Baseball By Paul

It has taken me awhile, but I had some other writing commitments to fulfill before I could sit down and hammer this out. A few weeks ago, I covered the best hitters for the 2nd half over the past three seasons. That group of players has shown a penchant for performing very well after the All-Star Break. Whether or not some sort of switch comes on for them or if it’s anything more than a mere coincidence is up for debate. I feel that using a three-year sample does show that at least something is going on that should be noticed. Some players can’t stand the cold and need the blistering heat of July and August to get their bat going.

Conversely, there is a group of players that don’t fare nearly as well after the break. Fatigue could be a big reason for this ineptitude. Given the way I gathered the data here, it could just be that they aren’t very good players in the first place. I pulled the sub-.750 OPS players with 200+ at-bats after the break from 2005-2007 and then isolated the players that appeared more than once. The first set of players are the ones that appeared all three times.

There isn’t much Earth-moving information contained in the above table. You have a catcher, several middle infielders and Pedro Feliz as the three-time offenders. All of them fare worse after the break than they do prior to it, but not a single one is a prime fantasy contributor to begin with, so while they might be in your lineup depending on league settings they aren’t part of a foundation.

The group of 25 players that follows appeared twice in the three-year span studied and within this list you will find some star power. The bulk of the list is still marginal mixed league talent and role playing single league players, but it isn’t completely useless to have a beat on who typically fades down the stretch.

Notice immediately two of this year’s best players in Justin Morneau and Kevin Youkilis. Thus far, both are bucking their trend handily. Morneau has a .255 average since the break, but a power surge has led to a .979 OPS with four home runs and 19 RBIs. Youkilis has had a similar power output with five home runs and a .961 OPS since the break. It’s too early to say they have eliminated that trend from their profile as there are still two months of play left in the season.

There are a lot of speedsters in this list and that is more a product of the fact that they don’t post very high OPS figures in general because they aren’t very powerful. That’s why I included the numbers before the break as a backdrop. I wouldn’t want someone to just see the names and think that the player is a guaranteed sell. Jose Reyes for example doesn’t experience a precipitous fall in production from one half to the next. Since they only appeared in two of the three seasons studied, that third season could have been a huge year that has their composite OPS over the .750-mark.

I would like to look at the biggest drop-offs for the second half on a percentage basis as opposed to just the players that are below a particular figure. I think a project like that will take much longer and might something suited for the off-season with the 2006-2008 data. We shall see.

Fantasy owners (and baseball fans in general for that matter) make a lot decisions and engage in a host of arguments based on different sized samples of data. The first thing that someone disagreeing with an assertion will generally look at is the sample size. If you tout an unknown commodity after a hot week, you’re likely to get hammered for putting far too much stock into a 7-day period. Sometimes when we’re on the side that wants to be right, we’ll ignore obvious sample size issues in hopes of not being called out.

What is a big enough sample size? Honestly, that depends on what you’re asserting or trying to prove. If you look back at a .417/.489/.590 week of a player you chose over the .231/.300/.378 player for your head-to-head league match, then you can pound your chest a little bit because you clearly made the right decision. But if the latter player is clearly the better player and you let that week make you chose player A again then you start getting yourself into trouble.

Does a month offer enough data to start making judgments on players? Again, it’s going to depend at what you’re hoping to accomplish. At the beginning of the seasons, fantasy team owners across the world are hypersensitive to the highs and lows that their players (and even those not on their team) are going through. Entering May 1st, Emil Brown had 25 RBIs. He has 49 through Wednesday. Josh Willingham was hitting .341, but he has .167 since (missed all of May) to leave him at .260 on the season.

