A few weeks ago I toyed around with a metric for comparing bullpen usage between teams. In that same post I highlighted a couple of relievers by plotting their pLI by appearance. Jeff Zimmerman took those second set of graphs and created a within-team usage comparison pair of charts (with some input from Tango) at Royal’s Review. Since then I’ve been pondering ways to quickly and graphically compare usage between teams. For a first attempt I narrowed the scope to just closer usage. My methodology was to bin each gmLI into one of four bins as specified by the below table

I then simply counted up the instances in each bin and charted the results. Here is the graph for the AL East, first with just raw totals

>

And then broken down by percentage

Read the rest of this entry »

As Eno pointed out last week over on the RotoGraphs side, Trevor Hoffman is struggling right now. He’s currently sporting a FIP of 10.50, XFIP of 6.62, and a WPA of -1.54. Is there anything in the data that sheds a little light on the problem? Perusing the data there are some things that stand out, notably pitch usage, contact stats, and an extreme fly ball tendency.

As Eno already noted, Hoffman is using his changeup much less this year than years past (21% compared to 29% career average) with his fastball picking up the some of difference (66% to 63% career average). In order to dig a little deeper I looked at Hoffman’s page in Texas Leaguers’ pitch F/X database and ran some splits by year on pitch utilization. The largest discrepancy was, not surprisingly, the changeup. The surprising part to me was that the drop from last year to this year was larger against RHB (32% in 2009 to 10% in 2010) than it was against LHB. This could be a key piece of information as most of the HR’s Hoffman has given up have been on fastballs to RHB. Could right handed batters just be ignoring the changeup and sitting on the fastball?

Also of interest is that Hoffman has struggled at getting swings and misses so far this season. Looking at Hoffman’s plate discipline stats, we see an abnormally high Contact% (88% compared to career 75% career). Also of particular interest are the results on pitches out of the zone. Hoffman is only getting swings on 20% of pitches out of the zone, and even when he does get swings he only gets swings and misses 22% (compared to 48% for his career) of the time. Turning back to Hoffman’s Texas Leaguers’ page, the culprit again appears to be his changeup, which has gone from getting whiffs 20% of the time in 2009 to only 9% of the time in 2010.

As I said in the intro, up until this point in the season Hoffman has become even more of a fly ball pitcher than normal. He currently leads all qualified relievers in FB% by a large margin with 71.8%, which is offset by an astonishingly low GB% of 12.8%. This is even more problematic when you sport a 21.4% HR/FB ratio as Hoffman currently does. Will the ratio remain that high? Probably not, but even if it comes down to a more respectable number it still will not solve all of his problems as his still high xFIP indicates.

I’ve thrown a bunch of various percents your way this morning, so I should probably summarize. Hoffman’s throwing his bread and butter pitch, his chanegeup, less than in years past. He’s getting fewer swings and misses, again notably on his changeup. He’s been giving up flyballs at a dizzying pace, which has also led to giving up HRs at a dizzying pace. So far this has all added up to disaster.

My goal at the start of the research for this article was to develop a quick methodology (i.e. not digging into individual game logs) to find out how effectively teams were deploying their relievers. My first cut was to take a team’s qualifying relievers ranked by gmLI and compare that to the same set of relievers ranked by CHONE ERA. The simple numerical representation I used was the absolute value of the difference between individual pitcher’s rankings summed over each team. For example a player that was ranked 2nd on his team in gmLI, but 3rd in CHONE ERA would have a difference of 1, so a team that has two players flipped would get a total score of 2. So, according to this metric, who is doing well so far? The Rays, Twins and Phillies all have a score of 2, and the next tier is the Cubs Marlins and Orioles with 4. The bottom list consists of the Indians and White Sox at 16, and the Rangers at 14*. Full list can be found here.

