Dan Szymborski has released his annual ZIPS projections for Tigers players and they are posted at FanGraphs. Pre-season projections should always be taken with a grain of salt, but ZIPS is one of the longest-running and accurate systems available.
The good thing about projections is that they are unbiased and not subject to the emotions of overly optimistic (or pessimistic) hometown fans eagerly awaiting (dreading) the upcoming season. The limitation of projections is that they do not take into account anything beyond the numbers. For example, it might be known that a player played through an injury last year or is scheduled to receive more minor league seasoning or any other non-statistical factor that could impact his upcoming season.
I like to think of a projection as a baseline rather than a prediction. They tell you how a player should perform based on his past record and the records of similar players. A knowledgeable fan can then take that projection and adjust it up or down according to what he knows about that player beyond the statistics. One thing that fans are particularly good at any where forecasting systems often fail is guessing playing time. Fans generally know more than computers which players are going to play regularly, which ones will ride the bench and who will be farmed out to the minors to start the season.
That being said, here is what I found interesting about the 2013 ZIPS:
Not surprisingly, the two players that improved the most last year - center fielder Austin Jackson and left fielder Andy Dirks - are expected to fall back to earth in 2013. When a player has a season where he performs way above what he has done previously, he typically does not do as well the next season. Conversely, a player who hits below his career norms in a given season is likely to bounce back the following year. In each case, the phenomenon is called regression to the mean and most forecasting systems will factor that in to next seasons projection.
After hitting a combined .271/.331/.387 in his first first two seasons, Jackson surprised many observers with his .300/.377/.479 line in 2012. ZIPS has him at .263/.341/.418 in 2013. I am also expecting a regression although maybe not quite as much. With his excellent defense, he'd still be a really strong player with those numbers though.
After hitting .251/.296/.406 in his rookie season, Dirks shot up to .322/.370/.487 in 2012. ZIPS has him pegged for .276./319/.418 in 2013. That backs up the Tigers reluctance to make him a full-time player and may explain why they are apparently still looking for right-handed bat. ZIPS does not think the Tigers' young outfielders - Nick Castellanos and Avisail Garcia - are quite ready for prime time projecting both to have OPS below .700 if they do play in the majors.
None of the Tigers regulars are expected to improve significantly in 2013, but the team should benefit immensely from three new players. First, right fielder Torii Hunter's estimated .782 ZIPS OPS would be a huge upgrade over Brennan Boesch's .658 mark last year. Then designated hitter Victor Martinez (.770 ZIPS OPS) should be a major improvement over Delmon Young (.707 in 2012). Finally, they will get a full year out of second baseman Omar Infante (.695 ZIPS OPS) instead of Ryan Raburn (.480 in 2012) and Ramon Santiago (.555). That's about eight or nine wins right there without even getting great seasons from the three additions.
I don't think there is any projection system out there that is able to figure out the unpredictable mound creatures, so I won't dwell too much on that. The most depressing projection is the 5.60 ERA for Bruce Rondon, the man being hyped as the Tigers possible closer. Jim Leyland indicated at TigerFest today that Phil Coke is the back-up plan and ZIPS has him at 4.25. On a more positive note, ZIPS thinks left-hander Drew Smyly is pretty much for real forecasting a 4.20 ERA in 167 innings.
That's all for now, I will wait until spring training winds down before I sift through all the projections and other data and come up with my own predictions.
Saturday, January 26, 2013
Thursday, January 24, 2013
WAR Baseline is All About Playing Time
For people trying to learn sabermetrics, one of the most confusing concepts is the replacement baseline used in the Wins Above Replacement (WAR) statistics. In simple terms WAR is the wins a player contributed to his team's win total above what you would expect from a replacement level player - a theoretical player who could be acquired for league minimum salary. An example of a replacement player would be a player in AAA, who is good enough to get some time in the majors, but is not regarded as a top prospect.
Why is replacement used instead of average or zero? When building a ball club, comparing players to league average can be problematic. If a team is forced to replace a player due to an injury, he is not likely to be replaced by an average player or even a slightly below average player. Average players are actually good players and are not generally available quickly or cheaply. In most cases, the injured player will be replaced by a player who is substantially below average.
