Don't be fooled by errors
Many teams with good fielding percentages aren't very good on defense
Ben Jedlovec
Baseball Info Solutions
ESPN Insider
In the piece I wrote last week about Cliff Lee's struggles, I pointed out that a lot of his problems had to do with the Phillies' poor defense. This led one commenter to write:
"This article seems to forget that the Phillies are the team with the second least amount of errors in baseball. Poor defense? No. Defensive runs saved is a very flawed statistic."
The commenter is correct about the errors. The Philadelphia Phillies have made the third-fewest errors and have the highest fielding percentage in the baseball this season. By this logic, you might be fooled into thinking they're a great defensive team behind their superb pitching staff. But there is so much more to good defense than not making errors.
<OFFER>Errors provide useful information. When the official scorer assigns an error, we know that the fielder has gotten to the ball and either bobbled it, made a bad throw or committed some other miscue on a play that would normally be made successfully. The fielder screwed up, and he should be penalized for it; hence, he is charged with an error.
However, there are well-documented issues with using errors to evaluate fielders. For starters, there is often room for disagreement in the official scorer's ruling. Additionally, errors don't appropriately account for a fielder's range; if a shortstop is a step slow and doesn't reach a ground ball through the hole, he's not likely to be charged with an error, although other shortstops might have made the play.
Many analysts have attempted to accommodate these concerns. For example, Bill James decided to focus on the plays a fielder does make, rather than the plays he does not. Along these lines, he proposed the original range factor formula, which simply counts the number of assists and putouts per nine innings. Players with great range tend to reach more balls and have a higher range factor than others at their position.
While an improvement over Fielding Percentage, range factor still has a few complicating factors. For instance, a shortstop behind a ground-ball pitcher would have more opportunities than a fly-ball pitcher.
Phillies pitchers have induced a ton of ground balls this year. The league batting average on ground balls is .227. The Phillies have allowed a .248 average on grounders, the third-worst in baseball behind the Arizona Diamondbacks and the Chicago Cubs. This is a strong indication that Philadelphia infielders are not getting to that many ground balls, and you can't make an error on a ball you don't get to.
Additionally, ground balls are not all equal. Obviously, a hard-hit ball in the hole is a much tougher play than a routine three-hopper right at the shortstop. Additionally, as discussed recently in an article about batting average on balls in play, ground-ball specialists like Trevor Cahill tend to allow a lower batting average than other pitchers on their ground balls. This unevenness is not guaranteed to even out over the course of the season.
That's why Baseball Info Solutions' video scouts plot the location and velocity (hard, medium or soft) of each ground ball over the course of the season. With this information, we can approximate the difficulty of each ball in play (not just ground balls) for each fielder. That's exactly what John Dewan did with the plus/minus system.
Given the location and velocity of the ground balls allowed by the Phillies' pitching staff year, a lineup of average infielders would have made about 627 plays. The Phillies' infield actually made 587 plays, a 40-play difference. Converting 40 outs into hits makes a difference of about 30 runs. Despite their low error total, the Phillies' infield has cost the pitching staff about 40 outs and 30 runs this year. Check out the chart below, which shows the teams with the biggest gap between their infield's ranking in fielding percentage and plus/minus.
<!-- begin inline 1 -->Major differences
The following five teams have the largest discrepancy between their infield rankings for fielding percentage and defensive runs saved. The Phillies come up first in one and last in the other.
<TABLE><THEAD><TR><TH>Team</TH><TH>Fielding Pct</TH><TH>Rank</TH><TH>Plus/Minus</TH><TH>Rank</TH></TR></THEAD><TBODY><TR class=last><TD>Phillies</TD><TD>0.988</TD><TD>1</TD><TD>-40</TD><TD>30</TD></TR><TR class=last><TD>Rangers</TD><TD>0.980</TD><TD>27</TD><TD>12</TD><TD>3</TD></TR><TR class=last><TD>Athletics</TD><TD>0.980</TD><TD>28</TD><TD>-14</TD><TD>12</TD></TR><TR class=last><TD>Dodgers</TD><TD>0.987</TD><TD>6</TD><TD>-24</TD><TD>21</TD></TR><TR class=last><TD>Mets</TD><TD>0.986</TD><TD>8</TD><TD>-22</TD><TD>19</TD></TR></TBODY></TABLE>
<!-- end inline 1 -->
The Phillies are an excellent team that will very likely win the National League East again. But that doesn't mean they have a good infield defense. In fact, despite what fielding percentage might tell you, they have a poor one.
