THT Fantasy Focus

Posted by Victor Wang at 1:00am

Thursday, October 02, 2008

FOX Sports Expert League recap and strategy discussion (Part 1)

Well, the season is in the books. At the end of every season, it's very important to analyze what went right, what went wrong, and the various processes that went into your decisions. Even if you won your league, did you luck into the win or was your process actually sound? This can be very important, because if you lucked into it and don't recognize your mistakes, you're apt to make them next year when luck might not be on your side.

I have officially won the FOX Sports Experts League I participated in this season. I finished with 102 total points, 15 ahead of second place, and I thought that this would make a good, real-life example of how we should go about evaluating our seasons.

Let's start with my draft.

Note: I originally intended for this to be much shorter—just one article—but with some year-end data-crunching still required and the final data still a few days away, I figured I'd expand the discussion while we wait. I realize that hearing someone gloat about his league isn't the most entertaining thing to read, so I tried to cut back on the gloating and focus only on the process involved in my decision-making. Hopefully this should give you a good idea of how you should be evaluating your own season.

Picks 1 to 3

I had numerous contingency plans in place, but my early round strategy was the same with all of them. I wanted to spend my first three picks on three players who were worthy of being drafted early but didn't have much downside (i.e. low risk players). When I received the second pick, I knew that I was in a decent position to enact the plan I favored most and target the three guys I wanted most. David Wright came first, followed (luckily) by Carlos Lee and Mark Teixeira.

As Victor Wang has been talking about, using past risk to predict future risk isn't the most sound method, but even ignoring risk, these three all warranted their draft positions at the time of the draft. Plus, I don't believe I was actually using past risk, considering other factors besides past performance. They were all at good ages with no obvious playing time or injury concerns, plus all had solid, high sample size skill sets.

Pick 4

In round four, when Russell Martin was taken just a couple picks before me, my decision came down to Troy Tulowitzki or Carlos Guillen, with Guillen a tick higher on my list. Being that it was an expert league, though, the competition was very good and it wouldn't be enough to take all low risk players. I needed to take some high upside players as well, as the winners of experts leagues are very often those who manage to hit on a couple of these high upside guys.

This was a great opportunity to take one given the low risk associated with my first three picks and how closely I had Tulo and Guillen ranked. Drafting Tulo obviously didn't work out, but the process was sound, and I wouldn't have a problem doing it again next year in a parallel situation. Guillen had little upside, while Tulo had lots of it. Unfortunately, he had much more downside than I imagined. Perhaps that's an important lesson going forward, to focus as much on downside as upside when considering drafting a high risk player.

Picks 5 to 11

Not getting Martin and being that it was a two-catcher league, I needed my first catcher now. Brian McCann worked out very nicely. I then shifted into pitching mode, taking James Shields, Daisuke Matsuzaka, and Tim Lincecum. All turned in great seasons, although Dice-K's was very lucky (we'll talk more about him soon).

Round nine came my favorite pick, Jermaine Dye. Loved him coming into the year and was excited he fell so far in drafts. Then I took Johnny Damon to help with steals and runs and followed with Hideki Matsui. Damon turned out solid, Matsui mostly busted, but the process was okay there. I needed a couple of solid hitters and some steals, and they fit the bill.

Picks 12 to 15

Knowing who I wanted at the end of the draft and relatively certain I'd get who I wanted, I decided it was time again to take a few risks. I had three very solid starters already, so I went with two high upside ones in Dustin McGowan and Pedro Martinez. Neither really worked out, but the process was strong. I was mixing high upside players with those who had established track records and good skills.

The way the draft had played out, I knew at this point that I wouldn't be very competitive with steals and didn't want to take a Michael Bourn type just for the steals. I decided to instead load up on power and hope to grab a waiver wire guy or make a big trade mid-season using the surplus power (even if that meant overpaying in the traditional sense). With Wright, Lee, Damon, and talk that Tulo might run, I thought I'd have enough to at least stay even with the back of the pack and then grab some points later on with a big acquisition.

Still without a middle infielder, I went with Ty Wigginton, who I had very high on my list. I saw him as a poor man's Dan Uggla, especially with the Crawford Boxes helping his power. Frank Thomas came next, as I was very high on his skills — especially the power. He got treated like crap by the Blue Jays and later got injured, but as a 15th round pick it wasn't a big deal. Again, more of a high risk pick, but I'd have no problem taking Thomas again next year if he falls into the right situation (albeit a bit later).

Picks 16 to 20

In Round 16, I saw Joakim Soria still sitting there. I wanted to wait another couple of rounds before I took my first closer, but I couldn't pass up a guy who was on par with those taken in the 10th and 11th rounds here. If nothing else, he could be trade bait in May or June.

I then went with Jose Guillen, who I thought was incredibly underrated, and a high upside guy in Edwin Encarnacion. Both turned out to be very solid picks. Freddy Sanchez came next (bust) because I needed another middle infielder. After missing out on an elite second catcher, I went with Mike Napoli next to go along with my newly adopted late-draft power strategy.

Picks 21 to 23

Kevin Gregg came next, and a guaranteed closer in the 21st round was a good pick. Then I went high upside with Evan Longoria, finally hitting on one. He was integral to my season. Rafael Betancourt was a no-brainer next. All the closers were gone, and he had great skills backing up a guy with terrible skills. Didn't work out due to his bad luck, but the process was perfect.

Then, to hedge my McGowan/Pedro bets earlier, I went with Andy Sonnanstine, who I saw as a very solid pitcher (albeit without much upside). Jason Kubel came next as a high upside guy/hedge bet with aging outfielders in Dye, Matsui, and Damon. Finally, I took Chris Carpenter with the second to last pick of the draft. I simply placed him on the DL as a potential future trade chip and then picked up Jeremy Accardo to hopefully get some early saves and to have another full-time closer if B.J. Ryan had a setback.

Thoughts

Overall, I was very happy with my draft. I got my specific targets in the first three rounds, a top notch catcher, three great pitchers without giving up a top five pick, several guys I considered undervalued, and several high upside guys. Looking back, I don't see any really big blunders. I diversified, followed my plan, and came away with a solid team.

Player evaluation is a different topic entirely, but I do think mine was pretty good. I missed on most of my high risk guys, but I wasn't counting on all of them to be effective anyway. Being able to distinguish between which players you really thought would help and which you simply drafted for their upside (and were willing to let them go if necessary) is an important thing to do when evaluating your team at this time of the year. Because I missed on so many, I will look over my process in more detail at some point (hopefully it was simply bad luck), and I think Victor's methods will help a lot in that area.

Closer strategy

So I'm sure you noticed that I didn't take my first closer until the 16th round and said that I had intended to wait even later.

On draft night, I had a couple of options with how to approach closers. With everyone being an expert, I knew that a lot of their time is tied up in writing columns (with deadlines) throughout the season, plus whatever they need to do if they have real-world jobs and personal lives besides. In addition, I noticed that a couple of the participants were already trying to balance a few other things while the draft was going on.

Since I'm inclined to wait on closers to begin with, these two conditions made it a pretty easy decision. I would attempt to be the first to the waiver wire to pick up closers. If that didn't work, I'd pick up guys who I thought could be inheriting the role soon since in a 12-team mixed league only a few setup men get drafted and are owned at any one time.

As a fall-back, I figured I could spin at least one of the starters I drafted before round ten for a closer at mid-season, if necessary. I also thought that closers with slightly better than borderline skills or who weren't getting save opportunities would make good targets and wouldn't even cost one of my three starters.

Overall, I had lots of "outs" with big reward potential, so I chose that path.

Concluding thoughts

In Part 2, I'll look at the early portion of the season.

Posted by Derek Carty at 12:01am

Friday, October 03, 2008

FOX Sports Expert League recap and strategy discussion (Part 2)

You can read Part 1 here, so let's jump right into Part 2.