Comparing players to zero is also not generally a great idea because your replacement is not likely to bat .000 for any length of time. Your replacement will usually be somewhere between zero and average. Based on examination of data over several years, analysts determine how good a player typically needs to be to get a decent amount of playing time. The threshold above which a player must perform in order to get consistent at bats is called replacement level. Different people use somewhat different replacement levels, but I'll follow the FanGraphs.com definition here.
If you are interested in playing general manager and are concerned about roster construction or how much money a player is worth,the replacement threshold is the way to go. If you want to do something else such as selecting hall of famers or award winners or you just want to know how many players on your favorite team are above average, you can use an alternative baseline.
If you do decide to shun replacement level for something more intuitive though, you should understand the consequences. It all comes down to how much credit you want to give for playing time. Whether you choose Wins Above Average (WAA) or Wins Above Zero (WAO) or WAR can make a substantial difference when there is a lot of variation in playing time among players.
Suppose, Gary Great and Sammy Solid were both second basemen with exactly 600 Plate Appearances (PA). They were both average base runners and average defenders and played in neutral parks. The only way they differed was that Gary was a much better hitter than Sammy. Gary had a .400 OBP, .540 slugging average and .398 Weighted On-Base Average (wOBA). Sammy had a .325 OBP, .450 slugging average and .335 wOBA. The question is how many wins was Gary worth compared to Sammy?
We would normally have to do a lot of calculations involving base running, fielding and park effects in order to calculate Wins, but the question is simplified by assuming that the two players were similar in every way except batting. Based on PA and wOBA, Gary had 40 Batting Runs which means than he contributed an estimated 40 runs above what would be expected from an average player in the same number of plate appearances. Since 10 runs is worth approximately one win, he was 4 WAA.
Sammy had 10 Batting Runs or 1 WAA. So, there was a a gap of three wins between the two players. (Note that we should actually be adding a fraction of a win for playing second base, but they both get the same fraction so we'll ignore it for simplicity.)
What if we use zero as the baseline rather than average? An average player is worth 68 runs over 600 PA, so Gary was 40 + 68 = 108 runs above zero (also called Runs Created) or 10.8 WA0. Sammy had 78 Runs Created or 7.8 WA0. Again, the the two players were separated by three wins.
Finally, a replacement player is worth 20 runs per 600 PA below an average player, so Gary was 40 + 20 Runs Above Replacement or 6 WAR. Sammy was 30 Runs Above Replacement or 3 WAR. So, one more time there were three wins between the two batters. There was a very big disparity in the number of wins a each player was credited in WAA, WA0 and WAR, but no difference in the number of wins separating the two players because they had the same number of PA.
It's another story when players are far apart in their numbers of PA Suppose Gary had a .398 wOBA in 300 PA while Sammy still had a .335 wOBA in 600 PA. In that case, Gary had 20 Batting Runs compared to 10 for Sammy. That comes out to 2 WAA for Gary and 1 WAA for Sammy. So Gary was one win better by that measure. Does this make sense? Is a great hitter who missed half the season worth more wins than an above average hitter in a full season?
Let's see what happens if we change the threshold. An average player is worth 34 runs in 300 PA, so Gary was 20 + 34 = 54 Runs Above Zero. Sammy was still 78 Runs Above Zero. In terms of wins, Gary was 5.4 WA0 and Sammy 7.8 WA0. In this case, Sammy was 2.4 Wins better than Gary.
Finally, if replacement is the baseline, Gary was 20 + 10 = 30 Runs Above Replacement or 3 WAR while Sammy was 10 + 20 = 30 Runs Above Replacement or 3 WAR. So, they were considered equal contributors to wins by this metric.
The lesson learned is that the baseline you choose can make a large difference in your evaluation of players. In the first case, Gary was the better player. In the second instance, Sammy was the better player by a substantial margin. In the third situation, they were equals. You don't have to use replacement level if you don't want, but it's important to be aware how much the results vary among baselines.
Why is replacement used instead of average or zero? When building a ball club, comparing players to league average can be problematic. If a team is forced to replace a player due to an injury, he is not likely to be replaced by an average player or even a slightly below average player. Average players are actually good players and are not generally available quickly or cheaply. In most cases, the injured player will be replaced by a player who is substantially below average.