Most are familiar with the awful start of C.C. Sabathia, but it bears mentioning that he was 1-4 with a 7.88 ERA. After a complete game shutout of the Cardinals tonight, he is 10-8 with a 3.30 ERA. How many watched as the owner of Sabathia in their league dealt him for a quarter of his true value because it wasn’t just a bad start or two, but rather a month? Furthermore, how many watch Mark Hendrickson get picked up because of a 4-1 start with a 3.68 ERA for the Marlins? You start looking for reasons, however flimsy, to believe these things to be true. For someone like Hendrickson, it was the previous track record of Florida arms and the home stadium that all of sudden made him worthwhile. Sabathia was out of shape and his workload had caught up to him.

If a player has an established track record (of success or failure), then you shouldn’t begin to take numbers seriously until June 1st rolls around if they are going against that previously established record. Even at that point, you should be cautious of making hasty moves that could end up hurting you in the long run. Remember that on June 1st, there are still FOUR months in a fantasy baseball season. I’m not suggesting you have to stand pat until two months or more into the season, rather that it can be foolish that make moves because of numbers posted in the first 2-2 1/2 months into the season if they are inexplicably against years of unlike data. I grant that there are variables to this line of thinking that can come into play. Injuries, age, team situation (lineup/ballpark/playing time) are primary amongst these factors.

I have spent the last 500+ words telling you which time periods aren’t long enough to make, so what time period of statistics do I find suitable enough to make judgments off of for players who performing against career norms? I think a year’s worth of data is pretty solid. Meaning if a guy has a great 1st half, how is he looking against his peers when it is combined with his previous season’s second half? Believe it or not, a half season of data—no matter how good or bad—can be a fluke especially when it’s complete contrary to a player’s history that spans much more than a half season.

I’m going to post some of the 365-day Leaderboards so we can get a better idea of who has been performing the best over the past “season” worth of games. Plenty of this year’s top performers will be found within these lists, but others not currently atop the league leaders for 2008 will be scattered across these lists as well. This exercise isn’t meant to be a predictive measure so much as it is a better look who has been performing the best for a more sustained period of time than just the nearly four months of the 2008 season.

This set of lists covers the Top 25 across the main roto categories of home runs, runs batted in, stolen bases and runs scored. I took OPS instead of batting average because I (as many do) find it a better measure of a player’s value. I realize average or even on-base percentage are used in fantasy scoring, but went with the OPS anyway.

Again, I’m not looking at these to be predictive of the final two months of the season, instead just another set of data to chew on when judging talent. Most of the names found within the above listings have been getting it done for a full season’s worth of plate appearances and it wouldn’t be a horrible idea to rely on them to get it done for you during the final two months of the season.

I leave for Philadelphia tomorrow and I’ll be gone until Sunday evening. While I hope to have some stuff up during that time period, I’m unsure if I’ll have time for much of anything. When I get back, I’ll most assuredly report on my trip to Citizen’s Bank Ballpark as the Phillies play host to the Atlanta Braves. I think we’re down for at least Saturday’s matchup which features Cole Hamels against Jo-Jo Reyes. We might go to the Kyle Kendrick-Jair Jurrjens showdown on Friday evening as well, but that is TBD at this point.

I got an email shortly after posting the Second Half hitter pieces asking if I had plans for anything on pitchers. I compiled the data to do a study on pitchers after the break, but the results were very light. I took the same three year data set of 2005-2007 and looked at the pitchers with 80+ innings and an ERA of 3.50 and below. The results were very underwhelming as only 63 results came back for all three seasons combined.

I could’ve loosened the ERA threshold, but even 80 innings of a 3.50 from a starter isn’t going to move the needle of a fantasy team’s ERA very much at all. I also could’ve lowered the innings threshold to 60+ or 40+, but 60 only brought in few more samples and 40 started to bring in relievers (and again, a 40-inning sample won’t dent an ERA).

