Remember me
Forgot your username/password?
 
Print
Home>Articles

Postseason Impact

By Dave Fleming
October 7, 2019
 
By Paresh Gupta
Guest Contributor
 
October is here and it’s the best time of the year for baseball. As teams make a run for a championship, a byproduct for some players is to further build their Hall of Fame case based on their postseason performance. Next year, Curt Schilling and Andy Pettitte will remain on the Hall of Fame ballot and a good part of their cases ride on their October successes. And in the coming years, we will have David Ortiz and Carlos Beltran who also had stellar postseason careers.
 
When it comes to Hall of Fame voting, every voter has different criteria, but for the most part, we hear that a player’s postseason play is considered either extra credit or his resume gets a "bump".
 
For instance, Kirby Puckett and Catfish Hunter have seven World Series rings between them, and that certainly helped them to get over the top for enshrinement. In the meantime, similar statistically Don Mattingly and Luis Tiant are still on the outside looking in. 
 
Back then, the evaluation for the Hall of Fame vote was different. Aside from statistics, much of it was also based on awards, "magic" numbers, and frankly, just overall feel. Nowadays, just like the rest of the game, analytics are coming more and more into play whereas the sabermetric darlings like Bert Blyleven and Tim Raines are justifiable in.  
 
Metrics like WAR, WAR7, and JAWS are now used to evaluate a player’s Hall of Fame candidacy but WAR and JAWS do not account for postseason play.  We don’t have similar objective statistics to evaluate postseason performance. The extra credit for postseason play has never really been defined or quantified.
 
Compared to the regular season, the postseason is a small sample size (SSS), but the games still count- they actually count the most!  Ultimately, the sport’s champion is determined on SSS. Even though the data pool is smaller, the data can still be quantified. And just like any other statistic, we need to be aware of the context.
 
Currently a nice objective stat to evaluate postseason value is WPA (Win Probability Added). WPA is a cumulative story telling statistic where you can also accumulate negative value and takes into account all aspects of a given game’s situation (inning of play, number of outs, men on base, etc). The problem with evaluating just WPA in itself is it doesn’t quite tell the entire story. We are unable to evaluate if a player has had an extensive playoff career or not. We’re also unable to decipher their peak playoff performance. Take for instance Eric Hosmer and David Freese. They both have the same exact postseason WPA (2.2) but they got there in very different fashion (Eric Homer 138 PAs versus David Freese 222 PAs).
 
So as a fun tool and general barometer to calculate and further assess playoff performance, we can play around with the WPA stat. And whether or not you subscribe to the notion of the postseason bump for Hall of Fame evaluation, it is still an interesting and entertaining analysis.
 
To calculate performance on a per plate appearance, we can take WPA divided by Plate Appearances (PAs) and then multiply by 100 (as an adjustment factor). We’ll call this statistic Postseason Impact (PSI). PSI will provide a relative gauge on how much a player contributed per at bat.
 
WPA/PAs x 100 = PSI
 
To calculate an overall career post season performance, we can multiply WPA by Plate Appearances, and then divide by 100 (as an adjustment constant). We’ll call this statistic Career Postseason Impact (CPSI). 
 
WPA x PAs / 100 = CPSI
 
CPSI is purposely weighting and rewarding playing time, as it is a career volume statistic. WPA is a cumulative stat but in terms of playing time it can work against a player. VPSI is offsetting that aspect and representinga player’s career postseason playing time. Overall, CPSI focuses on quality and quantity.
 
Now with that in mind and taking our Hosmer and Freese example, their PSI and CPSI numbers are as follows:
 
Player
WPA
(Postseason)
PSI
(Per PA Factor)
CPSI
(PA Weighted Factor)
Eric Hosmer
2.2
1.6
3.0
David Freese
2.2
0.99
5.8
 
Obviously, this can be viewed in two different ways. One can be "pro-Hosmer" in that he was able to accumulate just as much WPA in less plate appearances (portrayed by his higher PSI). Another can be viewed as "pro-Freese" in that he was able to sustain his excellence (portrayed by his higher CPSI).   
 
These numbers aren’t intended to sway you one way or the other. It’s just allowing you to understand the context where the numbers came from.
 
Listed here are some of the top players in career postseason WPA along with some other notable Hall of Fame candidates and recent Hall of Fame inductees from the wildcard era:
 
Postseason WPA Batting
David Ortiz
3.2
Albert Pujols
2.9
Lance Berkman
2.7
Justin Turner
2.6
Carlos Beltran
2.3
Eric Hosmer
2.2
David Freese
2.2
Miguel Cabrera
1.9
Larry Walker
0.93
Fred McGriff
0.9
Bernie Williams
0.88
Chipper Jones
0.41
Vladmir Guerrero
0.37
Edgar Martinez
0.34
Derek Jeter
0.02
Jeff Kent
-0.13
Scott Rolen
-0.43
Jim Thome
-0.46
Andruw Jones
-0.77
 
*Italics denotes Hall of Famer
 
 
The negative WPAs listed above are not listed to disparage the player but to put into context and appreciation for those with positive WPAs. Listed are the same players (minus the players with negative WPA) with their PSI and CPSI numbers:
 
Player
WPA
(Postseason)
PSI
(Per PA Factor)
CPSI
(PA Weighted Factor)
David Ortiz
3.2
0.86
11.7
Albert Pujols
2.9
0.88
9.7
Lance Berkman
2.7
1.2
6.0
Justin Turner
2.6
1.2
5.4
Carlos Beltran
2.3
0.88
5.9
Eric Hosmer
2.2
1.6
3.0
David Freese
2.2
0.99
5.8
Miguel Cabrera
1.9
0.82
4.5
Larry Walker
0.93
0.77
1.1
Fred McGriff
0.90
0.41
2.0
Bernie Williams
0.88
0.16
4.8
Chipper Jones
0.41
0.12
1.4
Vladmir Guerrero
0.37
0.20
0.70
Edgar Martinez
0.34
0.23
0.50
Derek Jeter
0.02
0.0027
0.15
 
*Italics denotes Hall of Famer
 
With keeping WPA in mind, some interesting takeaways from the PSI and CPSI batting numbers:
 
·         Bernie Williams has a lower WPA than Eric Hosmer and Miguel Cabrera and significantly lower PSI. But if you factor in total plate appearances, Bernie has a higher CPSI than both of them and has the four rings to show for it. 
 
·         David Freese’s and Justin Turner’s high postseason WPAs are not just due to a 2 or 3 series hot streak. In addition to their relatively high PSI, they have a high CPSI due to their significant number of postseason plate appearances (200 plus for both). 
 