Comparing players to zero is also not generally a great idea because your replacement is not likely to bat .000 for any length of time. Your replacement will usually be somewhere between zero and average. Based on examination of data over several years, analysts determine how good a player typically needs to be to get a decent amount of playing time. The threshold above which a player must perform in order to get consistent at bats is called replacement level. Different people use somewhat different replacement levels, but I'll follow the FanGraphs.com definition here.
If you are interested in playing general manager and are concerned about roster construction or how much money a player is worth,the replacement threshold is the way to go. If you want to do something else such as selecting hall of famers or award winners or you just want to know how many players on your favorite team are above average, you can use an alternative baseline.
If you do decide to shun replacement level for something more intuitive though, you should understand the consequences. It all comes down to how much credit you want to give for playing time. Whether you choose Wins Above Average (WAA) or Wins Above Zero (WAO) or WAR can make a substantial difference when there is a lot of variation in playing time among players.
Suppose, Gary Great and Sammy Solid were both second basemen with exactly 600 Plate Appearances (PA). They were both average base runners and average defenders and played in neutral parks. The only way they differed was that Gary was a much better hitter than Sammy. Gary had a .400 OBP, .540 slugging average and .398 Weighted On-Base Average (wOBA). Sammy had a .325 OBP, .450 slugging average and .335 wOBA. The question is how many wins was Gary worth compared to Sammy?
We would normally have to do a lot of calculations involving base running, fielding and park effects in order to calculate Wins, but the question is simplified by assuming that the two players were similar in every way except batting. Based on PA and wOBA, Gary had 40 Batting Runs which means than he contributed an estimated 40 runs above what would be expected from an average player in the same number of plate appearances. Since 10 runs is worth approximately one win, he was 4 WAA.
Sammy had 10 Batting Runs or 1 WAA. So, there was a a gap of three wins between the two players. (Note that we should actually be adding a fraction of a win for playing second base, but they both get the same fraction so we'll ignore it for simplicity.)
What if we use zero as the baseline rather than average? An average player is worth 68 runs over 600 PA, so Gary was 40 + 68 = 108 runs above zero (also called Runs Created) or 10.8 WA0. Sammy had 78 Runs Created or 7.8 WA0. Again, the the two players were separated by three wins.
Finally, a replacement player is worth 20 runs per 600 PA below an average player, so Gary was 40 + 20 Runs Above Replacement or 6 WAR. Sammy was 30 Runs Above Replacement or 3 WAR. So, one more time there were three wins between the two batters. There was a very big disparity in the number of wins a each player was credited in WAA, WA0 and WAR, but no difference in the number of wins separating the two players because they had the same number of PA.
It's another story when players are far apart in their numbers of PA Suppose Gary had a .398 wOBA in 300 PA while Sammy still had a .335 wOBA in 600 PA. In that case, Gary had 20 Batting Runs compared to 10 for Sammy. That comes out to 2 WAA for Gary and 1 WAA for Sammy. So Gary was one win better by that measure. Does this make sense? Is a great hitter who missed half the season worth more wins than an above average hitter in a full season?
Let's see what happens if we change the threshold. An average player is worth 34 runs in 300 PA, so Gary was 20 + 34 = 54 Runs Above Zero. Sammy was still 78 Runs Above Zero. In terms of wins, Gary was 5.4 WA0 and Sammy 7.8 WA0. In this case, Sammy was 2.4 Wins better than Gary.
Finally, if replacement is the baseline, Gary was 20 + 10 = 30 Runs Above Replacement or 3 WAR while Sammy was 10 + 20 = 30 Runs Above Replacement or 3 WAR. So, they were considered equal contributors to wins by this metric.
The lesson learned is that the baseline you choose can make a large difference in your evaluation of players. In the first case, Gary was the better player. In the second instance, Sammy was the better player by a substantial margin. In the third situation, they were equals. You don't have to use replacement level if you don't want, but it's important to be aware how much the results vary among baselines.