There were 14 pitchers that showed up twice (well, John Smoltz qualified all three seasons, but he’s done for 2008 anyway), but these names aren’t going to surprise and Smoltz is one of three that isn’t even available via trade or waiver wire rendering this entire process just about useless. Thankfully it didn’t take terribly long, so I’ll share the results with you all and move on to the next project:

Pitchers w/80+ IP of <3.50 ERA for the 2nd-half twice from 2005-07:
1. Andy Pettitte
2. Brandon Webb
3. CC Sabathia
4. Chris Carpenter
5. Derek Lowe
6. Jake Peavy
7. Jeff Suppan
8. Johan Santana
9. John Lackey
10. John Smoltz
11. Josh Beckett
12. Roger Clemens
13. Scott Kazmir
14. Tim Hudson

Don’t get me wrong, you should probably still target these guys as your hired gun for the duration of the season, but the point of the exercise was to highlight a group of players to target that might turn up some unknown jewels given the narrowed data set. This list would be targeted regardless of the split analyzed.

For those of you that know it from the message board, you’ve seen previous volumes of the Middle Reliever Hotsheet, but for many others this may be your first look. It has been two months since I last did an update and I don’t have any particular reason behind why I waited so long to do another one, but it’s basically a comprehensive look at the best middle relievers in baseball for the current season. Before the season, I covered the Middle Reliever Methodology, which is really just the utilization of middle relievers in various quantities to counter stockpiling mediocre or below average starting pitchers just because they are in the rotation.

Determining who the best will be from year to year has proven difficult and nearly as volatile as their parent subset of closers. You needn’t look any further than Rafael Betancourt‘s disastrous 2008 season for proof of this volatility. However, some of the success stories from last year, specifically Heath Bell, have proven to be solid bets once again this season. Anyway, I digress. I put together the MRH to update the middle reliever landscape so that those using the method will know where to look for their next diamond in the rough. Oftentimes players are held onto because of name recognition when better options exist (see: Broxton, Jonathan & Marmol, Carlos). But even when those established middle reliever names are performing, there still may be equal or better options available at a hugely discounted rate and that’s really the benefit of this strategy: top-flight production at a replacement-level cost. Bell & Hideki Okajima have paid dividends for their owners this season, but they were two of the hottest targets with Betancourt, whereas a Kyle McClellan or Matt Thornton was (or still could be) had for nothing as a waiver-wire pickup.

Volume IV turns in the lowest count of “Top Non-Closers” since I started doing this back in April with nine. I’ll show those nine in the charts that follow this writing, but for now I am going to look at some relievers that don’t meet some of the requirements for middle reliever stardom, but have skills that point to potential excellence going forward from now until the end of the season:

(numbered, but in no particular order)

1. Joel Peralta, Kansas City Royals – Peralta has seven strikeouts per nine and just 1.3 walks per nine in his 28 and 2/3 innings pitched for a Royals bullpen with some solid parts in it. OK, what’s the catch? A 5.65 ERA is enough to get Peralta overlooked in the search for a reliable middle reliever. His penchant for the blowup has done in the 32-year old. He has given up two runs in seven of his 23 appearances and that’s just not going to cut it. The quality peripherals point to talent, but perhaps he lacks that mental edge necessary in relieving.

2. Manny Delcarmen, Boston Red Sox – On May 27th, his ERA was 5.14. A month later, his June 27th clean inning had lowered the ERA to 3.27, but then he allowed six runs in his next three appearances spanning an inning and a third bringing the ERA back up to 4.72. That’s the long of saying that he is like Peralta in his gas can-like tendencies that balloon an ERA in a moment’s notice. His strikeout rate is at a healthy one per inning, but the walk rate is a tad high at 3.4.

3. Sean Green, Seattle Mariners – Usually the attention goes to J.J. Putz when you’re talking about the Mariners’ bullpen, but he’s been rendered ineffective due to injuries this season which left the door open. Unfortunately for Green, Brandon Morrow and 12.5 strikeouts per nine innings walked in and snapped up the available attention. That doesn’t mean that Green hasn’t been awesome. In fact, he has a 2.72 ERA despite the fact that he was dumped for nine runs in 13 innings during May (6.23 ERA). Green’s 53 innings are tied for 6th among relievers making his tiny ERA, solid 1.21 WHIP and quality 8.2 K/9 rate that much better.