·         Both Chipper Jones and Derek Jeter will rightfully be first ballot Hall of Famers. But it’s interesting that their teammates Fred McGriff and Bernie Williams, who played less postseason series, have a higher CPSI than their Hall of Fame teammates due to their higher PSIs.  
 
·         Of all of these players, Eric Hosmer has the highest PSI, but is on the lower end of CPSI. This indicates Hosmer has made the most of his limited opportunities, but his overall postseason career (in terms of volume) is still a little light compared to the others.
 
·         Border-line Hall of Fame candidate Carlos Beltran and Lance Berkman had won "only" one championship each (and Beltran was not really a huge part of the 2017 Astros championship run from a playing perspective) but they both had significant postseason successes with a higher per plate appearance PSI than Ortiz and higher CPSI than Bernie Williams and future HOF Miguel Cabrera.
 
·         Speaking of Ortiz, as we all know Big Papi had monster postseasons, and it’s not overrated- it is duly rated and especially conveyed in his CPSI. Same for Albert Pujols.
 
As mentioned previously, with WPA players can tend to rack up negative value which will affect their overall PSI and CPSI. And with WPA alone, we cannot evaluate peak performance. Thus, just like we have WAR7 which accounts for a player’s career regular season peak, we can also evaluate a player’s playoff peak by calculating the player’s best 5 postseasons’ series (utilizing the best 5 WPAs). PSI5 will evaluate a player’s performance on a plate appearance basis while CPSI5 will evaluate a player’s performance by weighting PAs (volume based).
 
Player
WPA
(Postseason)
PSI5
(Per PA Factor)
CPSI5
(PA Weighted Factor)
David Ortiz
3.2
2.8
4.0
Albert Pujols
2.9
2.1
3.7
Lance Berkman
2.7
2.6
2.5
Justin Turner
2.6
2.3
2.2
Carlos Beltran
2.3
2.1
3.2
Eric Hosmer
2.2
2.4
1.8
David Freese
2.2
2.4
2.5
Miguel Cabrera
1.9
2.1
2.0
Fred McGriff
0.93
1.8
1.5
Larry Walker
0.90
0.94
1.1
Bernie Williams
0.88
2.2
2.8
Chipper Jones
0.41
1.8
1.8
Vlad Guerrero
0.37
1.4
1.4
Edgar Martinez
0.34
1.3
1.3
Derek Jeter
0.02
1.2
1.7
Jeff Kent
-0.13
1.1
0.97
Scott Rolen
-0.43
0.76
0.68
Jim Thome
-0.46
1.4
0.66
Andruw Jones
-0.77
1.2
0.62
 
Once more, these peak numbers for each player are based on the same number of series from the wildcard era (5 series). Some takeaways from these numbers:
 
·         Again, David Ortiz is on the top of this list of peak values for both per plate appearance and overall performance. Big Papi is the gold standard for postseason performance, which should give him extra credit on his Hall of Fame resume. He is currently on the top of the list and other player’s potential October extra credit should not necessarily need to equal or exceed his numbers.
 
·         After Ortiz and Pujols, the next tier for CPSI5 are Carlos Beltran, Bernie Williams, Lance Berkman, and David Freese, indicating they all had stellar October peaks.  On a per plate appearance (PSI5), Berkman actually has the second highest on this list. 
 
·         Compared to McGriff, Chipper has a higher CPSI5 while their PSI5s are similar. Bernie Williams has significantly higher peak numbers than Jeter. 
 
·         Hosmer and Thome have significantly higher PSI5s than CPSI5s which conveys they had low amount of plate appearances in their peak series’ and were able to take advantage of their high leverage situations. In one series, Hosmer had a WPA of 0.59 in just 6 plate appearances. Thome had a WPA of 0.20 in just 2.
 
·         Jeff Kent, Scott Rolen, and Andruw Jones all had negative overall WPA but did have positive peak numbers. Expectedly they are lower on this list as well. They appear not to have enough playoff numbers to warrant a bump in their Hall of Fame resume.
 
This very same evaluation can be done for pitching. To calculate postseason impact for pitchers, we can replace plate appearances with innings pitched.
 
To calculate performance on a per inning basis, we can take WPA divide by Innings Pitched (IPs) and then multiply by 100 (as an adjustment factor). 
 
WPA/IPs x 100 = PSI
 
To calculate an overall career postseason performance, we can multiply WPA by Innings Pitched, and then divide by 100 (as an adjustment constant).
 
WPA x IPs / 100 = CPSI
 
Listed here are some of the top starting pitchers in career postseason WPA along with some other notable Hall of Fame candidates and recent Hall of Fame inductees from the wildcard era:
 
 
Postseason WPA Pitching
Curt Schilling
4.1
John Smoltz
3.6
Andy Pettitte
3.5
Jon Lester
3.4
Orel Hershiser
2.8
Madison Bumgarner
2.7
Orlando Hernandez
2.6
Tom Glavine
2.4
Justin Verlander
2.3
Mike Mussina
1.8
Cliff Lee
1.8
Josh Beckett
1.6
Randy Johnson
1.4
Greg Maddux
1.3
Pedro Martinez
1.2
Kevin Brown
0.7
David Cone
0.52
 
*Italics denotes Hall of Famer
 
 
The pitching PSI and CPSI calculations are just as intriguing:
 
Player
WPA
(Postseason)
PSI
(Per IP Factor)
CPSI
(IP Weighted Factor)
Curt Schilling
4.1
3.1
5.4
John Smoltz
3.6
1.7
7.5
Andy Pettitte
3.5
1.3
9.7
Jon Lester
3.4
2.2
5.2
Orel Hershiser
2.8
2.1
3.7
Madison Bumgarner
2.7
2.7
2.8
Orlando Hernandez
2.6
2.4
2.7
Tom Glavine
2.4
1.1
5.2
Justin Verlander
2.3
1.5
3.5
Mike Mussina
1.8
1.3
2.4
Cliff Lee
1.8
2.2
1.5
Josh Beckett
1.6
1.8
1.5
Randy Johnson
1.4
1.2
1.7
Greg Maddux
1.3
0.66
2.6
Pedro Martinez
1.2
1.3
1.2
Kevin Brown
0.70
0.86
0.57
David Cone
0.52
0.47
0.58
 
Some takeaways from this list:
 
·         Andy Pettitte has a smaller WPA than Curt Schilling and John Smoltz, but he's pitched almost twice as much as Schilling and much more than Smoltz. Pettitte’s volume CPSI (highest on this list, by far) conveys the point that he has pitched well for a longer period of time. Unfortunately, his per inning pitched PSI is on the lower half of this list. Similar with John Smoltz, who has the second highest CSPI but a middle of the road PSI.
 
