Baseball Player Won-Loss Records
Home     List of Articles



Event-Specific Park Factors

Beyond simply modeling expected runs scored, ballpark-specific base-out transition matrices enable one to assess more subtle differences between ballparks.

For some events, it is possible to deduce ballpark factors from these base-out transition matrices. For example, in general, a plate appearance which ends with the bases empty and the same number of outs as when the plate appearance started will have been a home run. Hence, one can estimate the relative home run frequencies across ballparks implied by the base-out transition matrices here.

Other less-obvious ballpark effects can also be induced. For example, based on a base-out transition matrix, one could impute park factors associated with on-base percentage (that is, plays that do not result in an out), doubles, triples, and even double plays (although double-play park factors calculated in this way tend to be highly erratic and exhibit extremely little year-to-year correlation by ballpark). One could then combine some of these factors (OBP, Doubles, Triples, Home Runs) to make a factor that roughly corresponds to slugging percentage (SLG) (note: this will not be exactly the same as SLG since the OBP factor here doesn't distinguish between singles, walks, and ROE (reached on errors)).

Park factors for all of these things, as well as runs, are presented below for all ballparks for 2009.

Component Park Factors Imputed from Base-Out Transition Matrices: 2009

American League Ballparks
City Ballpark Runs OBP SLG Doubles Triples Home Runs Double Plays
AnaheimAngel Stadium101.4100.5101.596.286.0110.5103.9
ArlingtonRangers Ballpark111.3106.6108.3107.6112.3114.8106.3
BaltimoreOriole Park at Camden Yards101.599.7101.599.7100.9109.990.7
BostonFenway Park104.397.3101.1119.9102.6104.899.0
ChicagoU.S. Cellular Field98.399.699.999.597.7101.482.6
ClevelandJacobs Field90.697.995.193.789.885.990.4
DetroitComerica Park101.5101.298.295.494.488.497.2
Kansas CityKauffman Stadium105.4106.8103.7113.4106.685.2112.8
MinneapolisMetrodome102.199.398.896.7108.897.393.1
New YorkNew Yankee Stadium100.0101.8101.890.1100.2108.7103.9
OaklandOakland Coliseum94.496.194.793.394.990.2103.6
SeattleSafeco Field97.799.899.094.895.698.9105.9
St. PetersburgTropicana Dome98.898.799.4101.0107.4100.292.4
TorontoSkydome92.894.797.098.8102.7103.9118.2
 
National League Ballparks
City Ballpark Runs OBP SLG Doubles Triples Home Runs Double Plays
AtlantaTurner Field93.798.196.086.689.094.294.7
ChicagoWrigley Field107.2101.9101.1104.496.096.6114.5
CincinnatiGreat American Ballpark99.4100.1102.8105.8101.6113.281.5
DenverCoors Field118.3106.6107.7114.9136.5102.4101.1
HoustonMinute Maid Park93.3100.9101.083.397.2113.6130.7
Los AngelesDodger Stadium91.795.294.6101.186.789.1106.9
MiamiPro Player Stadium110.1105.6103.299.065.7102.474.6
MilwaukeeMiller Field98.697.7103.1107.7118.2120.897.0
New YorkCiti Field90.895.797.492.095.1109.086.9
PhiladelphiaCitizens Bank Ballpark104.7101.0101.2103.7113.498.4117.8
PhoenixBank One Ballpark114.0104.8105.2121.9121.593.0103.6
PittsburghPNC Park99.2100.7101.599.187.2109.0100.8
San DiegoPetco Park77.492.989.078.394.877.8116.2
San FranciscoPac Bell Park109.2103.7102.9109.5106.494.594.9
St. LouisBusch Stadium II97.899.997.096.366.490.2101.2
WashingtonNationals Park94.895.296.296.4124.495.677.9


The next set of numbers are average Component Park Factors since 2000 for those ballparks which have been primary home parks for at least six seasons over this time period.