4. Chad Durbin, Philadelphia Phillies – Until Brett Myers was recently sent down to the minors, the Phillies had the same starting rotation all season long. This gave their bullpen a steadiness in roles that has no doubt contributed to their success. Durbin has been the link from those starters to the sweet back-end of the bullpen headlined by Brad Lidge, Ryan Madson and J.C. Romero. He has the most innings of any non-starter on the Phillies, but given his previous time as a starter the Phillies know they can stretch him out a little bit. He hasn’t disappointed with a 1.89 ERA and 7.4 K/9 in his 52.1 innings.

5. J.P. Howell, Tampa Bay Rays – Though prone to a walk (4.3 BB/9), it is due more to stinginess than a lack of control (6.0 H/9). This stinginess has suited Howell quite well as he’s become a true middle relieving ace for one of baseball’s best teams in 2008. He strikes out nearly a batter per inning (8.7 K/9) and has yielded a meager 2.68 ERA in his 57 innings (3rd-most amongst relievers). He is the least likely of these five to be available in your league, but he is the perfect anchor for a team using the MR Method.

To the charts we go:

You should aim to land one of the “Nifty Nine” found on the right if you’re employing the MR Methodology on your team. They represent the middle relievers with best, most complete skills to date. And given that we’re at the midpoint of July, sample size issues are eliminated from the equation meaning these are rather sturdy bets for at least the balance of the ’08 season.

The following two lists present the top skills in power and control regardless of anything else. These are just your premier flamethrowers and pinpoint control artists absent filters of innings pitched except that they reside in the top 200 relievers in terms of workload.

Finally, we filter down the top strikeout and walk rates based on workload. Using the top 200 innings-loggers, we get an average of 36 and a 1/3 innings which is what I used as the requirement to make either of these lists.

And finally, a look at how some combos of middle relievers have performed against some of the best starting pitchers in baseball:

Yesterday I posted Part I of a two-part piece on some historically strong performers in the 2nd half. For the study, I looked at post All-Star Break numbers from the years 2005 to 2007 and picked out the instances of players posting an .850 OPS or better. From there, I took the players that did it more than once and isolated them. The group that pulled the hat-trick were in yesterday’s list, while today will cover the remaining bunch that did it two out of three times. In the upcoming chart, you will see the three-year totals and averages for the 43 players that had an .850 or better OPS in two out of three seasons, meaning the season that they didn’t make it is factored in, too.

The email response to the first portion was great and I really appreciated it, thought some readers point blank asked what they are supposed to do with the information. You don’t have to do anything with it if you don’t want to, but it can be used for reference when entertaining trades for a stretch. It is by no means wholly predictive so don’t blindly follow it to the point of taking a bad trade because you think Player X is going to have a bust out second half. But if you are deciding between a few players, this information might lead towards the right guy to help your team from here until October. The Elite 16 list was filled with stars, but there were plenty of stars not listed meaning maybe you trade some stars you currently own for these ones that have shown a penchant for stepping it up during the dog days of summer.

The second list is decidedly less star-laden meaning you could find some real value by acquiring some of these players at a severely discounted price. Some of the players on the list are already having great seasons (Jason Bay, Milton Bradley, Adrian Gonzalez), but some others have seen their stock tumble based on a horrific first half (Robinson Cano, Matt Diaz, Todd Helton). Baltimore’s Luke Scott is having a breakout season of sorts due in large part to his job security, but if he holds to his usual pattern, he’s in for an even bigger 2nd half which would truly make 2008 a breakout season for him. He has the biggest pre to post All-Star Break delta in OPS at 261 points! Speaking of opportunity as a breeding ground for breakout potential, second to Scott in OPS improvement is recently traded Matt Murton. He should get a chance in Oakland to repeat his 194-point OPS improvements after the break especially given the fragility of Ryan Sweeney and suckiness of Emil Brown & Travis Buck. He’s batted second in back-to-back evenings since being recalled on July 11th.

