Sunday, July 28, 2013

Preseason projections revisited--Part 2-pitchers.

Here's the pitcher segment of the PECOTA projections compared with the current numbers.

Once again....I will be using the ERA/FRA/WARP for these numbers. No win/loss records. That is largely team dependent based on run support and other factors and we want to judge these performances based on how the pitcher pitched.

ERA-If you're reading this, you know what that is.

FRA explanation from Baseball Prospectus:

Fair Run Average differs from FIP in a few ways. While FIP is concerned only with what a pitcher is believed to control-typically strikeouts, walks, and home runs, though Prospectus includes hit batsmen in our FIP calculation-Fair Run Average takes things a step further. Pitchers receive credit for good sequencing, thus rewarding pitchers who seem to work out of jams more often than usual. Fair Run Average also considers batted ball distribution, base-out state, and team defensive quality (as measured by Fielding Runs Above Average).

So, without further ado...let's get on it.

Homer Bailey-

Projection: 4.10 ERA/4.46 FRA/2.2 WARP
Actual-      3.77/2.86/3.36/2.8 WARP

Bailey is having his best season, but it's peculiar as well. His underlying advanced stats actually say he's one of the best pitchers in baseball, but for some reason his ERA is higher than it should be. His BABIP is .310, so higher than normal, which means he's been a little hit-unlucky, but not extremely so. By WARP he's actually been the 4th best pitcher in baseball this year, behind only Wainwright, Harvey, and Kershaw. Among pitchers with 100 or more innings pitched, he's 7th in FRA. Basically, he's pitched better than his "traditional" stats indicate, so that makes him a candidate for a big finish to the year if he can continue his current trends. Also, his K/9 innings has jumped up quite a bit from his career averages. It is up to 9.1 this year, which is elite for a starter, and up from 7.3 last season and 7.2 the season before. This is likely due to the fact that oddly, his fastball velocity has went up. For the most part, pitchers lose velocity slowly as they age until it tends to level out in their late 20s or early 30s. A lot of guys throw harder in the minors when they are very young than they do when they mature enough as a pitcher to actually make it to the majors. According to Fangraphs, Bailey is throwing harder this year than he has since 2009, and he sucked in 2009 because he didn't know where it was going. His average fastball velocity in 2013 is 93.9, which is way up from 2012's 92.4 and 2011's 92.2. He did have a shoulder injury in 2011 and 2010, so maybe he's just healthy now and he always had the extra power to the fastball, but it's strange, because for most guys once you lose a mph or two you don't get it back, ever.

Mat Latos-


Pretty close. I don't think Latos gets enough credit from a lot of fans. He's fucking good. According to WARP Latos is currently the 13th best pitcher in baseball this year. The Reds have two of the top 13, and  the only other team that can say that is Detroit, which has 3, Verlander, Sanchez, and Scherzer. How funny is it that in a "down year" for Justin Verlander he's 11th in MLB in WARP? Fuck that guy's good.

Mike Leake-


Man, that ERA sticks out like a sore thumb, doesn't it? A guy who only strikes out 5.5 batters per 9 innings should not have an ERA that low. Fueled by a .271 BABIP in part, but the most striking thing to me is the fact that he's not getting killed by the home runs as much this year. His HR/9 is down from 1.3 the past two seasons to 1 flat this year. That doesn't sound like much, but it does help in the long run, but if that BABIP regressed to the mean Mike Leake could be in for a little regression the last two months. He's having the best year of his young career, but he hasn't pitched as well as the ERA would indicate. He allows more baserunners than either Bailey or Latos but less runs, and that just isn't sustainable. He's had ladyluck on his side. I'm not saying the guy sucks or anything, but he's largely a league average pitcher overperforming his true ability, and baseball is full of statistical anomalys like this. Remember Jack Armstrong? That guy sucked, but he pitched well enough for 3 months once to actually start the damn All Star Game. Good for the Reds Mike Leake is better than Jack Armstrong ever was, but let's not get too jazzed over that pretty ERA, because that's not who Leake is.

Bronson Arroyo-note that these numbers don't include the 7/27 game vs. LA. BP hasn't updated their stats for tonight's games yet. 