Early aggression

In the opening weeks of the season, many fantasy owners are afraid to make trades. They want to see a few weeks of player performance before they feel confident in making judgments that could impact their championship hopes. I hold the complete opposite opinion for reasons I've glanced over in the past and will discuss in more detail at a later date. Even if the rest of your league thinks that the first few weeks are meant to be calm and quiet and you can't work out out any trades, these weeks should still usually be your most active.

While it is absolutely imperative not to be seduced by small sample sizes, it is perfectly acceptable (encouraged, from my standpoint) to be very aggressive working the waiver wire early in the season. If a player is flashing legitimate skills, it is absolutely worth it to stash him at the back end of your bench and wait to see how he is doing once we reach a reasonable sample size. If, however, you wait until that point to pick players up, all the ones who can now be considered legit are long gone.

Early in the season, you want to act quickly and ask questions later. If a guy with good skills in the first two weeks regresses in the following two, drop him for another small sample, high skill guy. Don't be afraid to let your bench serve as an ever-moving carousel with lots of adds and drops. Make sure there's some logical basis behind the adds and drops, but aggression is a good thing.

There will always be players who come out of nowhere to have legitimately great seasons. If you can acquire one or two of those players for nothing, you gain a huge advantage. The huge rewards far outweigh the small sample size risks and the value of a bench spot, especially in deeper leagues.

So, how did I do?

Here were all of my pickups from April through the middle of May:
April 2: Mark Lowe
April 8: Scot Shields
April 12: Dustin Pedroia
April 13: Jeff Niemann
April 23: Scot Shields
April 24: Brian Fuentes
April 27: Max Scherzer
May 6: Christian Guzman
May 10: Ryan Franklin
May 14: Jose Guillen

As you can see, I was very active looking for saves, as per that portion of my strategy. I discovered early that I was capable of beating everyone to the wire for closers, which was important to know when deciding which strategies to pursue later on. I didn't get much out of Lowe or Shields, but I ended up with two very valuable closers in Brian Fuentes and Ryan Franklin. Added to Soria, Gregg, and Betancourt (Joe Borowski was falling apart), I felt very confident that I achieved my goals of competing in saves at a very low cost.

I did, however, miss out on many opportunities. Here are some of the guys who had significant value this year who were plucked off the wire:

March 25: Justin Duchscherer
March 30: Edinson Volquez
April 9: John Danks
April 9: Ervin Santana
April 9: Aubrey Huff
April 11: Todd Wellemeyer
April 11: Joe Saunders
April 13: Carlos Quentin
April 18: Cliff Lee
April 18: Jon Rauch
April 23: Jair Jurrjens
April 23: Ryan Ludwick
April 27: Jonathan Sanchez
April 30: Milton Bradley
May 7: John Danks (again)
May 7: Gavin Floyd
May 12: Ryan Ludwick (again)
May 12: Aubrey Huff (again)
May 12: Jay Bruce
May 18: Jon Lester

Obviously, some of the guys I really had no shot at. Some were taken before the season even started, others after just one week (equivalent to one start for pitchers). Still others, like Quentin, Lee, Ludwick, and Bradley, I really missed the boat on. Of course, I cherry-picked the pickups that worked out. For every one of these, there were probably three or four pickups made that didn't hold much value or busted entirely.

Again, though, this is a low risk, high reward strategy. While I could have done better, I did manage to get Pedroia and Guzman (in addition to picking back up Guillen, who I really liked but who was temporarily expendable because he got off to such a cold start). Those two really helped given Troy Tulowitzki's meltdown, Ty Wigginton's poor first half, and Freddy Sanchez being, well, not good (not that I really expected him to be; that's just how the draft played out).

I made a couple of young, upside pitcher pickups in Scherzer and Niemann, but neither really worked out. Should have gone for Lee and Danks.

As a side note, I picked up Scherzer with the intent to trade him, and failing to actually do so was a mistake I need to be careful not to repeat in the future. Several owners told me they weren't buying into the Scherzer hype, making him hard to move for the value I was hoping for, but getting something would have been better than eventually dropping him.

Midseason

Midseason was about looking for trades (I did this in the early portion of the season too, of course) and scanning the free agent wire for potential gems. I made tons of trade offers, and I don't know how many KFFL's Nicholas Minnix or RotoWire's Derek VanRiper said no to (sorry for the pestering, guys, but it was necessary). Eventually, I was able to work out my first trade on May 20.

I traded Franklin and Dustin McGowan for Johnny Cueto and Kerry Wood. I originally talked about the trade here. Cueto was looking excellent but didn't work out as well as I'd hoped, mostly due to bad luck (I think he makes a great 2009 sleeper). McGowan wasn't living up to the potential I drafted him for, and I thought Cueto was an improvement. Franklin's skills looked like a ticking time-bomb, and while Kerry Wood had concerns, I trusted him much more than Franklin. Overall, great process and good results.

I also spent the middle of the season accumulating closers. Rafael Betancourt didn't work out, but throughout the year I received saves from Dan Wheeler, Damaso Marte, Jonathan Broxton, Joel Hanrahan, John Grabow, Fernando Rodney, Aaron Heilman, Brad Ziegler, Frank Francisco, Luis Ayala, Matt Lindstrom, Chad Qualls, and Jensen Lewis. By the middle of July, I was rocking at least six closers at any one time.

Trade deadline

As per my own advice, I really tried to work out trades throughout June and July, but nothing really appealing came along. I knew that I would eventually need steals (it was my plan all along to acquire them at midseason), but I was staying relatively competitive in the category and wanted to make sure I built up a good lead in the power categories before trading some away for the steals. In my situation, it was best to wait until closer to the deadline because the homers and RBIs were more important than steals at that point.

On July 30, though, (two weeks before the deadline, exactly the time I was hoping to make my deal) I traded Tulowitzki, Kevin Gregg and a fresh-off-the-wire Fernando Rodney for Brandon Phillips. This filled a big hole at middle infield, gave me some of the steals I needed, and even helped out with the other categories. And the best part was that I gave nothing I valued much. Owning eight closers at that point, I could afford to lose two, and Phillips was a clear upgrade over Tulo (though looking at it from the other point-of-view, it wasn't a bad trade either; our situations were different).

Still, Phillips wasn't an elite basestealer, and I didn't think he alone would allow me to get all of the potential points in steals and runs. So John Halpin and I worked out a trade that ended up benefiting us both, although I believe this deal is what really propelled me to the top. I received Jose Reyes, Javier Vazquez, John Danks, and Mike Gonzalez for Tim Lincecum, Joakim Soria, Daisuke Matsuzaka, and Christian Guzman.

Most importantly, this netted me the elite basestealer and run scorer I needed. I had the lead in saves at this time, and I was in danger of losing only one point. Saves are saves, and I thought Soria for Gonzalez was a break-even bet. Even if Gonzalez busted entirely, I still would be okay making the swap.

Losing Lincecum and Matsuzaka might have seemed like a big hit, but I didn't see it that way. In fact, trading these two away symbolized my favorite part of my strategy for the year. I'll get into why in Part 3.

As I side note, I really should have traded Matsuzaka earlier. All the signs were there for an implosion, and while a number of the owners saw this and didn't have much interest, it ended up being a situation like Scherzer where I really wanted more than was being offered. Luckily, I was eventually able to move him (and when I did, John said that he was concerned about him), but getting greedy could have cost me.

Concluding thoughts

In Part 3, I'll look at the end of the season and give my final thoughts on everything.

Posted by Derek Carty at 12:58am

Saturday, October 04, 2008

FOX Sports Expert League recap and strategy discussion (Part 3)

You can read Part 1 and Part 2 here, so let's jump right into the final part.

Late season points in droves

I gained nearly 15 points from the middle of August to the end of September. It went from a heated battle for first to a run-away in just six weeks. Jose Reyes, Brandon Phillips, and a Denard Span pickup netted me a bunch of steals and runs points, but I also grabbed quite a few in the pitching categories.