That brings me to observation number two, a couple of guys that have already seen their role (read leverage) shift. Randy Williams of the White Sox is ranked 2nd in gmLI and 6th (out of 7) in CHONE ERA for his bullpen, and it seems like Ozzie has figured this out. Here’s a chart of his LI (in this case pLI as it was easier to scrape from the game logs) by appearance

That, my friends, is being put in the reliever doghouse.

On the flip side is Jason Bulger of the Angels. He ranks 4th in gmLI, but 1st in CHONE ERA. Scioscia also seems to be figuring this out as here’s his chart.

I’ll willingly admit that my quick little metric from above might not be the best way to judge a managers bullpen usage, but what do you guys think? Fans of the teams that were in the top or bottom lists, does your team appearing there make sense? Is it worth looking at this retrospectively (i.e. for 2009)?

*I didn’t adjust for Starter to Reliever conversions, so that likely affects the Rangers ranking.

Last week Dave, in his piece on Real Groundball Rates, and Erik, in a piece on Brad Penny’s first start, referenced Dave Duncan’s ability to get his pitchers to generate a great deal of groundball outs. In fact in Dave’s summary of the Cardinal’s organizational ranking he specifically referenced Duncan’s ability to take pitchers off of the scrap heap and turn them into serviceable MLB starters with his pitch to contact philosophy, and cited it as one of the reasons for the Cardinals recent success. With all of that in mind I thought it would be interesting to investigate how the batted ball data supports these positions, and what is the magnitude of the effect.

My methodology was a simplified version of Tom Tango’s WOWY. I found the difference between each pitchers’ ground ball rates for the two years after he joined the Cardinals and the two years before he joined (if available). Then I found the weighted average (weighted by the lesser of the two sets balls in play) of said differences. I limited the sample to just starting pitchers as that is who people reference when discussing Duncan’s “Magic”. With that in mind here’s the sample

Pitcher                With-Without
Brett Tomko             0.8%
Mark Mulder             2.8%
Jeff Suppan             2.4%
Todd Wellemeyer        -10.5%
Kip Wells               2.2%
Jason Marquis           8.7%
Chris Carpenter         11.9%
Kyle Lohse              6.2%
Joel Pineiro            9.1%

The weighted average of the difference works out to ~4.5%, so Duncan’s pitch to contact reputation is backed up by the data. All but Wellemeyer showed improvement in GB%, even Mulder who was a GB pitcher before coming to St. Louis. Clearly this analysis doesn’t speak to overall quality of pitching (for that see this piece by Kincaid) as groundballs and improving groundball rates aren’t the be all end all; however it has been a key component to the Cardinals success in the Tony LaRussa / Dave Duncan era.

One last off season related post to clear out of the outbox before we get too far into the season. Within its MLB preview SI listed the top performers (using 2009 WAR as the metric) within various 2010 salary bins. As could be guessed, most of the low salary bins (under $1M and $1-2M) had players that were still under team control. While this is interesting, I was curious about the best players signed to deals in those low salary bins, specifically those that signed to one year deals. With that in mind I set about to create a few lists of my own, a lineup for 2009 and a couple of potential scrap heap heroes to keep an eye on in 2010. Without further ado, the scrap heap heroes for 2009

Pos   Player            Money         War
C     Gregg Zaun        $1.5M         1.7
1B    Russell Branyan   $1.4M         2.8
2B    Craig Counsell    $1M           2.8
3B    Juan Uribe        $1M           2.8
SS    Omar Vizquel      $1M           1.1
OF    Gabe Kapler       $1M           1.3
OF    Fernando Tatis    $1.7M         1.7
OF    Jerry Hairston    $2m           1
SP    Carl Pavano       $1.5M         3.7
RP    Arthur Rhodes     $2M           1