In all statistical theories, there are always outliers. Guys who consistently buck the trend of what the advanced stats say. Bronson Arroyo almost always outperforms his underlying advanced numbers for whatever reason. He has his whole career, so his projections are always going to be low, because the system just doesn't factor in guys like him, and there are very few in baseball who can actually sustain "cheating the system" so to speak. Arroyo doesn't strike out enough guys, and he gives up too many home runs. But for whatever reason it works. For Mike Leake, he doesn't have a track record of things like that. Bronson does. If Mike Leake can have an Arroyoish career he will have done quite well. Note, however, that his FRA projection is almost dead on the money. An average BABIP is right around .300. Guys will have some seasons higher, and some seasons lower, but for the most part players tend to have their career BABIP around that level. Arroyo has only had an average or higher BABIP twice in his entire career, which suggests that he is one of the few guys who have consistent control over whether or not balls in play are hits or outs. For more on the BABIP thing (it's called the DIPS theory) read Voros McCracken's groundbreaking article from 2001. I'll link to it here for the newbs who don't have a clue what I'm talking about. It's kind of a must read and a theory that you must grasp if you want to get into sabermetrics, even a little bit. If you think it's a bunch of mularkey then sabermetrics probably aren't for you.

The McCracken article

I'm skipping Johnny Cueto because of injury, and skipping Tony Cingrani due to the fact that there wasn't much to project on him. PECOTA did a projection, but it was based on him spending most of the season in the minors, which isn't happening. He is exceeding his projection that they put out there, largely due to a sky high 10.1 K/9. If he could avoid the walks (3.7 BB/9) he could be a killer. Home run rate is a little too high as well. 

I was going to do a few bullpen guys, but I've been working on this thing for 2 hours, and nobody reads this stuff anyway, so the hell with it. 

Thursday, July 18, 2013

Preseason projections revisited.

From Wikipedia:

PECOTA, an acronym for Player Empirical Comparison and Optimization Test Algorithm, is a sabermetric system for forecasting Major League Baseball player performance. The word is abackronym based on the name of journeyman major league player Bill Pecota, who with a lifetime batting average of .249 is perhaps representative of the typical PECOTA entry. PECOTA was developed by Nate Silver in 2002-2003 and introduced to the public in the book Baseball Prospectus 2003.[1] Baseball Prospectus (BP) has owned PECOTA since 2003; Silver managed PECOTA from 2003 to 2009. He was responsible for the PECOTA projections for the 2003—2009 baseball seasons. Beginning in Spring 2009, BP assumed responsibility for producing the annual forecasts. The first baseball season for which Silver played no role in producing the PECOTA projections was 2010.[2]
One of several widely publicized statistical systems of forecasts of player performance, PECOTA player forecasts are marketed by BP as a fantasy baseball product. Since 2003, annual PECOTA forecasts have been published both in the Baseball Prospectus annual books and, in more detailed form, on the subscription-based website.[3] PECOTA also inspired some analogous projection systems for other professional sports: KUBIAK for the National Football League, SCHOENE[4] for the National Basketball Association, and VUKOTA[5] for the National Hockey League.
PECOTA forecasts a player's performance in all of the major categories used in typical fantasy baseball games; it also forecasts production in advanced sabermetric categories developed by Baseball Prospectus (e.g., VORP and EqA). In addition, PECOTA forecasts several summary diagnostics such as breakout rates, improve rates, and attrition rates, as well as the market values of the players. The logic and methodology underlying PECOTA have been described in several publications, but the detailed formulas are proprietary and have not been shared with the baseball research community.

Now that you know all this....I'm going to revisit the preseason PECOTA projections for several key Reds players and compare the results and see if certain players are going to overachieve, underachieve, or pretty much match their projections.

I will be using the standard "slash" stats (avg/on base/slugging), WARP (wins over replacement player) and certain other cherry picked numbers such as HR. I am not factoring in RBI as RBI is a team dependent statistic based on where you are batting, who is ahead of you, etc. I will also not be looking at pitcher W/L record, but for pitchers I will be looking at ERA, FIP (fielding independent pitching), and WARP. 

