Announcement

Collapse
No announcement yet.

PCA Request Line

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • PCA Request Line

    This is your sbaermetric DJ coming to you LIVE from the monotone jungle that is suburbia, USA....I've got all the hits...and the errors, and the strikeouts too right here at my fingertips! Send in your calls to 1-800-PCA-HITZ (legal disclaimer: this phone number does not exist!) or post them here and i'll do my best to lay down the baseball trax!!

    Seriously though...anyone want info from the PCA database...I can do my best to provide a critical overview of my findings to answer any question you can chuck at me.

  • #2
    Um, was Jim Kaat really the best defensive pitcher?
    Was Connie Mack the best manager?
    Where do you rank Blyleven among pitchers?
    How do you adjust rankings for 19th century players?
    How do you trade off longevity/dominance? (Clemens/Pedro, Wilhelm/Rivera)
    How do you evaluate catcher defense before 1965?
    Does Hornsby's fielding significantly downgrade him?
    A complete list of your PCA Gold Glove awards
    Mythical SF Chronicle scouting report: "That Jeff runs like a deer. Unfortunately, he also hits AND throws like one." I am Venus DeMilo - NO ARM! I can play like a big leaguer, I can field like Luzinski, run like Lombardi. The secret to managing is keeping the ones who hate you away from the undecided ones. I am a triumph of quantity over quality. I'm almost useful, every village needs an idiot.
    Good traders: MadHatter(2), BoofBonser26, StormSurge

    Comment


    • #3
      gonna move this to the Statistics, Analysis and Sabermetrics Forum

      Comment


      • #4
        eek!

        Let's take these one at a time...

        Was Jim Kaat really tghe best defensive pitcher?

        I am missing some data...I son't know for isntance how many of Kaat's assists were pickoffs. this makes a difference in how valuable he might be. But here is what PCA suggests about Kaat. I have him as being relatively pedestrian defensively. This is one area where I think my defensive analysis could use a little bit of tweaking...I believe I am successfully figuring out how much value each pitching staff on the whole has in preventing runs. But I think there may be a better way to measure individual pitchers' defense than to divide that value up based on claim points. I think a claim point per playing time (since we have exact measurements of how many defensive innings each pitcher pitched) approach would be a little more effetive. Nonetheless good defensive pitchers generally come out on top and bad defensive pitchers come out on the bottom...

        The top ten fielding pitchers by PCA in terms of wins/162 games:
        Code:
        First	Last    	Ps	W/Y	Full G (9-innings)
        John	Clarkson	SP	3.74	575
        Carl	Mays    	SP	3.21	459
        Freddie	Fitzsimmons	SP	2.69	471
        Hooks	Dauss   	SP	2.66	405
        Cy	Young   	SP	2.41	841
        Charley	Radbourn	SP	2.4	630
        Charlie	Buffinton	SP	2.26	606
        Greg	Maddux  	SP	2.22	760
        Harry	Howell  	SP	2.16	461
        Jack	Quinn   	SP	2.14	490
        I know of 8 of those 10 pitchers directly and they all had good defensive reputations...in fact great reputations in most cases. Anyway I can improve pitchings' individual fielding ratings probably...that's not an area I focused on this time around.

        Was Connie Mack the best manager...OK I think the way I would approach that question is compare a team's pythagorean win total to its actual win total and figure out which managers consistantly got more wins for their production...this isn't a foolproof way to go about it, but managerial skill is something that's hard to put into numbers...I'll build a query to answer this question the best way I know how and get back to you.

        Where do I rank Blyleven amongst the pitchers all time?

        In terms of total wins created, he is 19th all time...however this probably isn't the best way to rank players by greatness. I have come up with an index to rate greatness that combines the player's total wins created, a prorated total wins created over the course of a standard full career (this is to measure career scoring efficiency on something liek the same scale as career wins), and the total number of wins a player scored above and only above his career scoring average in individual seasons (if a player's average scoring rate over his career is 6 wins per 162 games, and in a full 162 game season he scored 11 wins, he'd get credit for 5 peak wins...whereas if he scored 5 wins in hat 162 game season, it would be entered as a zero)...this is designed to measure a player's capacity for great performances above and beyond what his scoring efficiency already tells us...

        And by this ranking method, Blyleven drops to 43rd all time.

        Presently I make no adjustment for the 19th century except to say that the entire system is dynamic...allowing me to see for instance that although pitchers threw many more innings, they were less important (per inning) to the prevention of runs because of the way in which they pitched (hence 19th century guys aren't getting too much credit simply because they pitched insane numbers of innings).