Wednesday, January 23, 2013
Earl Webb: One-Year Doubles Wonder
Photo Credit: findagrave.com
This post has nothing to do with the Tigers, but I've always been intrigued by Red Sox outfielder Earl Webb's 1931 season when he set a record with 67 doubles. If you asked some random baseball fans who holds the record for most doubles in a season and told them it happened in 1931, they might guess a hall-of-fame slugger like Hank Greenberg or Lou Gehrig. They probably never heard of Webb, but that's what makes the record interesting.
One of the fun parts of learning about players from past eras is that their paths to professional baseball were typically much different from players of today and Webb was no exception. According to Society for American Baseball Research historian Bob Nowlin, Webb was born on a farm in Blue Spring Cove near Bon Air Tennessee on September 17, 1897. When he was six years old, his family moved to nearby Ravenscroft. He wass involved in coal mining before and after baseball and first starting working at age 11 as a trapper opening and closing mine shafts for 5 cents per hour.
At 14 years of age, Webb started pitching for the Ravenscroft baseball team and would sometimes walk eight-miles to play local rivals. As he matured, his arm attracted the attention of professional scouts and he signed to play with the Memphis Chickasaws of the Southern League in 1918. He initially failed to report but, after some prodding from his father (also a coal miner) Webb agreed to try his hand at professional ball in 1921. Memphis was still interested and farmed him out to Clarksdale Cubs of the Class D Mississippi State League where he would play as a pitcher and outfielder.
Legendary Giants manager John McGraw saw him pitch in early April, 1922 and New York paid $10,000 plus two players to Memphis for the rights to his contract. He impressed more as a hitter than a pitcher and became strictly a pitcher by 1925. He saw his first action in the majors that year getting three at bats with the Giants at age 27. He would not see action again until he played for the Cubs in 1927.
He didn't play a lot prior to 1930, largely because he was a poor fielder by his own admission. According to the Baseball: The Biographical Encyclopedia by David Pietrusza, Webb once said: "I have seen some mighty bad outfielders, but none of them had anything on me." He broke through as a regular in 1930 at age 32 with the Boston Red Sox hitting .323 with 30 doubles and 16 home runs in 505 plate appearances.
Webb's amazing 1931 season was foreshadowed in March when he reportedly blasted a 550-foot home run in spring training. He only hit 14 home runs during the season, but rapped doubles like no player before or since. His doubles totals for the season were:
| Month | Doubles |
| April | 5 |
| May | 16 |
| June | 9 |
| July | 18 |
| August | 6 |
| September | 13 |
| Total | 67 |
By season's end, he had a major-league record 67 doubles. That's a lot of doubles for anyone especially a player who never hit more than 30 in any other season. Why so many doubles? Was everyone hitting doubles that year? Not really. The second highest total was only 47 by Tigers first baseman Dale Alexander. In fact, Webb collected more two-base hits (48) by July 31 than any other player had all season.
Was Webb helped by Fenway Park? It's hard to say how much of a park effect was present, but the Red Sox finished sixth in an eight-team American League with 289 doubles (including Webb's tally). Boston was seventh in doubles in 1930 and last in 1932.
Nowlin reported that Webb was well aware of his assault on the doubles record (previously set by George Burns of the Indians in 1926) and that he might have turned some triples into doubles by holding up at second. How many times could that have happened though? It wouldn't have started early in the season as he surely had no plans of setting a doubles record given his prior history. He was not a huge triples hitter anyway, legging out six in 1930 and nine in 1932 compared to three in 1931.
Webb finished in the top ten in almost every offensive category in 1931 including 7th in batting average (.333), 10th in on-base percentage (.404), eighth in slugging (.528) and 5th in OPS+ (149). He was clearly the hitting star on a team with finished last in the AL in runs scored and had no other regular with an OPS+ as high as 100.
Webb was not the same in 1932 posting a 101 OPS+ with just 28 doubles and was traded to the Tigers during the season. He would never again approach the brilliance of his 1931 season retiring in 1935 after seven seasons with the Giants, Cubs, Red Sox, Tigers and White Sox.
It's hard to explain Webb's 1931 season but, regardless of the reasons for his propensity for doubles his performance earned him a record that still stands 82 years later.