Overall my list of qualifiers averaged right around 0.4 WAR and a salary a little over $1M. Since opening day is now in our rear view mirror, let’s stop looking back and look some likely scrap heap heroes for this year. On the position player side Troy Glaus hopes to bounce back from an injury plagued 2009. CHONE projects Glaus to produce 1.7 WAR, and the biggest hurdle to doing that might be staying on the field. Russell Branyan will be bidding for a repeat appearance as CHONE projects him at 0.9 WAR. A couple others to watch are Eric Hinske, who just missed the 2009 list, and Rod Barajas, who could get starter’s playing time with the Mets. On the pitching side the scrap heap is usually populated with injury question marks, and 2010 appears to be no different. A couple of names of note that fall into the injury question mark bin are Erik Bedard and Justin Duchscherer. All in all there are some values out there to be had for your $2M, but don’t expect to hit the jackpot every time.

The Rangers are coming off of a 87 win campaign in which they finished 10 games back of the Angels. This year CHONE has them winning between 85-86 wins depending on the method, the FANS have them at 84, and PECOTA has them at 83. Those win levels are good for 1st in the West, but the margin is never more than 5 games, and mostly 0-2 games.

The infield returns all four starters from 2009 and project to be above average on the whole. Michael Young, whose offense rebounded in 2009 to post his highest wOBA (0.385) and WAR (3.8) since 2005, returns to man third base. He is projected to see a decline from that level and likely post something in the high 2s or low 3s. Chris Davis brings his feeble contact rates back to play first base again, and is the one of the few position players that projects to be below average at 1.5 WAR. Ian Kinsler projects to be the star of the group at 4 WAR as he projects to combine average defense with above-average offense. The most interesting, at least personally, member of the infield is Elvis Andrus. Yesterday Dave mentioned that he was one of the largest discrepancies between FAN projections (4.1 WAR) and CHONE (1.6 WAR). I could see that window of potential performances being very realistic, which would put his mean somewhere in the upper 2s.

The outfield of Josh Hamilton in left, Julio Borbon in center and Nelson Cruz in right all project to be in the neighborhood of 3 WAR. All three project to be average to above average with the glove and similarly above average with the bat. At DH they will see if they can squeeze some more life out of Vladimir Guerrero, with David Murphy providing a decent fallback option if Vlad’s knees spontaneously combust. The catching duo of Jarrod Saltalamacchia and Taylor Teagarden don’t project to be world beaters (1.5 WAR each), but that will get you by at that position.

The starting pitching is a lot like the position players in that none of them project to be stars, but they all project to be pretty solid. The closest to a star quality projection would be Rich Harden who projects to have a FIP in the mid-to-upper 3s. The question with Harden is the same as it always is, health. The de facto “Ace” is Scott Feldman and his cutter. The projection systems weren’t overwhelmed by Feldman’s 18 wins and project him to have a FIP in the 4.50 range. Colby Lewis, back from Japan, is a hard player for the forecasters to handle given his lack of MLB experience combined with a dominant year in Japan, but a FIP in the low 4s seems pretty reasonable. The back of the rotation looks like converted reliever C.J. Wilson and Matt Harrison at least for a little while (check out Matt’s piece on C.J.). The wildcard here is Neftali Feliz who is projected to put up a FIP in the mid 3s as a starter, but may spend some time in the pen.

The bullpen looks to be a strength again with closer Frank Francisco, lefty Darren Oliver, and the aforementioned Feliz all having projected FIPs in the mid 3s.

Add all of this up and you have a very solid team with few weaknesses that appears to have solid depth, so it’s no wonder that a lot of the projection systems have them at the top of their division.

The Brewers are coming off of an 80 win season in which their position players earned the right to pummel their pitchers into a big pile of replacementness. The position players accumulated 26.1 WAR, good for 2nd in the NL, while the pitchers only managed a measly 3 WAR, good for last in the NL. However, this post is supposed to be about their chances in 2010, so let’s not dwell on the past. The Fans and PECOTA have the Brewers at 78 wins and CHONE has them at 81, all of which gets them 2nd or 3rd in the Central with a 15-20% chance of getting into the playoffs. Clearly, they are not eliminated from the race before the season starts, but a decent number of things will have to go their way for a playoff berth.