·         Jon Lester CPSI is amongst Curt Schilling and Tom Glavine (two of the all-time greats) and his PSI is in the Top 5 on this list. However, he is someone we don’t always think of as a postseason star and is another future borderline Hall of Fame candidate. In 2013, his performance was overshadowed by Ortiz’s incredible performance and in 2016 the Cubs broke a 108 year-old curse.
 
·         Mike Mussina and Cliff Lee have the very same WPA but they both got there on different paths. Moose has the longer playoff career and higher CPSI, while Lee has the better per inning pitched performance and higher PSI.
 
·         Madison Bumgarner has the second highest PSI (after Curt Schilling) but his CPSI is closer to the middle. Like Hosmer in batting, this indicates Bumgarner has made the most of his opportunities but his overall postseason volume is less than the others.
 
·         Speaking of Schilling, his CPSI is in the very next tier down from Pettitte and Smoltz, and his PSI is the highest on the entire list- by far. Schilling has high postseason volume and has the best performance on a per inning basis as well.
 
And then if we take it a step further, their peak PSI5 and CPSI5 numbers:
 
Player
WPA
(Postseason)
PSI5
(Per IP Factor)
CPSI5
(IP Weighted Factor)
Curt Schilling
4.1
4.8
2.2
Jon Smoltz
3.6
4.9
1.3
Andy Pettitte
3.5
5.1
0.91
Jon Lester
3.4
4.8
1.7
Orel Hershiser
2.8
3.7
1.8
Madison Bumgarner
2.7
5.0
1.8
Orlando Hernandez
2.6
4.6
1.0
Tom Glavine
2.4
5.3
1.4
Justin Verlander
2.3
5.1
2.0
Mike Mussina
1.8
4.3
1.3
Cliff Lee
1.8
4.1
1.7
Josh Beckett
1.63
3.6
1.5
Randy Johnson
1.4
3.5
1.8
Greg Maddux
1.3
4.8
1.4
Pedro Martinez
1.2
4.9
0.71
Kevin Brown
0.7
2.9
0.81
David Cone
0.52
3.4
0.47
 
·         Justin Verlander has the second highest CPSI5 (after Schilling), and the highest PSI5, making a strong case that he is one of the best October peak pitchers in the wildcard era.
 
·         Orel Hershiser has a lower postseason WPA, PSI, and CPSI than Andy Pettitte and John Smoltz but has a higher peak CPSI5 than both of them, and Top 5 overall. During his peak, Hershiser pitched extremely well with more IPs, while both Smoltz and Pettitte pitched at a high level for a longer period of time.
 
·         Amongst the Braves Big 3, Smoltz is often regarded as the best postseason pitcher. But when it comes to peak, Glavine has the best numbers and Maddux’s CPSI5 is actually slightly better than Smoltz.
 
·         As mentioned earlier, Jon Lester has solid career playoff numbers, and also has very respectable peak numbers as well, higher than Cliff Lee, former teammate Josh Beckett, and Hall of Famers Mike Mussina and Greg Maddux.
 
·         Pedro Martinez, Smoltz, and Pettitte all have relatively high PSI5s while they have lower CSPI5s. This is due to their high WPA series coming from a lower number of innings pitched. For Pettitte and Pedro, they only pitched about 7 innings in 4 of the 5 series. For Smoltz, he had three series with less than 10 innings.
 
·         In contrast, Hershiser has a low PSI5 with a relatively higher CSPI5. This is due to Hershiser’s high number of innings pitched during his peak (average 14.1 IP per series).
 
 
So, now what? What does of all of this mean and how can it be used for Hall of Fame evaluation? Let’s start with the batters, and look further into these players: 
 
Player
PSI
(Per PA Factor)
CPSI
(PA Weighted Factor)
PSI5
(Per PA Factor)
CPSI5
(PA Weighted Factor)
David Ortiz
0.86
11.7
2.8
4.0
Albert Pujols
0.88
9.7
2.1
3.7
Lance Berkman
1.2
6.0
2.6
2.5
Carlos Beltran
0.88
5.9
2.1
3.2
Bernie Williams
0.16
4.8
2.2
2.8
 
Pujols is already an automatic first ballot Hall of Famer and he is listed here for comparison purposes only. And as mentioned previously, David Ortiz is the gold standard for Hall of Fame extra credit. Other players do not necessarily have to eclipse their numbers to merit "the bump".
 
Lance Berkman and Carlos Beltran have also had tremendous Octobers and have very high playoff impact statistic numbers across the board (peak and career). But unlike Beltran, Lance Berkman didn’t do quite enough with his regular season career for the postseason extra credit to really come into play. As a rough gauge, Berkman’s career WAR is 52.1 while the average HOF left fielder is 65.5. This past election, Berkman did not reach the 5% threshold on his first year and will no longer appear on the BBWAA ballot. 
 
On the other hand, Carlos Beltran’s candidacy will be coming up in 2023 and he has a very strong case on regular season numbers alone (career WAR 69.6). Based on his overall stellar PSIs and CPSIs stats, a voter can justifiably use these numbers to bump up his resume and put him over the top for election.
 
As for Bernie Williams, there are some that have indicated that his extensive October career is so strong that in and of itself warrants his election. Bernie did have a great postseason career and his peak postseason impact numbers (PSI5 and CPSI5) are amongst Beltran and Berkman. But Bernie’s overall postseason impact numbers are not as high and his PSI is one of the lowest on the list. Objectively speaking, if you look at his total resume (career WAR 49.6), it does not warrant him completely leapfrogging his way into Cooperstown. If Bernie were to do so, there would be a justifiable reason to do the same for Berkman, who actually has better overall postseason impact numbers.  
 
 
For pitching, let’s take a look into these players:
 
Player
PSI
(Per IP Factor)
CPSI
(IP Weighted Factor)
PSI5
(Per IP Factor)
CPSI5
(IP Weighted Factor)
Curt Schilling
3.1
5.4
4.8
2.2
Andy Pettitte
1.3
9.7
5.1
0.91
Jon Lester
2.2
5.2
4.8
1.7
Orel Hersheiser
2.1
3.7
3.7
1.8
Madison Bumgarner
2.7
2.8
5.0
1.8
Justin Verlander
1.5
3.5
5.1
2.0
 
Based on these numbers (peak and overall), Curt Schilling has distinguished himself as October’s finest. In comparison, Andy Pettitte has double the CPSI but a much lower PSI and CPSI5. On this list, Pettitte has the lowest PSI and peak numbers. Like Berkman, Pettitte was a fantastic player (career WAR 60.2, average HOF starting pitcher 73.2) and has incredible postseason volume but based on his regular season numbers it may be difficult to justify a considerable bump in his resume to get him into The Hall.   
 