Component Park Factors Imputed from Base-Out Transition Matrices: 2000 - 2014

American League
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
AnaheimAngel Stadium
Mean Values97.899.398.697.096.696.699.8
Std. Deviation4.32.32.64.96.97.36.8
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
ArlingtonRangers Ballpark
Mean Values104.3101.7102.8102.3108.3106.5101.9
Std. Deviation5.02.02.86.27.17.97.2
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
BaltimoreOriole Park at Camden Yards
Mean Values101.3100.3100.594.995.4105.399.0
Std. Deviation4.22.32.76.26.56.66.8
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
BostonFenway Park
Mean Values102.7101.2101.6113.298.196.2100.0
Std. Deviation3.32.21.74.17.26.46.2
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
ChicagoU.S. Cellular Field
Mean Values103.1100.9101.997.495.8109.498.3
Std. Deviation4.22.42.75.27.26.08.6
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
ClevelandJacobs Field
Mean Values97.099.298.699.194.296.6103.2
Std. Deviation3.51.82.05.68.07.38.6
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
DetroitComerica Park
Mean Values100.7101.099.694.1107.796.4100.4
Std. Deviation5.42.52.84.77.19.27.6
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
Kansas CityKauffman Stadium
Mean Values103.2101.9101.5106.4103.696.6104.0
Std. Deviation5.22.42.74.38.17.87.5
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
MinneapolisMetrodome
Mean Values100.299.099.3100.9106.098.799.7
Std. Deviation4.82.11.73.38.13.05.1
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
New YorkYankee Stadium
Mean Values100.999.8100.796.495.1107.894.4
Std. Deviation1.80.90.83.66.23.76.1
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
OaklandOakland Coliseum
Mean Values96.198.197.699.999.893.8103.4
Std. Deviation5.02.22.65.86.77.96.2
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
SeattleSafeco Field
Mean Values93.797.696.494.394.693.1100.6
Std. Deviation4.82.62.65.37.07.07.3
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
St. PetersburgTropicana Dome
Mean Values97.199.198.797.4102.197.395.4
Std. Deviation4.12.12.04.86.46.37.0
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
TorontoSkydome
Mean Values100.699.4101.1105.6101.6105.2102.2
Std. Deviation3.62.51.85.78.86.87.9
 
National League
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
AtlantaTurner Field
Mean Values98.2100.399.798.696.998.398.7
Std. Deviation7.42.73.110.514.17.49.8
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
ChicagoWrigley Field
Mean Values102.4100.7101.3100.198.8104.8107.0
Std. Deviation7.82.83.67.916.010.78.0
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
CincinnatiGreat American Ballpark
Mean Values102.099.6102.497.289.9120.3101.5
Std. Deviation6.22.43.27.415.710.110.6
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
DenverCoors Field
Mean Values123.4108.1110.8110.3128.6119.7104.8
Std. Deviation9.73.63.86.512.913.210.0
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
HoustonMinute Maid Park
Mean Values102.6100.4102.199.2101.2111.5100.7
Std. Deviation6.02.12.88.013.96.116.1
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
Los AngelesDodger Stadium
Mean Values95.097.697.593.574.3103.297.6
Std. Deviation6.33.83.57.514.69.69.8
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
MiamiPro Player Stadium
Mean Values94.699.297.194.1112.288.089.2
Std. Deviation6.52.42.78.520.08.310.8
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
MilwaukeeMiller Field
Mean Values101.299.4101.498.394.2113.197.0
Std. Deviation4.43.03.07.817.510.814.3
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
New YorkShea Stadium
Mean Values95.397.896.697.583.793.094.7
Std. Deviation4.02.71.96.018.19.79.6
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
PhiladelphiaCitizens Bank Ballpark
Mean Values102.0100.3101.5100.793.7108.4100.7
Std. Deviation4.61.83.06.212.513.910.4
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
PhoenixBank One Ballpark
Mean Values108.3102.5104.1106.8124.4105.9100.3
Std. Deviation7.22.93.38.516.912.79.4
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
PittsburghPNC Park
Mean Values99.0100.399.0107.286.390.1104.2
Std. Deviation6.42.43.48.413.411.28.7
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
San DiegoPetco Park
Mean Values84.994.992.688.6108.382.699.2
Std. Deviation4.71.73.06.713.112.511.0
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
San FranciscoPac Bell Park
Mean Values96.799.997.7104.6111.682.3102.4
Std. Deviation8.63.04.58.717.914.67.9
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
St. LouisBusch Stadium II
Mean Values96.299.497.197.391.588.0100.7
Std. Deviation2.31.52.28.114.87.48.6
RunsOBPSLGDoublesTriplesHome RunsDouble Plays
WashingtonNationals Park
Mean Values100.9101.0100.6105.391.297.3107.2
Std. Deviation2.93.22.48.129.013.614.4


