Oxtung

"Regression to the mean" is such a large part of not only baseball but of the world as a whole that it is not surprising that his model is dominated by it. Not everyday can be a good day and not everyday is going to be terrible. In fact most are going to be well, average (however one defines that for themselves). I don't really hear anyone arguing that his model is invalid because he uses regression to the mean but rather people seem to think that he is a little heavy handed with it. That's fine. Is that an accurate assessment with people's problem with his model? If so... I'd like to talk a little bit about the opposite of "regression to the mean". What many of you seem to be saying is that his model, and I'd argue most statistical models, doesn't do a good job of predicting natural variability. That is it doesn't predict enough "spikes" in the way a player plays, whether that is good or bad. It seems to be rather easy to understand some of the triggers for regression to the mean. Getting older. Getting hurt. Having surgery. A huge spike in performance, either good or bad. What about the triggers for natural variability? In order to incorporate that variability into statistical models we must know what the impetus is. Here is another way of asking the question: why was Mauer so much better in 2009 than any year before or since? Then ask yourself, does that data exist publicly? I think that is at the heart of the problem in this model, and really all baseball prognostication. This is a rather esoteric ramble but I thought it worth reflecting on. Enjoy, or not, as one pleases.

I would guess that most scouts saw Rosario on a semi-regular basis, not just because of Eddie himself but because on his team was also Sano as well as any top prospects they were playing against. For instance when they were playing the Cubs A+ affiliate on the field would have been Sano, Rosario, Jorge Soler and Javier Baez. My guess is that would be a very well attended game by scouts.

Considering that their predictions are ultimately derived from an underlying equation does it surprise you that it looks formulaic? Many people, and I'm not just talking to you here Pseudo, seem to be having problems with these predictions. IMO you are all looking at these too literally. Predictions like these aren't saying, "Hey this is exactly what is going to happen!" Rather they are saying, "Given the data we input, this has the highest probability of occurring." However, like any data set when your n=1, as is the case with Joe Mauer, their confidence level is going to be pretty darn small and their error bars are going to be very large. Now, when you apply statistical modeling to a team, and your accounting for not just 1 player but 25, now you have something that can be given a little bit more credibility.

In the last five years only 11 of the thousands of players in baseball have had an OBP of >.420. Mauer's K% has been rising steadily for 4 years straight and even with an absurd BABIP of .383 last season he was unable to even approach .420. I'm curious what you think will be different this year that allows Joe to achieve such a difficult task?

Yup, but all a projection system is going to see is that absurdly high BABIP and determine that he is ripe for regression. Frankly after looking at it a little closer I think he is going to regress as well. Will he OPS .750? That's probably a bit severe but I also don't think he'll approach .900 either. His K% is very concerning at first glance and he isn't going to have a .383 BABIP this season again to help obscure his increasing contact problems (in comparison to his previous seasons). The good thing though is that his BB% remains very strong. It's also possible that he hits more home runs since he's not getting beat up as much.

IMO it really comes down to what level of play are the Twins willing to accept at 3B. There is a spectrum ranging from elite to down right terrible. If the Twins believe they must have an above average defender at the hot corner then Sano is going to be moved. If they decide it's ok to be below average as long as his bat carries it's weight well then he might very well stick there. A good comp seems to be Cabrera, not that I think Sano is likely to hit like Cabrera or even field like Cabrera, but will his stick be enough to accept some defensive short comings. Unfortunately for us this won't be resolved until we see him in the majors in the next few years.

AT the end of the 2013 season Sickels ranked Rosario 65th. He stated he likes Rosario as a hitter and that "he has figured out second base." The only 2B that is ranked higher is Kolten Wong at 47. It will be interesting to see if he thinks the same for his pre-season 2014 list.

No. His first 2 weeks at NB aren't included in that .254 and 29% K rate. I used only games 14-67 at New Britain. Those are serious red flags. Even at Fort Meyers he had a 25% K rate which is very high. The reason FM looks so much better is that his BABIP was sooooo much higher it obscured the strikeout problem.

re:Mauer's projection We don't know what projection system he is using to derive his predictions. Some of them emphasize regression to the mean very significantly. Beyond that there are some very good reasons that he is projecting Joe to have a down year. First, as others have said, Joe is being compared to other 31 year olds who are entering their decline phase. Second, Joe's injury plagued year when he OPS'd .729 is being included in the data that drove the prediction, that is surely going to cause a regression to his statistics. Third, Joe had a high BABIP, it was .383 last season (.300 is about average) and a regression to the mean looks to be in order causing all of his stats to suffer. Fourth, Joe's K% has been steadily increasing over the last 3 years, from 11% to 18%. It really isn't surprising that the math predicts Mauer to significantly regress considering the inputs.