Jon Lester, Madison Bumgarner, and Justin Verlander are still active, so the books on them are not yet officially closed. Verlander seems to have done enough to be elected and his peak October performance will be the cherry on top. Lester has nice peak and career postseason impact numbers, but will his regular season stats get him to that point? Only time will tell. Bumgarner is still young and is listed here more for comparison purposes.
 
Orel Hershiser is a tough case. He has been on the Modern Day Committee ballot and amongst the larger group of pitchers investigated, his career PSI and CSPI numbers rank in the upper third. His peak CPSI5 is in the Top 5 but his PSI5 is relatively one of the lowest (as mentioned previously, due to the larger number of inning pitched). His prime years were before the wildcard era, so his CPSI and potentially his other postseason impact numbers could be even higher. Furthermore, it would be worth comparing Hershiser’s numbers to others from the divisional era. Overall for Orel, his total regular season statistics are on the lighter side (career WAR 56.0) and adding a postseason bump to get him over the top may be a stretch. But at the very least these relative postseason numbers give us a fair objective starting point and gives Hershiser his due for his stellar peak October performance.
 
Lastly, getting back to the Kirby Puckett example, here are some Hall of Fame candidates (for the Modern Day Committee) from the divisional era. For this era, since there were less chances to qualify for the playoffs and less rounds, we will utilize the player’s best four postseasons for the peak stats instead of their best five.
 
Player
WPA
PSI
CPSI
PSI4
CPSI4
Pete Rose
2.64
0.88
7.95
2.11
1.90
Thurman Munson
1.48
1.10
2.00
1.46
1.34
Reggie Jackson
1.18
0.37
3.75
1.97
1.02
Kirby Puckett
1.11
1.02
1.21
1.06
1.17
Steve Garvey
1.05
0.45
2.4
2.03
1.26
Dwight Evans
0.64
0.49
0.80
0.86
0.93
Keith Hernandez
0.48
0.35
0.66
0.62
0.74
Dave Parker
-0.30
-0.25
-0.37
1.04
0.55
 
Italics denotes Hall of Famer
 
For this era, Pete Rose is the gold standard and had an absolutely amazing October career. The next tier of listed players Thurman Munson, Reggie Jackson, Kirby Puckett, and Steve Garvey all seem comparable with each having their own claim. Munson has the highest PSI on this list, even higher than the Hit King. Garvey has a higher PSI and peak numbers than Mr. October himself. Based on these postseason impact stats, both Munson and Garvey deserve an in depth look with their October numbers playing a part of their resume.
 
And just a side note on Garvey, he doesn’t have the WAR numbers that everyone loves, but back in his day (right or wrong) getting base hits were greatly emphasized as opposed to walks and just getting on base. Again, context matters, and for a decade Garvey was amongst the leaders in base hits. Therefore with context in mind and his October numbers, Garvey along with Munson warrants a closer look.
 
In respect to analysis, you can draw your own conclusions based on the data and whether or not a player deserves that postseason extra credit. Again, we realize we’re working with smaller sample sizes, and these numbers should be viewed in that context and used as a comparative analysis. When it comes to Hall of Fame voting, we shouldn’t necessarily use low or even negative PSIs against a player, but rather use postseason impact numbers as a relative gauge to reflect how much a player actually contributed to their team’s success during the postseason. Also we are not advocating the inclusion of a David Freese-type as a viable Hall of Fame candidate, as postseasons should continue to be viewed as extra credit. Generally, it is helpful to have objective numbers like PSI and CPSI to refer to as a starting point and determine if extra credit may be warranted or not. And just like using a player’s career WAR solely doesn’t explain the full story, the same goes for these postseason impact numbers.   
 
The playoff careers of recent October heroes (like Carlos Beltran, Andy Pettitte, Curt Schilling, and Bernie Williams) and also current heroes (like Madison Bumgarner, Justin Turner, and Justin Verlander) are a little nuanced and not as easy to quantify as yesteryear. PSI and CPSI can provide a different perspective and be beneficial to the overall analysis.
 
Special thanks to JJ O’Connor, James Horne, and Dave Fleming. All source data obtained from Baseball-Reference.com. Twitter: @pgups6
 
 
 
 
​
 
 

COMMENTS (45 Comments, most recent shown first)

pgups6
jgk704,

Thanks for the feedback and the Jeter-Papi LI breakdown, that was interesting. I was curious how Rose racked up his WPA and may look into that further myself (peak Rose was a little before my time).

Concerning CPSI, it is a comparative factor that ties in volume and meant to be reviewed along with the PSI numbers as well.

The Delmon Young numbers are actually a good example. For his peak, on a per plate appearance his PSI5 number is comparatively very high 2.51. But his volume CPSI5 number is comparatively pretty low at 0.82.

So in terms of postseason impact, his volume is so low that it doesn't really merit extra credit. DYoung didn't come up big time and time and time again.

Understood that this could be viewed by just looking at a player's WPA along side their PA/IP numbers, but when doing comparative analysis, it's nice to have PSI/CPSI data.
8:33 PM Oct 9th
 
jgf704
Actually, I think I might have handed out too many games... A game needs to be divvied up between hitters and pitchers. In this article...

https://joeposnanski.substack.com/p/the-indis-part-1

... JoePos/Tango advocate giving 4/7 of games to hitters and 3/7 to pitchers. So, for hitters, G = PA/39 * 4/7, or G = PA/68. FWIW, it seems like that article assumes something more like 34.5 PA/game, or perhaps uses innings played somehow... it is not totally clear to me.
6:30 PM Oct 9th
 
jgf704
Paresh:

I think I understand what you are getting at with your CPSI measure, but I think an alternative formulation might be better. WPA is already a cumulative stat (it counts "wins"), so multiplying by plate appearances doesn't really make sense to me.

An alternative might be to do something like Tom Tango has done with his "Independent Won-Loss" records (a.k.a "The Indis"). Basically, you first create "game spaces" for the player in some rational way, and then divvy up the games into wins and losses. WPA = 0 corresponds to a .500 team, you would say:

Wins = 1/2 * G + WPA
Losses = 1/2 * G - WPA

Off the top of my head, I would say count game spaces as PA/39 (i.e. 39 plate appearances per game).