One thing worth pointing out is that these are aggregate value factors. If, for example, a ballpark generates 10% more home runs than average to left field but 10% fewer home runs than average to right field, this park would show up as having a Home Run Factor here of exactly 100 (assuming the same number of balls hit to left and right field). This is consistent with all of my work here, which attempts to measure pure value, not "true talent". In other words, these numbers do not say that any player who played in a ballpark with a Doubles Factor of 95 (e.g., Florida) and moved to a ballpark with a Doubles Factor of 110 (e.g., Colorado) would be expected to hit 16% more doubles (110/95 - 1). These numbers say that a player who hit 30 doubles in Florida is as valuable (all other things being equal) as a player who hit 35 doubles in Colorado.

One more note here: the numbers here are Ballpark Factors. That is, the numbers shown above for Colorado refer entirely to games played in Coors Field, NOT (as park factors are often reported) to all games played by the Colorado Rockies.

The next table shows two ways by which one can measure how stable these sorts of Ballpark Factors are from year to year.

The first of these is the standard deviation of these Ballpark Factors across all ballpark-seasons. This is calculated as follows: for every ballpark, average Ballpark Factors are calculated over the entire life of the ballpark. These are then used as the means in calculating the variance of these factors for each Ballpark. That is, if the average Home Run Factor in Anaheim Stadium was 99 over its 40+ year history (and it was), then for each of those 40+ seasons, the value 99 was used as the mean at Anaheim for calculating variance. The average Home Run Factor in Wrigley Field, on the other hand, was 108, so that is used as the mean for calculating variance for seasons at Wrigley.

The second row shows year-to-year correlations for Ballpark Factors for ballparks that were in existence in consecutive years (e.g., Memorial Stadium in Baltimore in 1979 and 1980, Memorial Stadium in 1980 and 1981, etc.).

Year-to-Year Consistency of Event Ballpark Factors

Runs OBP SLG Doubles Triples Home Runs Double Plays
Standard Deviation 16.615.515.617.219.419.917.5
Year-to-Year Correlation 30.9%11.9%21.4%31.2%37.5%48.6%5.4%


Keep in mind that all of these Ballpark Factors are expressed relative to all other ballparks. If other ballparks change, we shouldn't expect the Ballpark Factors of the remaining ballparks to stay unchanged, however. For example, in 1993, the Run Factors for the 12 National League parks that had been used in 1992 went down by an average of 3.1 with 7 of the 12 parks having lower Run Factors.

The reason for this, of course, was because the National League expanded in 1993, adding a team in Colorado, which played its home games in Mile High Stadium which had a Ballpark Run Factor in 1993 of 133.5. Other, less extreme, changes in ballparks have occurred in the past as well (the opening of Dodger Stadium, the Mariners moving from the Kingdome to Safeco Field, the Braves moving from Milwaukee to Atlanta, etc.).

In addition, I have not gone to the trouble of figuring out when ballparks may have been modified, which could lead to an expected change in Ballpark Factors. For example, the fences at Hiram Bithorn Stadium were moved out 30 feet between the 2003 and 2004 seasons. This had the perfectly predictable effect of decreasing the Home Run Factor for Hiram Bithorn Stadium from 133 to 86. I'm sure there are other examples of ballparks that were changed in ways that led to predictable and understandable changes in some of these factors.

In other words, there's no reason why we would expect these standard deviations to actually be zero or Year-to-Year correlations to be 100%. Given this fact, I am fairly satisfied with the consistency of most of these results - double plays being the notable exception.

Event-specific ballpark factors for all ballparks for the 2012 American League are presented here.

All articles are written so that they pull data directly from the most recent version of the Player won-lost database. Hence, any numbers cited within these articles should automatically incorporate the most recent update to Player won-lost records. In some cases, however, the accompanying text may have been written based on previous versions of Player won-lost records. I apologize if this results in non-sensical text in any cases.

Home     List of Articles