Thanks to some excellent (and lucky) pitching by Tim Lincecum and starting six or seven closers with great stats every day, I had a huge lead in ERA. Losing Lincecum and Daisuke Matsuzaka (who I wasn't starting anyway because of his control) was no big deal at all because they had already done what I wanted them to; I had my ERA lead and was doing well in WHIP. Going forward, they could only provide marginal value in wins and strikeouts.

They were actually near replacement level in the context of my team, because all I cared about for the final month-and-a-half were wins and strikeouts. Any additional stats a starting pitcher provided was meaningless. Jorge de la Rosa and Bronson Arroyo were nearly equivalent options, plus I got Javier Vazquez and John Danks to make sure I had at least some quality innings getting logged. Essentially, I leveraged Lincecum and Dice-K's excellent value for offensive points and lost nothing in the way of pitching points.

Up until now, I had sat tight with my starters. I didn't engage in spot starting and just rode the wave. After the deadline, though, I slowly but surely began to spot start. By the middle of September, I managed to keep my big ERA lead and was picking up at least one starter almost every single day (some days as many as three or four) to spot start and accumulate strikeout and wins points. By John Halpin's calculations, I was making "937 transactions per week."

This is a highly volatile strategy under most conditions, but I was in a somewhat rare situation where I could afford to do it with no risk attached. The WHIP point or two I put at risk were made up for by the strikeout and wins points I could gain without any risk at all. Start a lot of pitchers, and those two things will come.

Final standings

I ended up getting 12 points in both HRs and RBIs, as I intended with my draft day plan. I finished with 11 in runs and six in steals (gaining several points over the last few weeks). I also managed 11 in batting average, more than I expected (but I would have been fine without them).

I got 12 points for saves, which was huge considering I paid almost nothing for them. I got 12 points in ERA and 11 in WHIP, and finished with 8 in both wins and strikeouts (again, gaining several over the last few weeks).

Keys to success

One of the big keys to my success was avoiding injuries. One of my teams this year had some 15 or 20 injuries, while this one had relatively few.

My injured players were Phillips, Troy Tulowitzki, Carlos Lee, Evan Longoria, Hideki Matsui, Johnny Cueto, Dustin McGowan, Pedro Martinez, Ty Wigginton, Frank Thomas, and Mike Napoli. Seems like a pretty long list, but most of the the big ones came late in the year and had minimal impact. Tulo and Matsui were the only two that really hurt me.

I made no concerted effort to avoid injuries, as this really is a field that has yet to be properly researched, and I simply had some good fortune. Can't bank on that next year.
Hitting on some of my late round picks really helped. Evan Longoria was huge, and Joakim Soria, Kevin Gregg, Jose Guillen, and Edwin Encarnacion also all came past round 16 and were important. I got great value on Jermaine Dye, who did about what I expected.
That my big three (David Wright, Lee, and Mark Teixeira) delivered as expected was also huge.
Some good fortune with my starting pitchers (Lincecum, James Shields, and Matsuzaka all had lucky ERAs) helped immensely as it enabled me to pursue my late-season pitching strategy.
Some indirect good fortune came in my drafting Brian McCann. If Victor Martinez was sitting there (maybe even in round four)... oh boy. And if I remember correctly, I had intentions to draft Jorge Posada in round nine as my second catcher, but he was taken a little before then. Mike Napoli didn't do much, but using a high pick on Posada would have hurt.
Some good free agent pickups (Dustin Pedroia, Christian Guzman, Span, David Murphy, Adam Lind, all the closers) helped a lot as well.
And, of course, my mid-season trades and subsequent spot-starting were critical.

Concluding thoughts

That wraps it up. Hopefully this gives you a clearer idea how to go about evaluating your own seasons.

I'd like to say thanks to everyone in the league for making this such a pleasant season, and a big thanks goes out to Roger Rotter, John Halpin, and the other guys at FOX for extending the invitation.

Posted by Derek Carty at 12:01am

Monday, October 06, 2008

Player risk profile: Ben Sheets

For more background information and details on my methodology, read my first player risk profile article on Troy Tulowitzki here.

Background

Ben Sheets has been one of the more frustrating pitchers around. When he's on, there are few pitchers who are better than him. However, Sheets has had trouble staying healthy. For example, take a look at his last five seasons worth of inning totals: 237, 156.7, 106, 141.3, and 198.3. When Sheets is healthy, he is one of the most valuable fantasy pitchers in the game. Will Sheets stay healthy? I think the more important question is, how do you want to value Sheets given the risk he brings? To answer this, let's take a look at exactly what kind of risk he carries.

Qualitative Risk Assessment

Experience
Probability: Very Low
Impact: Very Low
Overall Risk: Very Low

While Sheets has not been completely healthy in the past, we still have a large sample size on him. Additionally, we've been able to see Sheets when he's healthy and dominant and when he hasn't been as healthy or dominant. Therefore, we have a pretty good idea of all Sheets can be.

Playing Time
Probability: Very Low
Impact: Very Low
Overall Risk: Very Low

Remember this risk factor is based on playing time risk not due to injury. Whoever signs Sheets in the offseason will likely view Sheets as their number one or two starter, meaning that Sheets would need to perform really, really poorly to see a major loss of playing time not due to an injury.

Age
Probability: Medium
Impact: High
Overall Risk: High

Research has shown that if pitchers survive to their late twenties, they will see improvements in their performance during that age until declining. Sheets will be 30 next year which means he is right near his peak right now. However, Sheets clearly had his best season in 2004 and hasn't been able to pitch 200 innings since then. While he did make 31 starts last year, his peripherals were well below those of past seasons. I think we can say there's a pretty good chance Sheets is past his peak.

Burnout
Probability: High
Impact: Very High
Overall Risk: High

I think we can all see that Sheets is very risky when it comes to his health. He is a guy who has a wide variance in his playing time projection. However, that small chance that Sheets could give over 200 innings of quality pitching might make him worth a gamble to some people.

Skill Risk
Probability: Low
Impact: Low
Overall Risk: Low

Sheets still has a very good skill set with the biggest risk being if he sees a moderate gain in his fly ball rate, which would elevate his home run rate into a dangerous territory. Still, Sheets is a very talented pitcher, and that's not going to change any time soon.

Overall Risk Level
High yellow. Sheets' skill set and experience are the pluses, but they don't quite make up for the big injury risk he presents. This might surprise some people as I can imagine many describing Sheets as a high-risk player or worthy of a red ranking. However, Sheets' low performance risk makes it so he's quite not in that territory, though pretty close.

Quantitative Risk Assessment

Breakout: 2 percent
Collapse: 35 percent
Beta: .89

PECOTA saw Sheets with a low volatility forecast. However, the volatility was in the wrong places as PECOTA did not see Sheets as having much upside but a lot of downside. While Sheets did throw his most innings since 2004, he also gained a year of age so I don't think his forecast will be too different next year. Sheets should have a good reliability score next year.

Overall Risk Assessment

Sheets has a lot of risk to him but almost all that risk is health related. I will disagree with PECOTA and say that Sheets does have upside but this upside is playing time related rather than performance related. However, since a lot of people will focus on the 198 innings he threw this year and quickly glance over his past innings pitched totals, Sheets might not be worth chasing. If you lean towards a more risk-seeking strategy, Sheets would be a guy I would consider targeting. However, if you are more risk averse, I would advise to look at other pitchers first unless you can get Sheets at a big discount.

Posted by Victor Wang at 12:01am

Tuesday, October 07, 2008

A tale of three third baseman

In the 2005 MLB amateur draft, considered one of the richest talent-wise in recent history, three third basemen were among the first five selections. Even though that was just three years ago, all three have made it to the major leagues. They are Alex Gordon (second), Ryan Zimmerman (fourth) and Ryan Braun (fifth). Let's look at each player's path to the majors, how they've fared, and what their outlook is for next season.

Path to the majors

Alex Gordon: Gordon was drafted out of the University of Nebraska after his junior year.