        I also dynamically era adjust fielding ratings so that the average scoring rate at each position in any given season is the same for the purpose of comparing fielding skill.

        How do I trade off longevity/dominance?

        In the past here at BBF and over at MLBC I've been relying on the principle of the supermargin to give players enough credit for peak performance that I could count up the wins and rank players based on total win counts. I have come to the conclusion recently that even though the supermargin accurately depicts peak performance...it will always be dwarfed by the mass of long careers. Which is what propelled me to invent my "greatness" index as described above. Using the greatness index, Wilhelm still beats big Mo...but Rivera moves up to third on the all time reliever list...and using the greatness index Pedro Martinez is third behind only Greg Maddux and Walter Johnson.

        How do I evaluate catchers prior to 1965?

        Even without SB/CS data, there are still catcher elements that make fielding analysis of catchers possible and even with good solid accuracy. I still have complete records of PO, A, E, DP, and PB...and I still have team strikeout records to subtract out from PO score to find non-K putouts...and I have offensive records of sac hits and steals so I can give the catchers a league average rate of runs allowed by stolen base and sac hit (I realize this isn't perfect...but it's the bes we can do with the data we have...and you can still see differences in catching skill because even with constant stolen base rates, the assist count changes as does the non-K putout count)...it works fairly well although if you'd like to confirm that, make some requests for catchers whose careers spanned the period where that SB/CS data suddenly becomes available...

        DOes Hornsby's fielding significantly dongrade him...

        Well...he's a below average fielder by PCA...scored a C- letter grade for his fielding, but I don't think it's so bad that it relaly has a massive impact on his overall rating.

        He earned about 34 dewfensive wins while the great defensive second basemen were earning as many as 75 defensive wins (72 wins for Bid McPhee)...so there is a difference...but it's rare to get a good defensive second baseman who can also hit...Eddie Collins...that's about it...and Collins wasn't as far ahead of Hornsby on defense as McPhee.

        A complete list of PCA gold glovers...

        I have a file made with all of the PCA gold glovers in each year and league (some of these may change as the new data gets processed obviously)....

        I make that file available to anyone who wants it...can't really post it all here because it would take al ot of work to format it correctly but if you want to leave an e-mail address I'll send you this file...it also contains a complete listing of defensive, pitching, and offensive letter grades based on mean/standard deviation of era adjusted scoring rates...and a complete listing of League MVPs and Cy Youngs.

        Comment


        • #5
          Matt

          Fundemental flaw in that P list

          The eye opener is that pre 1921, may bunts were laid down, thus more Opportunities for Pitchers to bolster stats

          I love Cy Young, but he was mearly above average, as was Harry Howell and Jack Quinn

          Maybe you should try and find a way to negate the effect of small ball and MANY grounders/Bunts/Bunts for hits during the deadball and 19th century

          One guy I think should be on the list is the man with the best pickoff mover of all time

          Terry Mulholland

          Comment


          • #6
            Fair point Potato...

            The fact that there were more bunt-for-hits, sac bunts, and groundballs to get the pitcher involved in would however mean that *all* pitchers should expect to have a larger share of the plays made by infielders...the way I divide up a unit score (units being pitching, catching, infield defense including the defense of the pitchers, and outfield defense) amongst the positions in that unit is to determine the league average ratio of plays made by each position on the unit to the plays made by the unit itself for each year and league...and use those to rate the magnitude of each position on each team in each year and league...if the pitchers were getting more bunts...then pitchers as a group would be scoring more plays per infield play...so I don't think I see that as an issue.

            Comment


            • #7
              Originally posted by SABR Matt
              This is your sbaermetric DJ coming to you LIVE from the monotone jungle that is suburbia, USA....I've got all the hits...and the errors, and the strikeouts too right here at my fingertips! Send in your calls to 1-800-PCA-HITZ (legal disclaimer: this phone number does not exist!) or post them here and i'll do my best to lay down the baseball trax!!

              Seriously though...anyone want info from the PCA database...I can do my best to provide a critical overview of my findings to answer any question you can chuck at me.
              Just how good was Barry Bonds for the last three years?
              I'm NickG, and I approve this message.

              Home page

              "In God we trust, all else must have data."
              -- Mets pitching coach Rick Peterson

              Comment


              • #8
                Originally posted by NickG
                Just how good was Barry Bonds for the last three years?
                Extremely.

