a Warning to Twins Fans (with pictures)
Small
Matt Fox and Lefty Gomez in the Same Sentence?
Untitled_small
Pavstache T-Shirt: Going... going...
Smallseal_small
The Twins' best one year wonders - hitters
Index_small
white sux schedule
Tiger_woods_ii_small
Outfield Defense and How to Fix the Problem in 2011
Small
The Revere callup becomes clearer
The_jet_small
Cuddyer Twins MVP?
Royconrad_fullthrottle_small
25 for 25: Twins

+ New FanPost All FanPosts >

Twinkie Town On Twitter

SPONSORS

SBNation.com Recent Stories

SEATTLE - JULY 10:  Starting pitcher Felix Hernandez #34 of the Seattle Mariners celebrates after defeating the New York Yankees 4-1 at Safeco Field on July 10 2010 in Seattle Washington. (Photo by Otto Greule Jr/Getty Images)

MLB Power Rankings: On The Challenge Of Identifying A League's Best Pitcher

ANAHEIM CA - SEPTEMBER 08:  Jeff Mathis #5 of the Los Angeles Angels of Anaheim is mobbed by teammates after hitting a  walk off sacrifice fly to score Torri Hunter form third base against the Cleveland Indians in the 16th inning on September 8 2010 at Angel Stadium in Anaheim California.   The Angels won 4-3 in 16 innings.  (Photo by Stephen Dunn/Getty Images)

Mathis' 16th-Inning Sac Fly Lifts Angels Over Indians 4-3

Philadelphia Phillies' Jimmy Rollins, left, slides into home to score past the tag of Florida Marlins catcher Brad Davis on a single by Carlos Ruiz in the third inning of a baseball game, Wednesday, Sept. 8, 2010, in Philadelphia. (AP Photo/Matt Slocum) +2 updates

Phils Top Marlins 10-6, Jimmy Rollins Leaves Game With 'Hamstring Tightness'

More from SBNation.com >


Editor-In-Chief

Twinkietown_small Jesse

Senior Writer

Small Bobomojo

Hrbek_small Jon Marthaler

The_jet_small cmathewson

Gladdentwins_small Adam Peterson

Hosken_powell_autograph_small RandBall's Stu

Special Contributor

Untitled_small Trevour

Small Karlee Kanz

Moderators

Chairmanmauer_small fischean