Sunday, January 20, 2013
No Agreement on Johnny Peralta as a Defender at Shortstop
Today, I'm going to revisit Tom Tango's Fan Scouting Report. Tango annually conducts a survey on defensive skills which many of you completed last year. For those of you who are not familiar with the survey, he asks fans to rate the fielding skills of players on their favorite teams just based on observation and instructed them not to use any stats at all. As you can see on the ballot, fans are asked to scout players on reaction/instincts, acceleration/first few steps, speed, hands, release/footwork, throwing strength and throwing accuracy.
The results for past and present Tigers are shown in Table 2. The table tells us that there where 61 fans who rated center fielder Austin Jackson. They gave him top 16% in MLB scores for instincts (84), first steps quickness (88), speed (86) and hands (79), above average scores on release/footwork (65) and accuracy of arm (55) and a below average score on arm strength (45). This resulted in an average rating of 72 which translates to 8 runs above average.
Remember that these observations should be based on skills rather than performance. So, Jackson's score says that, in the eyes of fans, he has the skills to be a center fielder who is +8 runs above average in 1,184 innings (his total for 2012). On the other hand, his +5 runs above on Defensive Runs Saved (DRS) is an estimate of his actual performance last year.
Table 2: Fan Scouting Report for Tigers, 2012
Table 3 compares the fan fielding estimate with other defensive measures. For the most part, the Fan Fielding results for the Tiger players correlate reasonably well with the advanced measures. Players that did well on the advanced statistics such as second baseman Omar Infante, center fielder Austin Jackson and right fielder Torii Hunter - were also given high scores by the fans. It should be noted though that Tigers fans rated Infante as average while Marlins fans had him well above average.
The fans and defensive statistics agreed that Miguel Cabrera was a a below average fielder, but his -3 on FSR was better than any of his other statistics. There was universal agreement that right fielder Brennan Boesch was a poor defender.
There was more disagreement on shortstop Jhonny Peralta than any other player. Peralta was rated worse by the fans (-11) than any of the advanced fielding measures. Even the defensive metrics did not come to a consensus with Peralta ranging from -6 on Baseball Prospectus' FRAA to +10 on UZR. There is some thought that DRS and UZR could be biased for some players due to human error in recording of location of batted balls. So, perhaps the lower scores on FRAA and RZR might be more reliable for Peralta.
Why might the fans score be lower than all the metrics? It could be that Peralta's skills were poor, but was able to perform above his skills for the year. In that case, we might be wary about his ability to keep it up in 2013. On the other hand, it could be that the fans are biased because he does not look athletic and that he's actually better than he seems. We may learn more this year.
Table 3: Comparison of Runs Across Different Fielding Measures
Tango then tabulates the results which can be seen in detail on his site. Each player ends up with a score between 0 and 100 on each of the 7 skills. The league average rating for each of the 7 categories is 50 and a player with a rating of 70 or better is in top 16% in the league.
Tango also has a method for turning these ratings into runs scored above average. First, weighted averages are created for each player based on the importance of the skills at his position. The weights are shown on Table 1 below. I'm not exactly sure how he decided upon these weights but they make intuitive sense. For example, a third baseman's arm strength carries more weight than his speed. On the other hand, speed and acceleration are very important for outfielders.
Table 1: Weights for skills on fan fielding survey
Pos | Instincts | First Steps | Speed | Hands | Release | Arm Strength | Accuracy |
C | 1.3 | 0.3 | 0.3 | 1.3 | 1.3 | 1.3 | 1.3 |
1B | 1.6 | 1.6 | 0.4 | 1.6 | 0.8 | 0.4 | 0.4 |
2B | 1.6 | 1.6 | 0.8 | 1.6 | 0.8 | 0.4 | 0.4 |
SS | 1.5 | 1.5 | 0.7 | 0.7 | 1.5 | 0.7 | 0.4 |
3B | 0.9 | 0.9 | 0.5 | 0.9 | 0.9 | 1.9 | 0.9 |
LF | 1.0 | 2.0 | 2.0 | 1.0 | 0.5 | 0.3 | 0.3 |
CF | 1.0 | 1.9 | 1.9 | 1.0 | 0.5 | 0.5 | 0.2 |
RF | 0.9 | 1.9 | 1.9 | 0.9 | 0.5 | 0.5 | 0.5 |
Once the weighted average is computed for a player, it is subtracted from the average score for all players at his position. Adjustments are then made to make it comparable to the distribution of the Ultimate Zone Rating (UZR) statistic which has a smaller standard deviation. For those who are interested in the details, the adjustments are explained in the comments here. I tried following that algorithm and got slightly different results (although very close) from what is listed on FanGraphs under FSR. The FanGraphs results are used in this post.