In his final college year, he batted .365, had a .493 OBP and slugged .754. He won plenty of awards following that season, including the Golden Spikes Award and the ABCA National Player of the Year Award, but more importantly got selected second overall with a sweet $4 million signing bonus.

In his first full year in the minors, already in Double-A, Gordon played phenomenally well. He put up a .325/.412/.588 line in 558 plate appearances. Slightly alarming was his high strikeout rate at 23.3 percent, but his above-average walk percentage of 12.9 and gaudy power numbers helped nullify that concern. More awards followed that season, including Texas League Player of the Year and Baseball America Minor League Player of the Year. More importantly he was promoted to the major leagues and given the opportunity to win the starting third base job from Mark Teahen.

Ryan Zimmerman: Zimmerman, the fourth overall selection, also was drafted out of college—the University of Virginia. Zimmerman was signed for $2.975 million, just over a million less than Gordon. He was signed relatively quickly and therefore able to get some minor league action in 2005. He was first placed in Single-A ball, where he stayed all of four games—posting a 1.530 OPS—before he was promoted to Double-A. There, Zimmerman's stats came down to earth. In 254 plate appearances, he posted a .326/.371/.528 line.

The Nationals were impressed enough to make Zimmerman a September call-up, and yes, we are still in the year 2005. I believe he and Braves (now A's) reliever Joey Devine were the only members of the 2005 draft to make it to the majors that year. Check out what Zimmerman did: He hit for a .397 average over 62 PAs. He had no home runs, but 10 doubles, and his .500 BABIP was justified by a 38.3 line drive percentage. So impressed were they with their 21-year-old third base prospect that the Nationals decided to play waitress and handed the starting third base job to Zimmerman on a silver platter.

Ryan Braun: Braun was selected fifth overall and signed quickly, receiving a $2.45 million bonus. He would spend the rest of that season unevenly split between Rookie League Ball and Single-A, where he recorded a combined 1.004 OPS. With Braun, the Brewers stood up much straighter than the Nats did with Zimmerman and the Royals with Gordon, starting 2006 with him at the A+ level.

Braun would split the 2006 season at the A+ and Double-A levels, playing in exactly 59 games at each. He struggled somewhat at A+, but nevertheless was promoted to Double-A where his stuff began to shine. Here's a quick table of Braun's 2006 Double-A half-season, and his 2007 Triple-A quarter-season:


Season	Team		League	PA	HR	RBI	AVG	OBP	SLG	BB%	K%	
2006	Huntsville (AA)	SL	252	15	40	.303	.361	.589	8.3	19.9	
2007	Nashville (AAA)	PCL	132	10	22	.342	.417	.701	11.4	9.4

As you can see, Braun showed that he had the raw skills to succeed at the higher levels of the minor leagues in his 2006 Double-A season, and also the plate discipline to thrive in the majors in his Triple-A stint. Actually, the ridiculous numbers Braun registered in two months at Triple-A proved he possessed everything a hitter needs to pan out in the majors. The Brewers agreed, probably influenced most by the much reduced 9.4 strikeout percentage, and promoted him to the majors on May 25.

In the majors

Zimmerman got his first taste of the majors in September and been the Nationals starting third baseman ever since. Gordon played all of one season in Double-A before winning a starting job in 2007 spring training and has started since. Braun spent the most time in the minors of the three, a whopping one and a half years, so you could argue that all three have been rushed to the majors. They have found varied success in the big leagues.

Gordon: Remember him? He was our Golden Boy. His trophy room was bursting with awards and he made it to the majors without playing in a single game at Triple-A. On his Baseball Prospectus PECOTA card was written "The number one prospect in baseball." They were being completely serious.

To make room for Gordon at third base, the Royals were willing to "banish" Mark Teahen to the outfield, and you bet they did. So now we've cleared a spot in the infield dirt and left a hole in the lineup for Gordon to fill... how'd he do with it?

Season	PA	R	2B	HR	RBI	SB	AVG	OBP	SLG
2007	601	60	36	15	60	14	0.247	0.314	0.411
2008	571	72	35	16	59	9	0.260	0.351	0.432

The quick answer: not well. Not terrible, but he certainly did not live up to expectations. An average under .250 is uninspiring, an on-base percentage of .314 is not helping the team—even the Royals—and his 15 home runs are rather pedestrian for a third baseman.

Regardless, there was still plenty of optimism surrounding Gordon heading into the 2008 season. The Royals definitely were going to continue starting him, and in fantasy drafts he was taken at an average position of 138.3, which correlates to the first pick in the 12th round of a 12-man league. Take a quick glance at the table above and you'll see Gordon did not live up to the expectations for a second straight season. His stats increased pretty much across the board, but only slightly. We've seen players who have been unlucky for one season; maybe Gordon has simply been unlucky for two?

Season	BB%	K%	LD%	BABIP	HR/FB
2007	7.00	25.20	19.50	0.304	8.50
2008	11.80	24.30	21.00	0.314	8.90

Ehhh, it seems that luck has not played a major factor in Gordon's major league career so far and that explains the surprising amount of consistency he's shown. Everything has remained relatively constant in the past two years. I don't believe Gordon ever will develop into a 40-homer threat—as indicated by his below-average HR/FB percentage and current gap power—so he's going to have to significantly reduce that strikeout rate if he's ever going to find success in the majors.

He needs to raise his 76 percent contact rate to at least the league-average 81 percent. His line drive percentage and BABIP all have been consistently around league average, so nothing to discuss there. This leaves only the walk rate, and it is the one promising aspect of his 2008 season. That jump form 7 to 12 percent is huge and shows the most important thing: development. Remember, Gordon did not have the luxury of developing his skills in the minors; he's having to do it now, in the majors, and thankfully that is what we're seeing.

Zimmerman: When we left off with him, he had been called up to the majors in September of '05 and made pitchers look like Girl Scouts selling cookies. The year 2006 was not much different. Take a look:


Season	PA	R	2B	HR	RBI	SB	AVG	OBP	SLG
2006	682	84	47	20	110	11	0.287	0.351	0.471
2007	722	99	43	24	91	4	0.266	0.330	0.458
2008	466	51	24	14	51	1	0.283	0.333	0.442

For a rookie season, a .290 average, 20 home runs, 80 runs, 110 RBI and 10 stolen bases is pretty impressive. Hanley Ramirez just edged him out for NL Rookie of the Year, but Zimmerman still finished the season with his head held high. At this time, Gordon was still mashing away in Double-A oblivion and Zimmerman already head a quality MLB season under his belt. You hear that sound? It's Zimmerman's service time clock ticking away to a huge free-agent contract.

His 2007 season was a slight step backward in most regards. Fantasy-wise, it was still solid but the 20-point drop in batting average had an ominous air. Still, plenty of players rushed to the majors have experienced much worse sophomore slumps than Zimmerman, if you can even call his a slump. Let's look at some of his indicating stats to see if luck played a role:


Season	BB%	K%	LD%	BABIP	HR/FB%   GB/FB%	O-Contact%
2006	9.0	19.5	21.8	0.329	11.4	1.18	54.51
2007	8.5	19.1	16.9	0.298	11.4	1.1	60.80
2008	6.8	16.6	19.8	0.312	11.5	1.35	69.34

Most of these indicators stay in line over the three years. The dip in batting average in 2007 is explained by the regression in BABIP, which is sequentially explained by the dip in line drive percentage. Coming into 2008, Zimmerman looked like a safe choice for decent production from third base. He was taken on average 96th overall and was considered by many to be the last secure third baseman available in a surprisingly shallow pool of players. If you missed out on Zimmerman, then you were left with a gamble of a player like Evan Longoria or Edwin Encarnacion.

But 2008 was a disappointing season for Zimmerman. It was not because he played terribly or missed most of the season, but rather a combination. He did miss playing time—almost two months from May 25 to July 22 with a shoulder injury— but still managed to accumulate 466 plate appearances. Problem was, he played unspectacularly in the four months in which he played. So a combination of injury and mediocrity left his end of the season numbers overall underwhelming.