Many teams with good fielding percentages aren't very good on defense
Ben Jedlovec
Baseball Info Solutions
ESPN Insider
In the piece I wrote last week about Cliff Lee's struggles, I pointed out that a lot of his problems had to do with the Phillies' poor defense. This led one commenter to write:
"This article seems to forget that the Phillies are the team with the second least amount of errors in baseball. Poor defense? No. Defensive runs saved is a very flawed statistic."
The commenter is correct about the errors. The Philadelphia Phillies have made the third-fewest errors and have the highest fielding percentage in the baseball this season. By this logic, you might be fooled into thinking they're a great defensive team behind their superb pitching staff. But there is so much more to good defense than not making errors.
<OFFER>Errors provide useful information. When the official scorer assigns an error, we know that the fielder has gotten to the ball and either bobbled it, made a bad throw or committed some other miscue on a play that would normally be made successfully. The fielder screwed up, and he should be penalized for it; hence, he is charged with an error.
However, there are well-documented issues with using errors to evaluate fielders. For starters, there is often room for disagreement in the official scorer's ruling. Additionally, errors don't appropriately account for a fielder's range; if a shortstop is a step slow and doesn't reach a ground ball through the hole, he's not likely to be charged with an error, although other shortstops might have made the play.
Many analysts have attempted to accommodate these concerns. For example, Bill James decided to focus on the plays a fielder does make, rather than the plays he does not. Along these lines, he proposed the original range factor formula, which simply counts the number of assists and putouts per nine innings. Players with great range tend to reach more balls and have a higher range factor than others at their position.
While an improvement over Fielding Percentage, range factor still has a few complicating factors. For instance, a shortstop behind a ground-ball pitcher would have more opportunities than a fly-ball pitcher.
Phillies pitchers have induced a ton of ground balls this year. The league batting average on ground balls is .227. The Phillies have allowed a .248 average on grounders, the third-worst in baseball behind the Arizona Diamondbacks and the Chicago Cubs. This is a strong indication that Philadelphia infielders are not getting to that many ground balls, and you can't make an error on a ball you don't get to.
Additionally, ground balls are not all equal. Obviously, a hard-hit ball in the hole is a much tougher play than a routine three-hopper right at the shortstop. Additionally, as discussed recently in an article about batting average on balls in play, ground-ball specialists like Trevor Cahill tend to allow a lower batting average than other pitchers on their ground balls. This unevenness is not guaranteed to even out over the course of the season.
That's why Baseball Info Solutions' video scouts plot the location and velocity (hard, medium or soft) of each ground ball over the course of the season. With this information, we can approximate the difficulty of each ball in play (not just ground balls) for each fielder. That's exactly what John Dewan did with the plus/minus system.
Given the location and velocity of the ground balls allowed by the Phillies' pitching staff year, a lineup of average infielders would have made about 627 plays. The Phillies' infield actually made 587 plays, a 40-play difference. Converting 40 outs into hits makes a difference of about 30 runs. Despite their low error total, the Phillies' infield has cost the pitching staff about 40 outs and 30 runs this year. Check out the chart below, which shows the teams with the biggest gap between their infield's ranking in fielding percentage and plus/minus.
<!-- begin inline 1 -->Major differences
The following five teams have the largest discrepancy between their infield rankings for fielding percentage and defensive runs saved. The Phillies come up first in one and last in the other.
<TABLE><THEAD><TR><TH>Team</TH><TH>Fielding Pct</TH><TH>Rank</TH><TH>Plus/Minus</TH><TH>Rank</TH></TR></THEAD><TBODY><TR class=last><TD>Phillies</TD><TD>0.988</TD><TD>1</TD><TD>-40</TD><TD>30</TD></TR><TR class=last><TD>Rangers</TD><TD>0.980</TD><TD>27</TD><TD>12</TD><TD>3</TD></TR><TR class=last><TD>Athletics</TD><TD>0.980</TD><TD>28</TD><TD>-14</TD><TD>12</TD></TR><TR class=last><TD>Dodgers</TD><TD>0.987</TD><TD>6</TD><TD>-24</TD><TD>21</TD></TR><TR class=last><TD>Mets</TD><TD>0.986</TD><TD>8</TD><TD>-22</TD><TD>19</TD></TR></TBODY></TABLE>
<!-- end inline 1 -->
The Phillies are an excellent team that will very likely win the National League East again. But that doesn't mean they have a good infield defense. In fact, despite what fielding percentage might tell you, they have a poor one.