Remember that these observations should be based on skills rather than performance. So, Jackson's score says that, in the eyes of fans, he has the skills to be a center fielder who is +8 runs above average in 1,184 innings (his total for 2012). On the other hand, his +5 runs above on Defensive Runs Saved (DRS) is an estimate of his actual performance last year.
Table 2: Fan Scouting Report for Tigers, 2012
player | n | instincts | first step | speed | hands | release | arm strength | accuracy | FSR | |
C | Avila, Alex | 47 | 52 | 40 | 33 | 56 | 67 | 63 | 63 | 2 |
C | Pena, Brayan | 18 | 33 | 23 | 23 | 31 | 29 | 36 | 29 | -4 |
1B | Fielder, Prince | 55 | 32 | 20 | 25 | 25 | 25 | 36 | 31 | -6 |
2B | Infante, Omar | 47 | 70 | 65 | 61 | 63 | 65 | 49 | 63 | +5 |
3B | Cabrera, Miguel | 62 | 49 | 30 | 27 | 48 | 47 | 79 | 58 | -3 |
SS | Peralta, Jhonny | 56 | 46 | 32 | 36 | 62 | 59 | 51 | 61 | -11 |
CF | Jackson, Austin | 61 | 84 | 88 | 86 | 79 | 65 | 45 | 55 | +8 |
RF | Hunter, Torii | 19 | 75 | 70 | 61 | 75 | 77 | 70 | 77 | +7 |
RF | Boesch, Brennan | 52 | 10 | 23 | 38 | 11 | 22 | 45 | 35 | -5 |
Table 3 compares the fan fielding estimate with other defensive measures. For the most part, the Fan Fielding results for the Tiger players correlate reasonably well with the advanced measures. Players that did well on the advanced statistics such as second baseman Omar Infante, center fielder Austin Jackson and right fielder Torii Hunter - were also given high scores by the fans. It should be noted though that Tigers fans rated Infante as average while Marlins fans had him well above average.
The fans and defensive statistics agreed that Miguel Cabrera was a a below average fielder, but his -3 on FSR was better than any of his other statistics. There was universal agreement that right fielder Brennan Boesch was a poor defender.
There was more disagreement on shortstop Jhonny Peralta than any other player. Peralta was rated worse by the fans (-11) than any of the advanced fielding measures. Even the defensive metrics did not come to a consensus with Peralta ranging from -6 on Baseball Prospectus' FRAA to +10 on UZR. There is some thought that DRS and UZR could be biased for some players due to human error in recording of location of batted balls. So, perhaps the lower scores on FRAA and RZR might be more reliable for Peralta.
Why might the fans score be lower than all the metrics? It could be that Peralta's skills were poor, but was able to perform above his skills for the year. In that case, we might be wary about his ability to keep it up in 2013. On the other hand, it could be that the fans are biased because he does not look athletic and that he's actually better than he seems. We may learn more this year.
Table 3: Comparison of Runs Across Different Fielding Measures
Player | FSR | DRS | UZR | FRAA | RZR | |
C | Avila, Alex | +2 | +6 | --- | -3 | --- |
C | Pena, Brayan | -4 | +1 | --- | -1 | --- |
1B | Fielder, Prince | -6 | -4 | -2 | +1 | +4 |
2B | Infante, Omar | +5 | +6 | +10 | +4 | +12 |
3B | Cabrera, Miguel | -3 | -4 | -4 | -10 | -18 |
SS | Peralta, Jhonny | -11 | -1 | +10 | -6 | -3 |
CF | Jackson, Austin | +8 | +5 | +4 | +9 | +11 |
RF | Hunter, Torii | +7 | +15 | +10 | +1 | +11 |
RF | Boesch, Brennan | -9 | -8 | -12 | -5 | -11 |
Subscribe to:
Posts (Atom)