Even if you exclude his first 2 weeks at AA he still only hit .254 and struckout 29% of the time. Neither of those are good. Nobody questions if he'll hit for power. But will he hit .240 or .280? That's the question, well along with do you think he'll play 3B or 1B.

Two interesting non-Twins placements: Chris Owings at 28 and Lucas Giolito at 13. Giolito has always had the "stuff" but he has hardly pitched since being drafted and is coming back from TJ surgery. Bundy seems to be a comp except that Bundy made it all the way to MLB before he had TJ surgery. Chris Owings is a bit like Pinto. He was never considered anything special until he hit AA and AAA where he exploded. Parks has him at 28. MLB.com has him at 77. That seems like quite the difference. Call me a Julio Urias Fanboi anytime you want. Man does he look like a stud in the making. Eddie Rosario is the 3rd 2B listed. Apparently his struggles in AA didn't effect his placement much for Parks. I wonder what it would take to trade for Luis Sardinas of the Rangers. He is a very intriguing player. Parks must think that Manaea's injury troubles really affected his play last spring. I think most prognosticators are in wait and see mode. I really like the approach that the Royals took in the 2013 draft. If Manaea can come back....what a coup!

I actually think he's undervaluing Lindor. Lindor is arguably is the best defensive player in the minors regardless of position. He drew more walks than strikeouts last year and stole 25 bases in just 400 at bats. Think about this for a minute. Andrelton Simmons who hits .300 with an OBP around .370 and steals 30 bases for you.

I have no problem with Sano's ranking at 14. His potential is wrapped up 100% in his bat and in his arm. If he strikes out a ton that reduces his offensive contributions significantly and if he can't hack it at 3rd and is forced to move to 1B his arm is worthless, not to mention the positional value decrease. If he can cut down on his strikeouts enough that he can hit .280 with an OBP around .360ish that is a big win. Then if he can play defense well enough to stick at 3B you have a player who might be an MVP candidate. On the other hand, if he only hits .245 with an OBP around .320 and he is at 1B well now you have a good but not great 1B. What a lot of you are missing is his BABIP. In the years that he has put up good batting averages his BABIP has been incredibly high. When his BABIP has approached more reasonable numbers his batting average falls too. The main difference between his season at A+ vs. AA? At Fort Meyers his BABIP was an absurd .397 so his batting average was .330. At New Britain his BABIP was .265 and so his batting average fell to .236. Both of those BABIPS are pretty extreme. If he can't fix his strikeout issue, then where that BABIP "normalizes" will ultimately determine his batting average.

I think the Twins should do whatever will help bring them back to contention. A higher draft position increases the odds of that happening. I have yet to read how Pelfrey will help. My own opinion, and those that I have read here as well, is that he was signed as a stop gap. A guy with pretty much no expectations that he will be good and really nobody will care how he pitches as long as it isn't as God awful as last season. That's a waste of an opportunity IMO.

Perhaps that is part of the discussion, but there were players they could have signed that helped in both the short term and probably the long term as well. They could have signed above average players and tried to contend this season. Assuming they weren't in contention they could have kept the players and then added a few more pieces next season or tried to flip the players for some good prospects that could help in the future. Either way the key was to add good players. On a separate topic, last season the difference between 73 and 74 wins was drafting 7th or 11th. Considering the new qualifying rules that is a pretty big difference. In 2011 the difference between 71 and 72 wins could have meant drafting 9th instead of 5th.

Without getting into his analysis of other teams, that I don't know well enough to comment on, I found his Twins blurb to have a lot of merit. Many posters seem to be glossing over his intent and skipping straight to the number of the ranking. He writes that we are hoping that Nolasco, Hughes and Pelfrey all improve. Well, how many people here have commented that they hope Hughes will be better after getting out of Yankee Stadium or that he is still young enough to learn? Aren't many of us hoping that Pelfrey has a bounce back a second year after TJ surgery? Aren't we hoping that Nolasco is the 2013 version with his 101 ERA+ instead of the '09-12 version with an 87 ERA+? That seems like a lot of wishing for improvement to me and yet the most likely outcome is that they remain who they have been for their whole careers. If that happens we will have a rotation filled with back half of the rotation pitchers that are signed for years to come. It is clear that the Twins could have a potent lineup as early as 2015 if everything fits together perfectly. Yet that will mean very little without great starting pitching. His point seems to be that if you're going to hope and take a risk, then do so on players that have a legitimate shot at being a great starting pitcher. A player that could be slotted into the front of the rotation when the Twins young bats arrive because they have precious few arms that fit that description arriving from the minors in the next year or two. If you don't like Tanaka, fine, there are other pitchers who fit his description of "hope and risk". If you can get past his ranking and who he names specifically as upgrades and drill down to his larger points I think he brings up a very valid point of view.