In terms of luck stats, not much changed. The decrease in strikeout percentage was nullified by a similar decrease in walk percentage. And the small curb on his power numbers was the result of him putting the ball on the ground more often. The one thing that did change significantly was his O-Contact percentage. Simply put, that is the percentage of pitches a batter makes contact with when the ball is outside the strike zone and when the batter is swinging. Not much research has been done on the effects of making contact with pitches outside the zone, but the general assumption is that it's undesirable. Zimmerman's O-Contact percentage obviously has shot up steadily over the past three years. Just something to keep an eye on.

Braun: If you remember, Braun was the player handled most sensibly. His promotions were deserved and the next one didn't come until he proved himself at that level. If you also remember what he did for two months at Triple-A in 2007, you'll realize that his May 25 promotion to the majors was well deserved. At that point, Zimmerman had been in the majors for one year and two months and Gordon just two months. Gordon was struggling mightily, batting just .194. Zimmerman wasn't having a great season either, so far with a .245 average. With those being the numbers produced by supposedly superior players (based on where they were drafted) and of players with more major league experience than Braun, who was expecting much of anything from him?

Expectations obviously did not matter, as Braun absolutely tore up the major leagues from June to September. Check it out:


Season	PA	R	2B	HR	RBI	SB	AVG	OBP	SLG
2007	492	91	26	34	97	15	0.324	0.370	0.634
Season	BB%	K%	LD%	BABIP	HR/FB%    GB/FB%			
2007	6.0	24.8	16.3	0.367	22.1	0.86

An absolutely monster season all-around. In less than 500 plate appearances, Braun was able to blast more than 30 home runs and fell just short of the 100 plateau in runs and RBI. It netted him the NL Rookie of the Year Award, just ahead of Troy Tulowitzki.

Fantasy teams that picked him up midseason benefited tremendously from his condensed production and often won leagues. Taking a look at his "luck" stats, however, you see that luck played an integral part in Braun's season. He struck out too often, did not draw walks often enough, hit for a low percentage of line drives, had an abnormally high BABIP, and owned a possibly inflated HR/FB percentage. Regardless of all of those bad signs, Braun was still picked on average 16th overall, in the second round. That was way too early in drafts for me to select him, and I did not select him in an of my leagues or mock drafts. Let's see how he did in 2008:


Season	PA	R	2B	HR	RBI	SB	AVG	OBP	SLG
2008	663	92	39	37	106	14	0.285	0.335	0.553
Season	BB%	K%	LD%	BABIP	HR/FB%	GB/FB%			
2008	6.4%	21.1%	17.3%	0.308	17.5%	0.88

Braun certainly did see a regression in his stats in 2008, but he was so good in 2007 that he was still a roto monster in '08. His BABIP regressed significantly and consequently his batting average took a dive as well, yet floated from .280 to .300 for most of the season. He maintained his home run total, albeit not at the same PA or flyball rate of last year. Overall another solid season for Braun—his second in two tries—and he removed any of the doubt surrounding his ability.

Outlook

Gordon: Alex Gordon has not yet had his breakout season and it is due to come over the next three years. It could be next year, but I feel he will provide the most value in 2010. I'm not sure how far he will fall in drafts because there usually tends to be one Gordon-lover who selects above where I'm comfortable. But if your league is without one, feel free to target him toward the mid-to-late rounds. However, I do feel that 2009 will be another learning season for Gordon, the hype will be gone entering 2010, and he will be a great sleeper that season. Who knows what fantasy baseball will be like then...

Zimmerman: Ryan Zimmerman has the trickiest future to predict of the bunch, and I do not know what to expect from him next year. The good news is that you probably won't have to expend a high draft choice to select him, so he could have value if he falls as far in drafts as I think he may.

Braun: Ryan Braun is an absolute monster, no doubt. I'd have zero hesitation in selecting him in the third round of drafts, but he tends to go sometime in the second. Probably not for me.

And just a nice little note to sum things up: Notice how Ryan Braun, who was drafted last of the three, is now by far the most accomplished. So much for scouting.

References and Resources I used a whole bunch of sites for this one, including The Baseball Cube, FanGraphs, Cot's Baseball Contracts, Baseball Prospectus, and, of course, our own Hardball Times stats.

Posted by Paul Singman at 1:01am

Wednesday, October 08, 2008

First mock draft of the offseason

Oct. 2, just one day after the regular season ends, and I'm already mock drafting. Oh boy. Thanks to Mike Muschiano of the Poughkeepsie Journal for the invite.

Now, I didn't do any kind of preparation for this mock and I didn't even have a really clear idea what kind of strategy I would employ. I went in very flexible and just wanted to see how the draft played out, to get an early, rough feel for how the market value of players has changed.

This being said, here's my roster, followed by some comments about my team and about some other selections I found interesting. At the end of the article, you'll find some thoughts on the importance of mock drafting in general.

+-----+-------------------+-------+
| Pos | Player            | Round |
+-----+-------------------+-------+
| C   |	Joe Mauer         | R5    |
| 1B  | Carlos Delgado    | R7    |
| 2B  |	Ty Wigginton 	  | R13   |
| SS  |	Michael Young 	  | R10   |
| 3B  |	Miguel Cabrera 	  | R1    |
| CI  |	Adrian Beltre 	  | R11   |
| MI  |	Carlos Guillen 	  | R15   |
| OF  |	Carlos Beltran 	  | R2    |
| OF  |	Carlos Lee 	  | R3    |
| OF  |	Vladimir Guerrero | R4    |
| OF  |	Corey Hart 	  | R6    |
| UT  | Nelson Cruz 	  | R17   |
| SP  |	Edinson Volquez   | R8    |
| SP  |	Javier Vazquez 	  | R9    |
| SP  |	Aaron Harang 	  | R12   |
| SP  |	Wandy Rodriguez   | R14   |
| SP  |	Jonathan Sanchez  | R16   |
| SP  |	Randy Johnson 	  | R18   |
| RP  |	Manny Corpas 	  | R19   |
| RP  |	Huston Street 	  | R20   |
| RP  |	Frank Francisco   | R21   |
+-----+-------------------+-------+

Because I'm still in the process of reviewing player seasons, please keep in mind that the following are general thoughts that could change a little down the road.

Right now, though, I like my team a lot. I got six starting pitchers who I believe could be among the top 30 on my final draft board and didn't take the first until round eight.

I took Miguel Cabrera as my first pick, and I think at pick eight that's a good selection.

My outfield looks pretty stacked; I was glad to get Carlos Lee in round three, but Vladimir Guerrero in round four was a slight shock.

Joe Mauer in the fifth might have been a bit of a stretch in a one-catcher league.

I'm not sure if I'll ultimately condone taking Carlos Delgado in the seventh round, but I do like him.

Ty Wigginton continues to be undervalued. I kind of wanted to leave him out there to see how far he'd fall, but there will be plenty of time for that in the coming months.

If Nelson Cruz's MLEs were accumulated in the majors, no way does he last this long. Of course, MLEs aren't quite as good predictors as actual major league stats, but I still think he's a great pick in round 17 who could be a five category contributor.

With Brian Fuentes likely leaving Colorado, I think there's a good chance Manny Corpas will be closing in 2009, making him a great end-game pick.

I also believe that the A's will try to trade Huston Street at some point, and as long as they think he can pitch effectively, the best way to build that trade value is by having him close. He's my bet to open 2009 pitching the ninth.

Frank Francisco has better skills than C.J. Wilson and did well closing games at the end of the year, and while a free agent might also be an option, in the last round I think he was a solid pick.