Take Carlos Beltran: 256 post-season plate appearances with 2.26 WPA. So 259/39 = 6.56 games. So Wins = 6.56/2 + 2.26 = 5.54, and Losses = 6.56/2 - 2.26 = 1.02.

Players with alot of post-season plate appearances will end up with lots of "wins", but also lots of "losses".

Bernie Williams (545 PA, 0.88 WPA) ends up with a record of 7.87 wins and 6.11 losses.
6:15 PM Oct 9th
 
pgups6
2012 ALCS MVP Delmon Young's Peak Postseason #s- PSI5 2.51, CPSI5 0.82

On per plate appearance, short of Papi and in the Berkman- Hosmer range. But very low on volume CPSI, less than half of Hosmer who was already considered on the light side.

DYoung made the most of his part-time role!
5:46 PM Oct 9th
 
bhalbleib
MarisFan61:

"I think you would have trouble finding a whole lot of other players with even just 100 post-season plate appearances who had as good a slash-line."

Here's one (in 184 post season Plate Appearances): .337/.397/.627.

I'll let you figure who it is (but I'll give you a hint, he's the Real Mr. October)
2:44 PM Oct 9th
 
ksclacktc
I applied Season Scores to the Post Season Stats and treated it as a season. The method proved very satisfying to me.

The Top Hitters were:

Jeter-310
BernieW-244
Manny-243
Albert-208
Big Papi-195

The Top Pitchers:

Mariano-469
Smoltz-283
Pettitte-233
Schilling-197

12:06 PM Oct 9th
 
MarisFan61
OK -- "as measured by WPA."

That's very different.
11:48 AM Oct 9th
 
jgf704
And, since you asked, of the 305 players with 100+ post-season plate appearances, Jeter's 0.838 OPS is tied for 59th with Delmon Young. Yes, *the* Delmon Young.

His WPA of 0 is tied for 151st.

:)
11:17 AM Oct 9th
 
jgf704
Please don't put words in my mouth. I never wrote anything resembling "Jeter's post-season performance wasn't real good".

Derek Jeter is a future HOFer. He's an all-time great shortstop (#3 all time on Bill's most recent ranking).

I simply wrote that his post-season performance, as measured by WPA, seems to be not as good as his reputation. And then I took a deeper dive into some specific high-leverage plate appearances to show that indeed, Jeter's results were not as good as David Ortiz's (admittedly, a high bar).
7:05 AM Oct 9th
 
MarisFan61
.....BTW, in case some of you doubt that the post-season 'run environment' in general is lower than in the regular season or that post-season batting stats tend to be lower (I know that many people don't realize it; we see indications of it all the time), how about some figures.

Here are the teams' slash lines for the World Series, ALCS, and NLCS for the last few years.
I won't label them.
Trust me. :-)

222-303-388
180-249-302
233-327-384
219-337-385
223-305-328
232-296-382

230-297-467
205-290-393
187-271-294
205-274-347
258-366-515 (nice!)
156-193-299

249-316-404
237-321-371
168-215-329
201-251-283
238-297-426
210-292-301
4:33 AM Oct 9th
 
MarisFan61
GADZOOKS!
The link in the comment below does work as a link, but for some unfathomable reason it puts you to a totally different thread on Reader Posts, a discussion from 10 years ago.

In any event, a definite way for you to get to the actual thing, if you want to, is to go onto Reader Posts:
boards.billjamesonline.com/forumdisplay.php?2-Main-Forum

.....and click on "My data on Win Probability Added, re: top-and-middle-of-order."
3:23 AM Oct 9th
 
MarisFan61
.......I finished the 'study' that I outlined before.
Rather than take up yet more space on here with the details, I put it on Reader Posts:
boards.billjamesonline.com/showthread.php?11901-My-data-on-Win-Probability-Added-re-top-and-middle-of-order-(pls-dont-read)-)&p=189487#post189487
(may not work as a link; in any event should work as a copy/paste into the address bar)

As I said in the previous post about this, it didn't come out nearly as extremely as it looked from the first several sets that I looked at.
There [i]does[/i] seem to be some such effect, to the extent that it looks like there are more cases where a top-of-the-order hitter will have less WPA than a middle-of-the-order hitter with comparable offensive Win Shares, than the other way around. But in most instances, there's 'no hit neither way' -- the relative WPA's seem to follow the offensive Win Share amounts pretty well.
(Disclaimer: My sample size wasn't real big.
Disclaimer #2: The results depended on my judgment.) :-)

Here's the summary:

26 "matched sets"

12 came out neutral -- the WPA's were in line with the offensive Win Shares of the players.

9 of the sets do show a disadvantage for top-of-the-order; with one of those examples, the disadvantage is just slight. In the other 8 sets, it's clear and pronounced.

2 sets show the opposite --i.e. the top-of-order players are relatively [i]higher[/i] on the WPA than the middle-of-order comps.
3 additional sets show a [i]possible/arguable[i] such effect, but slight.

The magnitudes of the disadvantages for the top-of-the-order hitters in those 9 sets, for the most part, are greater than the magnitudes of the "opposite" effects in those 2 + 3 sets.​
3:18 AM Oct 9th
 
MarisFan61
JGF: Someone reading your posts and seeing material like "why Jeter's post-season reputation might exceed his actual post-season performance" would think that his actual post-season performance wasn't real good.

In fact, his actual post-season performance was more than real good.
Do we need to repeat the numbers? Maybe we do.....
(and please nobody forget, post-season offensive numbers tend to be lower than regular season numbers; the 'run environment' is lower -- so these numbers are even better than they look)

158 games
734 plate appearances (conveniently, as was mentioned, just about a "full season")
200 for 650 (yet more convenience -- you can't make this stuff up) :-)
In fact that's a batting avg we can figure in our head! (pretty cool)
It's the same as 4 for 13, which, as I learned from Joe Adcock's stat line for the 1958 World Series (didn't everybody? :-) is .308.

32 doubles, 5 triples, 20 HR, 61 RBI
(the latter probably a big part of why the WPA isn't better)
.308 / .374 / .465

I think you would have trouble finding a whole lot of other players with even just 100 post-season plate appearances who had as good a slash-line.

Did you forget what his post-season record was?
3:05 AM Oct 9th
 
jgf704
A big reason why Jeter's post-season reputation might exceed his actual post-season performance isthat the Yankees generally were very successful in the post-season during his career. Thus, people tend to remember and accentuate the successes.

FWIW, both Jeter and Ortiz failed in their respective highest leverage post-season plate appearances:

* In the bottom of the 8th with the bases loaded and 2 outs and Yankees down by 2 in game 5 of the 2002 ALDS (LI = 6.05), Jeter struck out.