Please note-defense and baserunning is also figured in WARP for offensive players.

Joey Votto-projection- .300/.401/.532, 29 HR, 6.0 WARP
Votto at the break-      .318/.434/.506, 15 HR 4.3 WARP

On base is ahead of projections, slugging is slightly down, mostly due to a lack of doubles, HR is slightly below pace but a hot streak could flip that quickly. Joey Votto continues to be one of the best players in baseball, despite what the talking heads in the Cincinnati media would want you to believe. 

Shin-Soo Choo-projection-.276/.368/.462, 21 HR, 19 steals,6 caught stealing, 4.0 WARP
At the break-                     .287/.425/.468, 13 HR, 11 steals, 6 caught stealing, 3.8 WARP

Choo has greatly exceeded expectations, mostly due to his high on base percentage. He could basically be a crappy player the rest of the year and meet his WARP projection. He's been pretty bad defensively, being 4 runs below average, which is almost 0.5 WARP. He's gotten half those runs back with baserunning, and it could be better if not for the 6 caught stealing already. 

Jay Bruce-projection-.255/.326/.481, 28 HR, 2.9 WARP
                                .277/.325/.507, 19 HR, 2.7 WARP

I don't have much to add to this. Bruce has consistently been an above average player but not a superstar, and that is what he continues to be. If he could turn that on base percentage into the .350 or .360 range he would probably be an MVP candidate, but that's probably not going to happen at this point. He's 26 now, he is who he is, and he's pretty good.

Brandon Phillips-projection-.266/.314/.425, 20 HR, 2.4 WARP
                                           .266/.320/.413, 12 HR, 1.6 WARP

Projections are pretty much right on the fucking money.

Some more guys...but I don't want to type all this out:

Todd Frazier is right in his projection ranges, except for slugging and HR. He's been quite good defensively as well. 
Everyone's favorite whipping boy, Zack Cozart, has been a little bit under projection, but not super under. He's not that good of a player at his best, so I don't know what people are expecting. He's a slightly less than average shortstop (because there aren't that many great shortstops right now) and as Dennis Green would say:
Ryan Hanigan has been bloody awful, and way below projections, but I wonder if injuries are playing a part, or if he's just gotten old. Same thing with Chris Heisey, although he's not old.

I'll do pitchers on my next post. 

 Is anybody even reading these things? Leave a comment if you like this stuff I'm writing. Even if no one is reading it I find it fun to look this stuff up and analyse it, so I'll keep doing it regardless.

Monday, July 1, 2013

Is Brandon Phillips REALLY having a great season because he has a shitload of RBI?

Baseball Prospectus has a stat buried on their stat pages called OBI%, which is, simply, the percentage of batters on base that a batter has driven in.

Without looking, I thought that BP was likely just having his typical production, with a inflated RBI total due to the high OBPs of Choo and Votto. What I found is, well, I'm right, but I'm also wrong.

Brandon Phillips is leading the 2013 Reds in OBI% with 19.9%. No one can guess who is second. I bet you anything. Go ahead. I'll wait.

Think of it yet? It's Xavier Paul! 16.1%!

However, I'm still correct in the hypothesis that Phillips is leading the Reds in opportunities. He is. However, that OBI% is also the highest of the last 5 years for him. Here are those numbers.


Pretty consistent other than 2010, right?

This season Phillips has 179 plate appearances with runners on base. Jay Bruce is third with 162. (and a respectable 15.1%)

Second is Joey Votto with 171, however, Phillips has had 11 more runners in scoring position. A higher percentage of Votto's runners have been on first base, but Votto isn't cashing his runners the way he should be. He's at 10.6%. Part of that could be being pitched around, I'm not sure. Those extra 11 runners on second and third aren't worth the 23 RBI lead Phillips has on Votto. If you raised Votto's OBI to 15% he would have 10 more RBI. You cannot assume that that would be 10 extra runs for the Reds, because the guys in the lineup behind Votto have knocked in some of those runs. In fact, the decrease in Votto's OBI may have contributed to the extra opportunities for Phillips, which currently, he is taking advantage of.