Surprise Picks

Pick No. 1: Hanley Ramirez
Round 2: Evan Longoria
Round 2: Tim Lincecum
Round 4: Dustin Pedroia
Round 5: Victor Martinez
Round 5: Joey Votto
Round 5: Francisco Liriano
Round 5: Alexei Ramirez
Round 7: Troy Tulowitzki
Round 8: Chris Davis
Round 8: Derek Jeter
Round 10: Jay Bruce
Round 11: Matt Wieters
Round 11: Pablo Sandoval
Round 11: David Price
Round 14: Mike Pelfrey
Round 14: Milton Bradley
Round 15: Carlos Guillen
Round 17: Miguel Tejada

We see a clear trend of young (often second-year) players getting taken early. I don't know if this is the best idea, although they'd all need to be analyzed individually, of course.

Hanley Ramirez went number one over guys like Alex Rodriguez and David Wright. If I received the number one pick, there's a very real chance I would have taken Hanley, and I'm actually hoping that my final rankings put him number one. The drop-off at shortstop is pretty big, and Hanley helps you in so many ways.

Tim Lincecum in the second round was a shocker, especially with some people thinking he could break down next year (for what it's worth, I probably wouldn't have a problem taking him at a reasonable price because evidence is inconclusive that pitchers who throw a lot of innings are at a higher risk of injury). A pitcher in the second round isn't always a given, and seeing Lincecum there was very interesting.

I was very curious to see where Pedroia ended up, and we see he went in the fourth round. I think he could put up very similar numbers to last year, so I really don't have much of a problem with the pick. In the second round, well, then I probably would.

Francisco Liriano, I was hoping, would fall into the teens this year, but he managed to get taken earlier than he was going last year. His ADP will be fun to watch this off-season.

Someone's expecting a bounce back year from Victor Martinez, taking him in round five as the fourth catcher off the board. I'd have a hard time taking him this early.

I know I already covered young guys, but Matt Wieters in the eleventh round was quite unexpected. He's got talent, but he's played just one year of professional baseball, accumulated just over 200 at-bats at Double-A, never played at Triple-A, and isn't guaranteed the starting spot for the Orioles. Seems early to me, especially in a one-catcher league.

Milton Bradley fell to round 14 after a tremendous year. This could be because he wasn't very high on the Mock Draft Central default rankings (which aren't completely tailored to 2009 yet being it's so early in the off-season) and got overlooked. I filled my outfield early and liked Nelson Cruz for my utility spot later, which is why I didn't pick him up.

Carlos Guillen was no mistake. His value appears to have plummeted. I was the one who took him, but he's on the list because of how late I got him. I'll talk about him some more in a few weeks and see how early it would be reasonable to take him. Miguel Tejada is another shortstop whose value fell off the table, and Derek Jeter in round eight was relatively late as well. Troy Tulowitzki, however, went much earlier than I'd expected. People loved going for the upside in this draft, it seems.

Click here if you'd like to see everyone's picks.

Thoughts on mock drafting in general

Mock drafting is a great practice for everyone to get into. Obviously this is incredibly early and it isn't necessary to start in October, but mock drafting will benefit you greatly when the time comes for your actual draft. Frequently mock drafting throughout the off-season will give you an idea how player value is changing, more so than simply looking at an Average Draft Position (ADP) report will.

For example, I'd bet that Lincecum's ADP ends up somewhere around the start of the fourth round. However, there may be great variation in where he is taken. In this draft, he went in the second round because there wasn't as great of a concern about a breakdown. In a draft full of risk-adverse owners or owners who don't believe in drafting starters early, however, he could fall to the seventh or eighth round.

Simply looking at his fourth round ADP doesn't give you this information. By being involved in a number of mock drafts, you may be able to see these trends evolve more clearly.

Once you know the situations in which Lincecum goes in round two, those where he goes in round four, and those where he goes in round seven, you can apply this knowledge to what you know about the owners in your league to estimate how long you could reasonably wait to take him yourself.

If you didn't do the mock drafting, you might see the raw ADP and take him in round three when you could have waited a full two or three rounds longer. That is huge value you're giving up.

Concluding thoughts

I'll be participating in another mock draft on the 20th I believe, so I'll be sure to check back following the draft and let you know how my drafting habits changed and where some of the players in the "Surprise Picks" section went.

Posted by Derek Carty at 12:01am

Thursday, October 09, 2008

Player spotlight: Ben Sheets

On Monday, Victor Wang went over Ben Sheets' risk profile. Today, I'd like to discuss his specific skills and check out a little PITCHf/x data.

Numbers

+------+-------+------+----------+-----------+-------+------+---------+------+-------+-------+
| YEAR | IP    | ERA  | LIPS ERA | DIPS WHIP | K/9   | BB/9 | K/BB RI | xGB% | BABIP | HR/FB |
+------+-------+------+----------+-----------+-------+------+---------+------+-------+-------+
| 2004 | 237.0 | 2.70 |     2.77 |      0.99 | 10.03 | 1.22 |    1.35 |   42 | 0.292 |  10.0 |
| 2005 | 156.7 | 3.33 |     3.21 |      1.10 |  8.10 | 1.44 |    0.76 |   37 | 0.281 |  10.0 |
| 2006 | 106.0 | 3.82 |     2.91 |      1.00 |  9.85 | 0.93 |    1.35 |   40 | 0.342 |   7.6 |
| 2007 | 141.3 | 3.82 |     4.01 |      1.25 |  6.75 | 2.36 |    0.23 |   36 | 0.287 |   8.7 |
| 2008 | 198.3 | 3.09 |     4.00 |      1.20 |  7.17 | 2.13 |    0.34 |   40 | 0.285 |   7.0 |
+------+-------+------+----------+-----------+-------+------+---------+------+-------+-------+
|   1H | 108.0 | 2.85 |     ---- |      ---- |  7.90 | 2.05 |    0.55 |   42 | ----- |   9.1 |
|   2H |  75.3 | 3.46 |     ---- |      ---- |  5.97 | 2.27 |    0.00 |   36 | ----- |   4.0 |
+------+-------+------+----------+-----------+-------+------+---------+------+-------+-------+

While his ERA might not reflect it, Sheets definitely had a down year — his second in a row actually. His LIPS ERA sat just above 4.00 in both 2007 and 2008. His raw skills did improve a little bit this year, as seen most clearly in his K/BB Run Impact and expected ground ball rate (xGB%) (LIPS ERA doesn't show it because LIPS is a relative stat that is calculated according to each year's specific run environment &mdash this year's NL must have been a little easier than last. EDIT: Also, Sheets induced more infield flies in 2007; his K+IFFB/9 was 8.21 in 2007 compared to 8.12 in 2008), but he is clearly a worse pitcher than he was from 2004 to 2006.

Sheets will be 30 next year, and echoing Victor, "I think we can say there's a pretty good chance Sheets is past his peak."

If we look at his half-season splits, we notice that his first half was significantly better than the second half. Higher strikeout rate, higher expected ground ball rate, and lower walk rate. I suppose this could give us some hope that he might be able to bounce back in 2009 since he pitched pretty well for at least a portion of 2008. Of course, the opposite side of the coin is that he was pretty bad in the second half, and if that carries over into 2009, he and his owners would be in big trouble.

One interesting thing to note is Sheets' Batting Average on Balls in Play (BABIP). Aside from a horrible 2006 (during which he threw the fewest innings), it has been much better than league average (which is around .300). This is a repeatable skill for a select few excellent pitchers, and there's a pretty good chance Sheets possesses that skill.

Also take note of his Home Run per Fly Ball Rate (HR/FB). It has been significantly better than league average (11 percent) three years in a row now. After three years, it's time to consider the possibility that this is a legitimate skill that Sheets has picked up. We must consider, though, that Sheets has pitched just 445.2 innings over those three years — the equivalent of two seasons for many pitchers. We could also take pause when we notice that it was much closer to 2004 and 2005 levels in the first half of the season (although that may not mean much).

Make of this information what you will. If he can continue to keep the BABIP down and the HR/FB trend is based on legitimate skill, Sheets' actual ERA could continue to beat his LIPS ERA going forward. This would allow Sheets to remain a top pitcher despite his peripheral decline, although I don't know if I'd bet on it.