* In the bottom of the 9th with the bases loaded and 2 outs in a tie game in game 4 of the 2004 ALCS (LI = 6.39), Ortiz popped out to 2B.
7:51 PM Oct 8th
 
jgf704
FWIW, I used BB-Ref to download all post season plate appearances by Derek Jeter and David Ortiz.

Jeter had 22 plate appearances with LI >= 3. He generated positive WPA in 7 of these (totaling +0.90), and negative WPA in 15 (totaling -1.53).

Ortiz had 13 plate appearances with LI >= 3. He generated positive WPA in 8 of these, totaling +1.70, and negative WPA in 5 (totaling -0.54).

As you can see, on average, Ortiz's positive WPA events in these ultra-clutch situations (+0.21 per event) were "better" than Jeter's (+0.13 per event), though, on average, their negative WPA events were about the same (-0.10 per event for Jeter, -0.11 per event for Ortiz).

Jeter's 7 positive WPA events: 2 1B, 1 2B, 3 BB, 1 FC (no walk-offs)
Ortiz's 8 positive WPA events: 2 1B, 1 2B, 1 HR, 4 BB (2 walk-off hits: a single and a HR)


7:44 PM Oct 8th
 
jgf704
* On the one hand, it would make sense to me that, all things being equal, WPA might be higher for middle-of-the-order hitters, mainly because I would expect that middle-of-the-order hitters would see a higher average leverage index (LI) for their plate appearances than would top-of-the-order hitters.

* On the other hand, don't forget the obvious reason that Jeter's WPA might be low: in sum, he performed basically average (WPA pretty close to zero). And so maybe our perceptions and memories of his performance aren't completely reliable.
6:46 PM Oct 8th
 
MarisFan61
Progress report: Looking like there is such a tendency (i.e. WPA giving advantage to middle-of-order vs top) but not as uniform as it looked before.
It's not "shooting fish in a barrel."
I won't be doing posts on every twist and turn -- none till it's done -- just wanted to moderate it right away.
3:15 PM Oct 8th
 
MarisFan61
About the work I did, and am going to do:

There was a mistake there (one that I know of), but it doesn't change the tentative conclusion because of the magnitude of what I was seeing.
Anyway I won't make that mistake when I do it for real.

The mistake was that I was using each player's TOTAL Win Shares for the year. To be fair to WPA, we need to leave out fielding Win Shares, because, as we said before, WPA doesn't (yet) deal with fielding.
And actually to be totally fair we'd need also to leave out whatever base-running amounts are in the Win Shares, but I wouldn't know how to do that, and anyway those amounts are presumably small enough that they don't mess up the results either -- provided that the magnitude of what is seen continues to be so great.
12:02 PM Oct 8th
 
pgups6
Hal Smith- PSI 8.4, CPSI 0.05
Lou Gehrig- PSI 1.5, CPSI 3.5 (PSI4 2.6, CPSI4 1.8)

Gehrig's peak numbers better than the Divisional Era players listed, and career numbers pretty much at the top as well.

11:07 AM Oct 8th
 
DaveFleming
Ksclacktc mentioned 'Championship Win Probability Added), which is a weird stat. It tends to favor relief pitchers (Rivera is #1, and Fingers is #3).

And it's going to take a LOT of work to convince me that Hal Smith added more probability in the eight plays involving him during the 1960 World Series (.648 CWPA) than Lou Gehrig added in his 153 plays.

But it's a quirky stat, and it probably tells us something. Thanks for the nod to it, Ksclacktc.

9:32 AM Oct 8th
 
pgups6
Re: bearbyz

I purposely left off ARod, Manny, Bonds, and Rocket to avoid the whole nauseating PEDs topic. But ARod's numbers are real interesting:

WPA 1.00
PSI 0.30
CPSI 3.3
PSI5 2.9
CPSI5 3.8

His peak numbers are better than Pujols and pretty much at Papi level (ARod had a monster 2009 postseason). But his career numbers would be on the lower part of the list. None of the other players looked into had such a dramatic swing.
8:19 AM Oct 8th
 
ksclacktc
I appreciate the effort!!!

I think you should take a look at cWPA.


www.thebaseballgauge.com/post.php?tab=players_at

www.thebaseballgauge.com/glossary.php?tab=cWPA
7:10 AM Oct 8th
 
MarisFan61
.....Looked a little more, and it seems to be a VERY pronounced tendency.
And BTW, it's extreme fun looking for it, including because it's so easy to find and to see. The "reinforcement" is quick, and extreme.

Dave -- Great job on having the idea.

Within the next few days (maybe later today, not sure), I'll present some substantial data on it, unless you guys do it first, which I invite you to do. As I said, it's pretty fun. Shooting fish in the proverbial barrel.

BTW, because it's so pronounced, you don't even need to be as restrictive on the matches as what I said. Like, don't worry if the differences in the Win Share amounts are a bit more than 1, provided it's the top of the order guy who has MORE -- because chances are he'll still have a lower WPA.
5:38 AM Oct 8th
 
MarisFan61
......Couldn't resist looking at a few teams to see if there's some clear 'preliminary indication.'

THERE IS.

It looks like that's it.
I don't have enough to say it as an answer yet, but it looks like, borrowing Dave's wording, WPA tends to favor middle-of-the-order hitters over top-of-the-order guys, and quite strongly.
5:19 AM Oct 8th
 
MarisFan61
......Here's a way you can do it (and I will, unless one of you guys wants to do it first), and I don't think it has any flies in the ointment.
Forget post-season, look at it for the regular season, and it ought to be valid about the post-season too:
Find players on the same team who had the same amount of Win Shares for the year (or let's say within 1), one of them top-of-the-order, the other middle-of-the-order, also make sure they had about the same amounts of plate appearances, and compare their WPA's.

If there's some pretty consistent pattern on the WPA's from that, I think that would be convincing, whatever it shows -- unless you guys can see any issue about it. (I don't.)

BTW, if you feel like doing the same thing but using "WAR" instead of Win Shares, I'm not sure I'd be convinced (really), but I suspect they'd both show the same.
5:09 AM Oct 8th
 
MarisFan61
About this from Dave Fleming:
"....does anyone know if WPA tends to favor middle-of-the-order hitters over top-of-the-order guys?
I could see a scenario where it might, particularly in a non-DH context. A leadoff hitter in the NL would have fewer high leverage at-bats than, say a #3 hitter on an AL team. I can't imagine that WPA makes an adjustment for that, but I haven't looked into the metric in a long time, and the details of it are fuzzy."