The Brewers’ everyday lineup is built on two stars, a high upside young guy, and some average-ish filler. This model in not too dissimilar from their division rivals the Cardinals, it just so happens that the Cardinals players are projected better at most of the positions. As an interesting aside, both teams have their two stars projected to amass ~45% of the WAR for their starting eight. Those two stars, Prince Fielder and Ryan Braun, both project to hit well above average, which more than makes up for their below-average defense, and places both in the top 20 of projected WAR for position players.

Alcides Escobar’s mean projection is a slightly above average to above average glove and a slightly below average bat at a premium position, but he probably has the widest distribution of the Brewers’ position players given the lack of data off of which to project. Carlos Gomez is somewhat similar in that he projects to be more glove than bat and probably has a pretty wide range of outcomes considering his relative lack of experience still. The big question about Gomez is how bad the bat will be. CHONE projects a 0.323 wOBA, while most other systems have him hovering in the 0.300 range. At 0.323 he is likely a well above average player, but it would be hard for his defense to be good enough to be anything more than average with a 0.300 wOBA.

Rickie Weeks, if healthy, could surpass his projections as they are all based off of a fairly low playing time assumption. Corey Hart is two years removed from a 4.5 WAR season, and has seen his offense and defense decline. His projections meet in the middle of his last two years and that one good one. The rest of the starting eight, Gregg Zaun and Casey McGehee, project to be below average to sniffing average.

As a whole the Brewers’ staff projects to be better than the 3 WAR they put up last year. Yovani Gallardo is back at the top of the rotation after posting a 3.97 FIP last year and will probably be good for 3-4 wins this year. Randy Wolf comes over from the Dodgers and slides into the number 2 slot after posting numbers similar to Gallardo last year, but with a lower upside for the upcoming season. Next up in the rotation is Doug Davis, who is back for a second go-around with the Brewers and projects to be right around average. The last two spots theoretically should go to David Bush and Manny Parra, but that implies that the Brewers will see Jeff Suppan and his 12.5M as sunk cost.

The bullpen also projects to be better this year, with Trevor Hoffman back slinging changeups in the closer role. They also have solid depth with LaTroy Hawkins, Carlos Villanueva, Mitch Stetter, and Todd Coffey all projecting to have FIPs in the high 3s or low 4s.

Clearly the Brewers have some pieces in place to be contenders, but their chances this year will likely hinge on a Weeks comeback, a big jump for Escobar, and some substantial improvement on the starting pitching side.

Every season some teams spend money on players that are going to be on other team’s rosters. For example, according to Cots Contracts, the Red Sox will be paying $9.25M for Julio Lugo to play for the Cardinals. I thought it would be interesting to see which teams had the most “dead money” on the books for 2010. The list, sorted by % of payroll devoted to “dead money” is as follows (all data from Cots)

Team           Dead Money       % of Payroll
Blue Jays      $16M                23%
Dodgers        $16.6M              16%
Angels         $16.1M              14%
Rangers        $6.8M               10%
Brewers        $8.5M               10%

The rest of the list can be found here.

Outside of the Blue Jays, the teams at the top of the list all expect to be in their respective pennant races for a large portion of the season (and the various projection systems expect it too). However, I will also point out that, at least according to those same projections, the Angels, Dodgers, and Brewers appear to be chasing the leaders by a couple of wins in their divisions and this extra money could help close or erase that gap.

The fact that the dead money leaders are going to be in the mix is not entirely surprising as “dead money” is an indictment of past contractual transgressions and does not necessarily reflect on the current management (see Mariners and Yuniesky Betancourt). Also, one years worth of information is hardly enough to draw conclusions from no matter how the teams fell out on the list, but I still found the list interesting.