PITCHf/x

Since Josh Kalk is having some trouble with the PITCHf/x data (completely understandable after nearly a full season's worth of pitch tracking), I won't go too in-depth with Sheets. We'll do some basic scouting and wrap things up. Because of this, take note that the data here only reflects his starts up until July 29, although once we get the rest I'll be sure to take a look and see if anything changed in his physical skills between his very good first half and his not-so-good second half.

+-------+-----+-------+------------+------------+
| PITCH | %   | SPEED | MOVEMENT_X | MOVEMENT_Z |
+-------+-----+-------+------------+------------+
|    FB | 62% |  93.3 |      -4.85 |      10.43 |
|    CB | 33% |  80.6 |       3.15 |      -2.76 |
|    CU |  5% |  87.1 |      -6.29 |       6.76 |
+-------+-----+-------+------------+------------+

Sheets tends to get away with throwing (primarily) just two pitches—rather rare for a starting pitcher. He throws his change-up (a solid pitch) just five percent of the time and relies heavily on his fastball. The fastball, strangely enough, gets less horizontal movement than a league average fastball. Sheets makes up a little for it with good speed (93 MPH) and great rise (a.k.a. vertical movement)—nearly 10-and-a-half inches.

It's harder to make a definitive judgment about the curve. The averages for it are quite unimpressive (if we only looked at the averages, it would have fallen under the "crappy" classification I talked about in my curveball article), but there is a wide range of variability. Some actually break in on a right-hander, while others break away as far as 10 or 11 inches (these would fall under the power "slurvy" curve classification — similar to what Francisco Rodriguez throws — which I found to be most effective).

Is this intentional? Can he control how much curve each particular pitch gets? My guess would be yes, at least to some extent.

It seems to me that it would be nigh impossible to throw a pitch a full third of the time, have that much variability, and still post a 2.0 BB/9. Looking at the spin direction and spin rate of his curves (feel free to contact me if you'd like to see the exact data), I also see pretty wide variability. This article indicates that Sheets changes his grip on his curve, but that was all I was able to dig up. If you've found something else that verifies this, shoot me an e-mail.

I can't say for sure, but I'd guess that Sheets controls how much spin he puts on his curve depending on the situation. If he needs a "get me over"-type breaking ball, he imparts less. If he's going for the kill, he imparts more. If this is the case, I can definitely see why people rave about the pitch. It's essentially like he's using two different breaking balls. One is a filthy slurve/11-to-5 combo curve, while the other comes in looking like the former, deceiving the batter by failing to break as much.

Unfortunately, we don't have data from before 2007, so we have nothing from Sheets' glory days to compare all this to. Once I get a hold of the full season's worth of data, however, I'll be sure to check all of this plus some of Sheets' non-physical skills (i.e. location, usage by situation, etc) from the first half against the second to see if he changed anything. Hopefully this was interesting enough for now.

There was a great article about Sheets written by David Gassko (who was aided by our Chris Neault) back in May. Definitely a great read if you're interested in more about Sheets' PITCHf/x data, injury history, and mechanics.

Market value

Not a lot of rankings are out yet, so for now we'll have to make due with what's available. As the off-season goes on, I'll be adding additional rankings to the list to give us a more complete picture of how our competition views each player.

RotoHog Value: 12th SP
Yahoo! Big Board: 21st SP
ProTrade Value: 28th SP
CBS Sportsline: 31st SP
Mock Draft #1: 40th SP (R13)

We're definitely looking at a small sample and some year end data that may not actually be measuring what we're looking for, but it appears there are widely differing opinions on Sheets. Even at #40, though, he went in the 13th round of my first mock draft. Mike Podhorzer of the Fantasy Baseball Generals took him in the tenth round of his first mock draft.

Overall, I'd say there aren't many sound conclusions we can draw about his market value yet. It's just too all over the place, and with the prior caveats, I'm going to decline to make any real guesses about where it will ultimately end up.

Concluding thoughts

Overall, I doubt I'll be drafting Sheets this year. I've never owned him before, and his decline in skills (combined with the great 2008 ERA) and the constant injury risk makes him someone I doubt will fall far enough for me to take. If you feel otherwise, feel free to comment or send me an e-mail. Discussion is always welcome.

Posted by Derek Carty at 12:01am

Saturday, October 11, 2008

Not just elite closers

It's been well-documented that you should avoid drafting elite closers in drafts. Well, when you do not get the elite ones, you are left with drafting the even riskier closers and the cream of the set-up men to fill the relief pitcher spots. There is nothing wrong with that... it is the strategy I advocate. But how reliable are the elite set-up men? Is it worth even a 17th-round selection to draft one? These are some of the questions I explore in this article.

This year

The set-up men drafted the highest in 2008 drafts are as follows, with the average draft position (ADP) after the dash:
{exp:list_maker}Carlos Marmol—181
Rafael Betancourt—203
Jonathan Broxton—211
Hideki Okajima—212
Heath Bell—216
Bob Howry—219
Pat Neshek—226
Scot Shields—229 {/exp:list_maker}
Now let's take a look at the top eight set-up men at season's end, sorted by ERA:
{exp:list_maker}Joey Devine—.59(!)
Brad Ziegler>—1.06
Grant Balfour—1.54
Jose Arredondo—1.62
Hong-Chih Kuo—1.69
Scott Downs—1.78
Craig Breslow—1.91
Joe Nelson—2.00 {/exp:list_maker}
Not one name on the first list made it onto the second list. Very interesting. So, of the middle relievers we thought would have the best seasons, none ended up in the top eight in ERA. I'm not saying that ERA is the only or even the best way to judge performance>—certainly other factors like strikeouts and possibly holds should be considered>—but for fantasy purposes, ERA is basically what you're going for. Let's see what how the ERAs of the players from the first list ended up:
{exp:list_maker}Marmol—2.68
Betancourt—5.07
Broxton—3.13
Okajima—2.61
Bell—3.58
Howry—5.25
Neshek—4.73
Shields—2.70 {/exp:list_maker}
Six of the eight had solid seasons, with Betancourt and Howry posting ERAs above 5.00. The rest of the ERAs fell between the mid-2s and mid-3s, which puts them a whole point of ERA behind the league leaders.

Although I have the feeling that this trend will continue if we look at the numbers from past years, let's examine them anyway and confirm my inclination. Problem is, I do not know where ADP numbers are for past seasons. (If anyone does, let me know!). I'll probably download and save this year's data, so I'll have a solid collection of past ADPs in the future, but that doesn't help me now. We will have to use a different "stat" to determine who the best middle relievers were in past seasons.

I know if I use ERA, I will get yelled at by many readers, so that won't work. I'll instead use FIP, which attempts to factor only the things a pitcher can control, like strikeouts and walks, and ignores things pitchers cannot control, like defense and luck. Many people consider it a great predictor of future ERA—better than ERA itself—so FIP it is!

Past years

We've already compared 2007 to 2008, so let's compare relievers from two more sets of years: 2005 to 2006, and 2006 to 2007, using FIP as the perceived value and ERA for the actual results.

Here are the top middle relievers of 2005 in terms of FIP, the top ERA relievers of 2006, and finally what the original pitchers did in 2006:

Player As            05 FIP  Player Bs           06 ERA Player As           06 ERA
Rudy Seanez           2.15   Dennys Reyes         0.89  Rudy Seanez        4.92
Aaron Heilman         2.35   Cla Meridith         1.05  Aaron Heilman      3.62
Juan Rincon           2.46   Joel Zumaya          1.94  Juan Rincon        2.91
Arthur Rhodes         2.54   Pedro Feliciano      2.09  Arthur Rhodes      5.32
Joaquin Benoit        2.54   Rafael Soriano       2.25  Joaquin Benoit     4.86
Rafael Betancourt     2.58   Rheal Cormier        2.44  Rafael Betancou    3.81
Mike Timlin           2.70   Dan Wheeler          2.52  Mike Timlin        4.36
Jose Valverde         2.73   Scott Cassidy        2.53  Jose Valverde      5.84
Kyle Farnsworth       2.75   Brandon League       2.53  Kyle Farnsworth    4.36
Scott Linebrink       2.76   Jonathan Broxton     2.59  Scott Linebrink    3.57

Much of the same for 2005 and 2006. None of the players on list A found their way onto list B, with many if the players on list A imploding in 2006 with ERAs from the mid-4s and up. Middle relievers with ERAs above 4 will find themselves on the free agent wire pretty quickly, unless the general sentiment is they have been getting unlucky. And I guarantee no one thought Mike Timlin's 4.36 ERA was the result of bad luck.