If all you want is an improvement over last season, well that shouldn't be hard to come by. That isn't enough for me. IMO, every move should be made with returning the playoffs in mind. If it doesn't help us get there then don't make the move. There in lies the crux of my problem with the Pelfrey (and the Hughes too) signing. I don't see how he helps get the team there in either 2014, 2015 or beyond. There were pitchers out there who could have helped with that goal. If you can show how Pelfrey helps the Twins to the playoffs in 2014 or beyond and how he was a better option than the likes of Colon, Hudson, Haren, Johnson, Kazmir, or any number of other pitchers that have or will sign 1 or 2 year deals I would love to hear it.

As Thrylos pointed out Worley has been a good pitcher except for last season. I know that you have the impression that he is a poor pitcher but that doesn't change the fact that he has been a better pitcher than Mike Pelfrey over the last few years. As for your polling I disagree. If you asked fans "do you want 1 more win this season, a season which we'll still be below .500, or do you want to draft up to 4 spots higher?" I think you'd get a lot of people answering draft.

I agree with you that worst to first isn't reasonable to expect and that moderate improvements need to be made. But Pelfrey isn't a moderate improvement. He was here last year, was terrible, has been below average for his career, is going to be 30/31 and there were better options out there that would have cost roughly the same amount of money. So by all means let's improve, but let's do it with players who are actually above average. So far this off season we've signed zero of those. How can a team expect to get to the playoffs with a bunch of below average guys?

If you look at bWAR then he averages .75 WAR per season. So, while he could exceed the value of his contract it is anything but a given. However, even if he does how does that help the team return to contention? Let's use your 3 WAR over 2 seasons. For the Twins, which won't sniff the playoffs, that is pretty much useless. He is just a filler while we wait for somebody better (which we could have signed this year or signed a 1 year contract and then signed a better pitcher next off season which looks like it could be a good crop of FA pitchers). But we could trade him you say! Just looking at recent trades, back of the rotation starters bring back very little of value (which shouldn't be too surprising). At best they seem to bring back a usable bullpen piece on a short contract like Joe Saunders in 2012. Of course they often bring back nothing of value, like Joe Blanton that same season. So, the upside here seems to be 3 WAR, a bullpen piece loaner or some combination there of. The downside is we are going to lose a young pitcher who could be just as good as Pelfrey. There is also the chance that this signing blocks Gibson or another prospect for a while. Finally, let's say he sticks around and we get a couple of WAR out of Pelfrey, we still aren't contenders and now we have a worse draft pick. Mike Pelfrey makes sense on a team with 4 good to great starters that is looking to shore up the back end of their rotation for a playoff push. For the Minnesota Twins, not so much.

Let's keep the perspective straight, you're right. Pelfrey is a back of the rotation starter. So yes, he might outperform his salary, but what does that get the Twins? He will still be a below average pitcher over the length of his contract. When your team has a dearth of front of the rotation options and a glut of back of the rotation options signing Pelfrey makes little sense. Will he be better than Worley, Deduno, Hendriks, etc... perhaps but what exactly do the Twins gain by having a marginally better #5 starter?

Wise One, you seem to agree with me whether you can see it or not. Your second sentence is exactly what I'm saying. You sign someone because of what you think they bring to the table and then judge your results based on those assumptions. In fact you continuously reference the knowledge base at the time the decision was made when you use the words "validate your judgement" and "meet your expectations". As for Privacy Laws, it is impossible for me to know that which I cannot know, therefore I can only base my opinions upon that which is known to me. So, as I've asked you many times before, please help us understand the inside workings of the Twins decisions so that we too may benefit from their knowledge. Until you do however I can only base my opinions on my current knowledge.

Of course the same could be said about the Mets and Pelfrey. Doesn't it speak volumes that the Mets didn't even tender Pelfrey a contract? Your argument doesn't seem to work well since a team could only ever known the in depth analysis of one half of the decision. For the Twins they knew Baker but not Pelfrey. Why should I blindly follow the Twins decision when they didn't follow the Mets decision? No, I'll stick with my original statement. Baker was the better signing at the time the decisions were made. He was just as risky but he had considerably more upside.

A signing is either good or bad based upon what was known before the signing occured, not because of what happened after. Based upon what was known publicly at the time of the signings Baker was the superior choice.

The funny part is the author of the Yahoo! article makes the same argument. He buries this little nugget in the middle of that article. So in the end, for all his railing on "team chemistry", it even comes back to talent for him.