That sounds very possible (doesn't it?), and it could be why Jeter shows so mediocrely, at least past of why.

Main thing I want to say, though (dam, that word has become poisoned, hasn't it, but it's still a word) :-)
....is that I wanted to do a little study on it, but couldn't, as I'll explain. A couple of things that I saw in the process of it suggested support for that idea, but most basically, I couldn't do it.

I tried two different things.
My first idea was to look for "matched pairs" from the position player charts in this article:
-- Take two players, one of them top-of-the-order, the other middle-of-the-order, who are very close to each other in the rankings here....
-- Look to see how their actual post-season stats compare to each other.
-- If Dave's idea is so, what we'd expect to find is that even though the two players are very close in the rankings here, the top-of-the-order players had better actual post-season stats.

Why I couldn't do it:
There aren't enough top-of-the-order players in those charts.

IS THERE anyone besides Rose and Jeter? I think not really.

I'd suggest that this fact in itself gives some support to the idea, but it's not much.

So, I figured I'd try something else: Look at some post-season series from recent years, again look for matched pairs of players, this time looking for players who had very similar OPS and its components, and see how their WPA for the series compared. I was going to look only at 7-game series (to get at-bats amounts as large as possible), and only at players with high number of at-bats (it would be stupid to bother with pairs of players who had similar numbers in 5 at-bats), and only at pairs with similar numbers of at-bats, and only players on the same team.

I thought all those "similars" were important to do, but that made it impossible to find enough pairs on this either. I looked at every 7-game post-season series since 2003 (why starting then? I can't stomach the few years before that) and found only one pair that I thought met enough criteria. That one example -- Allen Craig and David Freese, 2011 World Series -- did show WPA's in line with what I was looking for: Freese, who mostly hit 6th, had a much high WPA than Craig, who mostly hit 2nd (1.09 vs. 0.18). But, besides that it's just one pair, it's not even a great pair, for various reasons.

I know that there could be infinite other ways to try to get at what I was trying to get at. Maybe one of you wants to give it a try, or maybe just to say how you think one might do it.
I think it would be very hard to come up with a way that wouldn't have built-in problems. I was trying to minimize them with all those requirements.
2:25 AM Oct 8th
 
MarisFan61
I can't let this thing go -- and if you guys think any of what I'll be saying here is wrong, please don't let it go either. :-)

Dan wrote: On Jeter: his postseason stats look basically like his regular season stats. Which makes sense, because as you play more games, you'll move toward your established level of play.

My Cliff's Notes on that is: NO. :-)

I'm confident that if you looked at the post-season records of 10 other randomly selected players with long careers who also played very many post-season games, you'd be unlikely to find that more than 1 or 2 of them (if any) had post-season hitting records that were ANYWHERE CLOSE to their regular season records.

I would say the absolute opposite of what Dan said: The best chance for a player to have his post-season hitting record be close to his regular season record is for him to have as few post-season games as possible -- because hitting numbers in the post-season tend to be lower than in the regular season.
(2 reasons: colder weather, better pitching)

The remarkableness of Jeter's post-season hitting record, notwithstanding how well known it is, is commonly underestimated because of a lack of recognition of that. I think the most common impression is, 'yeah, he hit the same in the post-season as in the regular season.'
That's not true.
He hit far better in the post-season -- if you take proper account of the run environment.​
1:26 AM Oct 8th
 
bearbyz
How does Alex Rodriquez come out in this formula as he always had a reputation as a poor post season player, but one article I say he had a high WPA?
10:19 PM Oct 7th
 
thedanholmes
Maris,

Thanks, Gotcha

On Jeter: his postseason stats look basically like his regular season stats. Which makes sense, because as you play more games, you'll move toward your established level of play.

I remember Jeter different in the postseason. Sure, I recall the hit that earned him "Mr. November" status, and "The Flip," and one or two more key hits. But, I also remember him popping the ball up or striking out too many times with men on base. He had a few "signature" plays, but otherwise, he wasn't a consistent PS monster, like say George Brett or Will Clark or Marquis Grissom. He was basically Derek Jeter.
9:53 PM Oct 7th
 
MarisFan61
Dan: Just to be clear: My comment about fielding not being in WPA wasn't related to what I said about Jeter. They were separate.
7:44 PM Oct 7th
 
BarryBondsFan25
This was a good read. Thank you Mr. Gupta.
7:15 PM Oct 7th
 
pgups6
And yes, I do realize WAR is used in the article, but it's meant more for a general gauge, not end all be all.
7:09 PM Oct 7th
 
pgups6
And yes, I do realize WAR is used in the article, but it's meant more for a general gauge, not end all be all.
7:09 PM Oct 7th
 
pgups6
Re: MarisFan61

Of course! The bar stool discussions is one of my favorite parts. Going on a soapbox for a second...but I feel like the community nowadays is relying too much on WAR as the end all be all and there's no room for discussion anymore. But there's more than one way to value a player.

Anyways, I'm glad with this topic it turned out to be multiple factors/numbers and it leaves so room for thought.
7:01 PM Oct 7th
 
thedanholmes
I'm not sure adding fielding to the measurement would help Derek Jeter. Outside of "The Flip", Jeter's defense in the postseason was as mediocre (or worse) as it was in the regular season. Weak to average arm, bad range to his left, a mediocre first step. Jeter was basically a "guess fielder" --- he took a step in the direction where he guessed the ball might go, to get a head start. He had a long throwing motion, and if my opinion doesn't sway you, look at the numbers. Jeter's career Win Shares from fielding rates 40th among players who made shortstop their home. That despite playing more innings at the position than anyone except Cal Jr. In fact, Jeter's Defensive Win Shares are about the same as those of Jhonny Peralta, who played about 65 percent as many innings. No one (who was objective) ever watched Jeter and thought he was a good shortstop.

Adding fielding to the metrics of postseason evaluation will hurt Derek Jeter.
5:44 PM Oct 7th
 
MarisFan61
Also: I did realize it was possible that you did realize Chipper is already in and that you just didn't want to make that sentence more complicated.
5:32 PM Oct 7th
 
MarisFan61
PG: So, you're the author!!
I much appreciate your reading the Comments and participating in the discussion. I'm sure others do too.
It is too rare that the authors participate. When they don't, we don't even know if they have cared to look at the discussion.

I realized that you didn't necessarily mean that non-metric aspects don't play a role any more, but, if we were to take the grammar and tense literally, which perhaps I do too often :-) (sometimes those do indicate what a writer meant!), it does seem to indicate that.
In any event, it seems like you think it counts a lot less than it used to, and I don't think that's so -- but maybe you didn't mean that either!