Now let's look at the same type of table for the 2006 and 2007 seasons:

Player As           06 FIP      Player Bs      07 ERA Player As                07 ERA
Chad Bradford        2.53   Carlos Marmol       1.43  Chad Bradford          3.34
Justin Duchscherer   2.79   Rafael Betancourt   1.47  Justin Duchscherer     4.96
Juan Rincon          2.84   Rafael Perez        1.73  Juan Rincon            5.13
Dennys Reyes         2.87   Peter Moylan        1.80  Dennys Reyes           3.99
Cla Merideth         2.93   J.C. Romero         1.92  Cla Merideth           3.50
Jason Davis          2.97   Lee Gardner         1.94  Jason Davis            5.84
Kiko Calero          2.97   Heath Bell          2.02  Kiko Calero            5.75
Alan Embree          2.97   Manny Delcarmen     2.05  Alan Embree            3.97
Manny Delcarmen      3.02   Manny Corpas        2.08  Manny Delcarmen        2.05
Jonathan Broxton     3.13   Scott Downs         2.17  Jonathan Broxton       2.85

Finally! A player on the first list made his way onto the second. That player is Manny Delcarmen, who finished 2006 with a 5.06 ERA, and yet a 3.02 FIP. Why such a discrepancy? Delcarmen had an incredibly unlucky 2006 season, registering a .385 BABIP and 65 percent left-on-base percentage. In reality, Delcarmen had the skills of a 3.00 ERA pitcher, not a 5.00 ERA pitcher, evidenced mostly by his solid strikeout and walk ratios.

Now, a 3.00 ERA is very different from the 2.05 ERA he boasted in 2007. Just chalk that up to over-regression; one unlucky season followed a lucky one. Besides Delcarmen, it was much of the same, with a couple of last year's best relievers posting solid ERAs while most of the others saw their ERAs skyrocket upwards.

I should point out that FIP is not the best statistic for determining perceived value, especially for fantasy baseball purposes. A big reason a person would draft a middle reliever is because the reliever is behind a shaky closer on a short leash, and that is not considered in the FIP formula. Regardless, we can conclude that best set-up men of one year often are not the best the next year. So this brings me back to my original question: Is it worth drafting elite set-up men?

My advice

The biggest reason to continue drafting them is that it costs, what, a 17th-round pick? And that's for the most expensive ones. Most players you would draft otherwise in these rounds are either low-upside role players or high-risk, high-reward types who often don't pan out. Plus, the low WHIP and ERA of a top middle reliever can be surprisingly valuable in roto leagues where these are often the categories with the least diversion. With an elite set-up man, you'll get consistent production with the chance of getting a save or win here and there.

Or will you?

I'm pretty sure we just established that drafting set-up men can be somewhat of a crap shoot. So ignore my sarcasm in the last paragraph. Here is the plan I advise you to follow: If you are going to select a set-up man, weigh more heavily his chance of becoming the closer than his actual pitching abilities. This does not mean that pitching ability should not enter into the equation. If the incumbent closer does get dethroned, then the low-ability reliever probably won't be selected as the next closer anyway. And even if he is, chances are he will not last long.

The most valuable set-up men are those like Joey Devine, Manny Corpas and Carlos Marmol, all of whom have both the skills to be a closer and a solid chance of becoming one as well. A guy like Hideki Okajima, who did own a 2.61 ERA with 60 strikeouts in 62 innings, is not as valuable pitching before one of the most entrenched closers, Jonathan Papelbon.

I've ended an article with it before and I'll say it again: Drafting in the late rounds is all about upside. Nothing else, just upside.

If you feel like using your 17th or 18th or whatever round pick on a set-up guy with little upside above being a solid ERA, WHIP and strikeout guy, go ahead. But just be aware that players like Aubrey Huff and Mark DeRosa were drafted on average in the 17th round of drafts this year. Although picks like that do not usually prove rewarding, those rare times they do make up for all the failures in between. And if that means avoiding elite setup men along with the elite closers you originally are avoiding, know that surprise relievers like Grant Balfour appear every year and can be picked up for nothing. There is just as good a chance Troy Percival will get injured as any other closer.

Posted by Paul Singman at 1:03am

Monday, October 13, 2008

The problem with measuring forecasting accuracy

David Gassko recently published an article on how various projection systems performed. I am not a big fan of these types of articles. This is not an attack on David or other forecasters out there. They do great work, and there is no way I could create a projection system that matches what any of the forecasters do. Rather, this should be taken as a general warning about the information we can receive from various projections.

The first step in these kinds of articles is to choose a playing time cutoff. For example, David chose a cutoff of 200 plate appearances to compare player projections to their actual results. Note that we have no way of knowing, before the season starts, how many plate appearances he'll have. A player may get hurt, or he might perform so poorly that he doesn't get the opportunity to reach the plate appearance cutoff.

This means we don't receive certain information from these players—some of the most valuable information fantasy players seek. Most projection systems match up comparably when it comes to forecasting a player's rate stats such as on base percentage or OPS. However, how useful is this if we're not sure what the chances are of a player reaching a certain amount of plate appearances?

So what information are we receiving? We learn how accurate a forecasting system is if a player plays for a certain amount of time. But what kind of players will reach that plateau? It will include players who stay healthy for at least part of the season. It will include players whom managers saw as "everyday" players before the season started. It will also include players who may not have been considered "everyday" players but performed well enough to force the manager's hand.

What does this all mean for fantasy players and GMs? If we consider these results as a measure of accuracy, the projections that we use in general will be too optimistic. We will not know what the chances are that a player falls off the map. We will not know the chances that a player performs so badly that he doesn't reach a certain playing time cutoff. We will know in general how accurate a projection is if a player plays long enough. However, we don't have a projection for the chances a player has of actually reaching that requirement!

This is what I see as one of the biggest problems in forecasting. Forecasting systems are pretty accurate right now in projecting a certain subset of players. However, it is the other subset of players, those who provide basically no value, that can cripple a fantasy owner. We can't really project what the chances are that a player will fall in a certain subset. So consider this as a word of caution as you begin preparing for your next draft or auction while using those projection systems as a guide.

Posted by Victor Wang at 1:02am

This is Page 1 of 4 THT Fantasy Focus pages 1 2 3 > Last »

Now shipping

Search THT:

THT's Toolbox

Wednesday, October 01, 2008

Background

Qualitative Risk Assessment

Quantitative Risk Assessment

Overall Risk Assessment

Thursday, October 02, 2008

Picks 1 to 3

Pick 4

Picks 5 to 11

Picks 12 to 15

Picks 16 to 20

Picks 21 to 23

Thoughts

Closer strategy

Concluding thoughts

Friday, October 03, 2008

Early aggression

Midseason

Trade deadline

Concluding thoughts

Saturday, October 04, 2008

Late season points in droves

Final standings

Keys to success

Concluding thoughts

Monday, October 06, 2008

Background

Qualitative Risk Assessment

Quantitative Risk Assessment

Overall Risk Assessment

Tuesday, October 07, 2008

Path to the majors

In the majors

Outlook

Wednesday, October 08, 2008

Surprise Picks

Thoughts on mock drafting in general

Concluding thoughts

Thursday, October 09, 2008

Numbers

PITCHf/x

Market value

Concluding thoughts

Saturday, October 11, 2008

This year

Past years

My advice

Monday, October 13, 2008