Last week I presented American League playoff probabilities based on a simulation I had created (which I came to find is similar to the one produced by xls sports if you want to play around with this yourself. They even have incorporated home field which I have not yet). If you are interested in the details of the simulation, give last weeks post a quick read. Remember, these are based off of standings generated from the FANS projections. Without further ado, let’s take a look at the National League results.

East       Div Win %    WC Win %   Playoff %
Braves           56%          9%         65%
Phillies         23%         11%         34%
Marlins          12%          7%         19%
Mets              7%          4%         11%
Nationals         2%          1%          3%

Central    
Cardinals        63%         6%          69%
Cubs             12%         5%          17%
Brewers          11%         5%          16%
Reds             10%         5%          15%
Astros            3%         1%           4%
Pirates           1%         0%           1%

West      
Rockies          35%        12%          47%
Diamondbacks     31%        13%          44%
Dodgers          18%        10%          28%
Giants            9%         6%          15%
Padres            7%         5%          12%

and again some other useful data points compared to historical data

Division        Avg. Wins      2002-2009 Avg Wins
East              94           95
Central           93           93
West              94           91
Wild Card         91           91

The big surprise here is the Braves/Phillies flip-flopping in the East. It would be interesting to go back through the FANS projections for the two teams and see who the FANS are higher/lower on than other projection systems. In the Central, the Cardinals have the tightest hold on any division in baseball according to the FANS, while the West looks like it will be a dogfight.

Next on the simulation to-do list is to simulate some of the most probable playoff scenarios.

Last week David posted projected standings generated from the FANS projections. I thought it would be interesting to dig a little deeper and investigate what it means for example to have the Yankees projected at 98 wins and the Red Sox at 94. My metric of choice will be playoff probabilities and my method will be simulation.

Before I get to the results of the simulation for the AL, let’s get into the nuts and bolts a little bit. In order to run the simulation needs the season’s schedule and each teams true talent win percentage. The simulation is a simple Monte Carlo that determines the winner of each game using random draws bounced up against log5 based winning percentages. For example, if we want to simulate the outcome of a game between Team A that has a 0.600 true talent win percentage and Team B that has a 0.450 win percentage, we first calculate the probability that A beats B using the log5 equation linked above. That calculation says that Team A should have a 0.647 winning percentage against Team B.

To simulate a game between these teams then, the simulation draws a random number between 0 and 1 and if the number is less than or equal to 0.647 then Team A wins, otherwise Team B wins. This process is repeated for all of the games for the entire season. Run the simulation for 10,000 such seasons and you have your results. Also built into the simulation is some up front uncertainty about the true talent win percentage. Before each of the 10,000 simulated seasons, the true talent win percentages for each team are varied slightly by using a random draw from a normal distribution centered at the input win percentage (which is based off of the projected standings) with a standard deviation of 0.030. For example, some seasons the Yankees will simulate as a 0.605 team, sometimes a 0.600 team and sometimes a 0.610 team. The standard deviation was derived through testing (read trial and error) and some of the comments in this thread at The Book Blog.

Now on to the results, starting with the East

East       Div Win %    WC Win %   Playoff %
Yankees          53%         27%         80%
Red Sox          26%         31%         57%
Rays             20%         28%         48%
Orioles           1%          3%          4%
Blue Jays         0%          0%          0%

Central    
Twins	         38%         1%          39%
White Sox	 24%         1%          25%
Tigers           19%         1%          20%
Indians          13%         1%          14%
Royals            6%         0%           6%

West      
Rangers          38%         2%          40%
Mariners         27%         2%          29%
Athletics        19%         2%          21%
Angels           16%         1%          17%

and finally some other useful data points compared to historical data

Division        Avg. Wins      2002-2009 Avg Wins
East             101           99
Central           89           93
West              90           96
Wild Card         94           96

Overall nothing too shocking. According to the FANS all of the divisions should offer plenty of intrigue be it in the form of a dogfight for the division title with likely no Wild Card safety net (the West and the Central) or the powerhouses taking it to each other all season (the East).

Next up I’ll do the same with the NL.