Some other potential gems exist in the form of Carlos Delgado, Travis Hafner, Paul Konerko and Adam LaRoche. Althought Delgado and LaRoche have both been strong in July, so the window to acquire them cheaply might be closed.

Feel free to share your tales of trading for some of these guys as Steve from Iowa did via email. Steve sold high on BbP favorite, Justin Ducscherer in a deal that landed him Matt Holliday. Steve dealt Duchscherer and Bill Hall for Holliday. Like I said, Duchscherer is a huge favorite of Baseball by Paul, but this is an amazing deal for Steve.

I’ve never done crack, but I imagine that the withdrawals from quitting the stuff are similar to what I suffer through during the All Star Break. While likely not nearly as dramatic, I suspect many others search for ways to keep themselves engaged, but occupied. This naturally leads back to your league’s homepage in search of ways to shore up the weaknesses for the stretch run. The dust has settled on the “first half” (technically the 81-game mark hit weeks ago) and every fantasy owner should know full well what they need to get to their desired result.

Is your team in need of offense? I have compiled a group of players that have shown a propensity for second half excellence in the past three seasons by registering an .850 or better OPS in at least two of the three seasons. Of the 59 players that qualified, 16 of them completed the hat-trick and the names within that list are unlikely to surprise many. In fact, the entire list is rather star-laden but that doesn’t render it useless. We know that stars are supposed to perform, but if you’re acquiring the high-dollar players from a team dumping and looking at 2009, this list will guide you towards the players you are likely to get the most gains from during the last 2 1/2 months of the season.

I will start with the upper echelon of players. As I mentioned, this group gave .850 or better in each of the three seasons studied (2005-2007). In fact, the lowest three-year average in OPS is Jeff Kent‘s .911. By and large, this group of players will represent the guys you ought to be targeting for the second-half surge. Some of them are in the midst of otherworldly seasons already so a continuation of their post-All Star Break mashing could lead to some record-breaking totals.

Here is a detailed look at the group sorted by OPS. The far right column is their difference in OPS from before to after the All Star Break. The 3-year average column directly below each player is their total line divided by three… pretty self-explanatory stuff there!

Seeing David Ortiz & Ryan Howard top the list should please owners of these guys quite a bit. For Ortiz, it was a painfully slow start followed by an injury currently shelving him that his derailed his season while Howard had delivered little more than raw power until the beginning of July. Even at his worst (like a .234 batting average), Howard’s power production makes him rosterable in just about any format. He is already turning things around in July with both the power (seven home runs in 10 games) AND the batting average (.375, 15-for-40). The time to buy low has probably vanished by now, but he should still be a superstar target, especially if he can be had at ANY amount of discount.

The usual suspects round this list though Kent has been rather “blah” so far this season. The Dodgers could really, really use a hot summer from him considering their lineup’s immense struggles. If he achieves his average output from the past three seasons, he will be very close to last year’s final numbers and that’s pretty darn good for a 40-year old second baseman.

Moises Alou might be done forever with his latest injury likely sidelining him for all of 2008. He is (was?) a truly amazing hitter that would get the job done whenever he was the lineup. Unfortunately, at age 42, his body is just not holding up anymore. I decided to leave him on the list essentially to pay homage to his greatness because I knew he wasn’t going to be a viable target for fantasy owners in the 2nd half of 2008.

Tomorrow I’ll post the remaining 43 players on the list. It is the group of players that reached the .850 OPS or better mark two out of the three seasons I looked at for this study. That list contains top-flight superstars, mid-level semistars, average players and somewhat scrubby players that you might be able to pick up off the heap and speculate on to see if they continue their post-All Star Break hitting trend.