BTW: No need to reply further to this, and in general there's no need to reply to every little thing.
It's great just to know that you're on here.
5:31 PM Oct 7th
 
pgups6
Re: MarisFan61

Yes, WPA does not take into account fielding, nor base running as well. So for Jeter, it does not include "The Flip" nor his exceptional ability to take the extra base. I included that info in the initial write-up but the article was already long enough.

Honestly, I was surprised at the Jeter numbers as well. Time and time again, he came up with the "clutch" hits. My guess (and it is only a guess) is that he often came up with the game tying hits or go ahead runs but he didn't often break the game open with a 3 run homer or bases loaded double. Again, just a guess and if someone else has a better explanation please chime in.

I think with all stats (not just this one), we need to be mindful of the context and where they came from.

The article doesn't indicate that non-metrics aspects don't play a role in HOF voting, it indicates that analytics are coming more and more into play, which it is.

And yes, I do realize Chipper is HOF and listed him as such in the tables, but in the text I should have stated "Both Chipper Jone and Derek Jeter is and will rightfully be first ballot Hall of Famers." But again, the article was long enough.
3:10 PM Oct 7th
 
pgups6
Re: MarisFan61

Yes, WPA does not take into account fielding, nor base running as well. So for Jeter, it does not include "The Flip" nor his exceptional ability to take the extra base. I included that info in the initial write-up but the article was already long enough.

Honestly, I was surprised at the Jeter numbers as well. Time and time again, he came up with the "clutch" hits. My guess (and it is only a guess) is that he often came up with the game tying hits or go ahead runs but he didn't often break the game open with a 3 run homer or bases loaded double. Again, just a guess and if someone else has a better explanation please chime in.

I think with all stats (not just this one), we need to be mindful of the context and where they came from.

The article doesn't indicate that non-metrics aspects don't play a role in HOF voting, it indicates that analytics are coming more and more into play, which it is.

And yes, I do realize Chipper is HOF and listed him as such in the tables, but in the text I should have stated "Both Chipper Jone and Derek Jeter is and will rightfully be first ballot Hall of Famers." But again, the article was long enough.
3:10 PM Oct 7th
 
DaveFleming
Building on something Maris alluded to....does anyone know if WPA tends to favor middle-of-the-order hitters over top-of-the-order guys?

I could see a scenario where it might, particularly in a non-DH context. A leadoff hitter in the NL would have fewer high leverage at-bats than, say a #3 hitter on an AL team. I can't imagine that WPA makes an adjustment for that, but I haven't looked into the metric in a long time, and the details of it are fuzzy.
3:03 PM Oct 7th
 
pgups6
Re: DaveFleming

Thank you Dave. Concerning PSI5 and CPSI5 (peak stat), yes the numbers did tend to sway a little in favor of those with more opportunities. And that's why it's especially important to differentiate the divisional playoff era versus the wildcard era. Where the analysis can get a little interesting is when players with less or the same amount of postseason playing time have higher peak numbers than their counterparts, like the Bernie "vs" Jeter and Glavine "vs" Smoltz examples listed.
2:52 PM Oct 7th
 
pgups6
Re: Mike137

Yes, sure those numbers can be looked into as well but they are more team dependent and do not convey overall playing time. And I do think the situational aspect of WPA is important especially in Oct when a player's ability to capitalize on a high leverage situation can make or break a championship run.
2:43 PM Oct 7th
 
MarisFan61
Four things, the first three theoretical; the other, specific.

For a thing like "post-season impact," at least how I think of it, it's mostly about individual games -- and to a lesser extent but still greater than overall-lumped-together-performance (which this is), crucial individual plays. A method that looks at an overall total for a player doesn't address those; it averages it away. Sure, you could say that if (say) Don Larsen was in 10 other World Series games and blew them (he didn't; just saying as an example), how could we laud his "post-season impact" because of one great game (which also happened to be pivotal) if overall he was bad? Well, to me such a player would still score very high on post-season impact. I suspect most people here would disagree, but for something like Hall of Fame voting, which is referenced here, I suspect that the view I'm expressing would be common: emphasis on individual games (and individual series), not a lumped total performance.

This method doesn't distinguish one game from another. Of course there's no such thing as an un-meaningful post-season game, but some are most critical, and conceptually many of us give those games the most emphasis.

Finally (as to the theoretical points), there's a big basic problem with the method that isn't the author's fault; it's the fault of WPA as it currently stands (if I understand right):
It doesn't look at fielding at all.



My specific gripe is (please pardon) :-) Derek Jeter.
Most of you probably know that his overall post-season stats are superb. How he comes out so relatively mediocrely on this, I don't know; not that I don't trust the result -- I do -- but it's hard to square with the reputation, memories, and, significantly, the stats.
Of course one could counter.....well, you could counter in many ways, including that just one bad result (if it be such) doesn't incriminate a system, and I would agree, but my only pointing out Jeter doesn't mean there's just one bad result; it's just one that jumped out immediately.

I agree with the article that looking at WPA in this way is "a fun tool," and I'd agree that it's probably sort of somewhat of "a general barometer," provided we stay mindful of "sort-of-somewhat-of."

--------------

Two stray points:
-- The article is mistaken in indicating that non-metric aspects don't still play a major role in Hall of Fame voting.
-- It looks like the writer didn't know that Chipper Jones is already in the Hall of Fame.
12:33 PM Oct 7th
 
DaveFleming
One of the first things I look for in any kind of new metric is whether or not it passes the test of past interpretations - that is, whether or not the conclusions of the new metrics generally dovetail with earlier understandings.

In that regards, these metrics holds up satisfyingly: we'd expect the likes of Ortiz and Beltran and Schilling to do well by a metric measuring postseason impacts, and those players rate out as excellent postseason contributors.

I am slightly less convinced by the Career Postseason Impact-5 (CPSI5) version than the career version, just on the basis that players on better teams will have more postseasons to choose their best five years from, but that's a small preference.

Lastly, I enjoyed the broader frame for this article, which a) asks what is a method by which we can reasonably evaluate postseason performances, and b) uses that to determine which players we could legitimately give some extra credit for playoff performance...which players merit an extra 'tick' in their credentials for, say, the Hall-of-Fame.

Thanks, Paresh!
11:32 AM Oct 7th
 
Mike137
Why not just use RBI, and Runs scored or allowed? Much simpler and less situationaly dependent.
10:03 AM Oct 7th
 
 
©2024 Be Jolly, Inc. All Rights Reserved.|Powered by Sports Info Solutions|Terms & Conditions|Privacy Policy