                The Giants were sabermetrically worth 369.9 Wins as a team in the four years from 2001-2004. Barry Bonds was worth about 85.1 of those wins! That's the difference between a team sabermetrically winning 57.1% of its games and a team winning 43.9% of its games! Barry Bonds made a team that was not even remotely competitive into a team that nearly won a world series and appeared in three of four post-seasons.

                Those 85.1 Wins occured while Bonds consumed 1105 batting outs out of the team's approximately 17,400 total batting outs...roughly 6.4% of his team's chances to score! His 85.1 Wins per 1105 batting outs (now thi includes his defensive contribution as well...those aren't just batting wins...nonetheless) are a rate of about 37.4 Wins per full season (486 Batting Outs = Full Season)...which is third only to Ted Williams and Babe Ruth for the greatest scoring average in 4 consecutive seasons. That is an extraordinary accomplishment...clouded though it may be by the likelihood that his good physical health and increased strength were related to his use of performance enhancing drugs. (no...steroids cannot make you into a walk taking machine nor can they raise your batting average...they give you health and raw power and nothing more)

                BTW Bonds won the PCA National League MVP in all four of those seasons and the runner up was generally never close.

                Comment


                • #9
                  Incredible. Every time I see Bonds's dominance measured, I'm simply amazed.

                  I would even dispute that he was healthy - I think Stan (not Victor) Conte did a hell of a job keeping him in playing order, given the condition of his hamstrings and knees, and that the surgeries he is going through now are an almost direct result of the past few years. This was bound to happen sooner or later, although I'm surprised it's taking as long as it is.
                  I'm NickG, and I approve this message.

                  Home page

                  "In God we trust, all else must have data."
                  -- Mets pitching coach Rick Peterson

                  Comment


                  • #10
                    Originally posted by SABR Matt
                    Was Jim Kaat really tghe best defensive pitcher?

                    I am missing some data...I son't know for isntance how many of Kaat's assists were pickoffs. this makes a difference in how valuable he might be. But here is what PCA suggests about Kaat. I have him as being relatively pedestrian defensively.
                    I knew it!

                    The top ten fielding pitchers by PCA in terms of wins/162 games:
                    Code:
                    First	Last    	Ps	W/Y	Full G (9-innings)
                    John	Clarkson	SP	3.74	575
                    Carl	Mays    	SP	3.21	459
                    Freddie	Fitzsimmons	SP	2.69	471
                    Hooks	Dauss   	SP	2.66	405
                    Cy	Young   	SP	2.41	841
                    Charley	Radbourn	SP	2.4	630
                    Charlie	Buffinton	SP	2.26	606
                    Greg	Maddux  	SP	2.22	760
                    Harry	Howell  	SP	2.16	461
                    Jack	Quinn   	SP	2.14	490
                    Does seem strange that almost all are old-timers

                    <Was Connie Mack the best manager...OK I think the way I would approach that question is compare a team's pythagorean win total to its actual win total and figure out which managers consistantly got more wins for their production...this isn't a foolproof way to go about it, but managerial skill is something that's hard to put into numbers...I'll build a query to answer this question the best way I know how and get back to you. >

                    Sounds good, thanks!

                    <Where do I rank Blyleven amongst the pitchers all time?

                    In terms of total wins created, he is 19th all time...however this probably isn't the best way to rank players by greatness. I have come up with an index to rate greatness that combines the player's total wins created, a prorated total wins created over the course of a standard full career (this is to measure career scoring efficiency on something liek the same scale as career wins), and the total number of wins a player scored above and only above his career scoring average in individual seasons (if a player's average scoring rate over his career is 6 wins per 162 games, and in a full 162 game season he scored 11 wins, he'd get credit for 5 peak wins...whereas if he scored 5 wins in hat 162 game season, it would be entered as a zero)...this is designed to measure a player's capacity for great performances above and beyond what his scoring efficiency already tells us...

                    And by this ranking method, Blyleven drops to 43rd all time.>

                    Wait'll I tell ElHalo!

                    <How do I evaluate catchers prior to 1965?

                    Even without SB/CS data, there are still catcher elements that make fielding analysis of catchers possible and even with good solid accuracy. I still have complete records of PO, A, E, DP, and PB...and I still have team strikeout records to subtract out from PO score to find non-K putouts...and I have offensive records of sac hits and steals so I can give the catchers a league average rate of runs allowed by stolen base and sac hit (I realize this isn't perfect...but it's the bes we can do with the data we have...and you can still see differences in catching skill because even with constant stolen base rates, the assist count changes as does the non-K putout count)...it works fairly well although if you'd like to confirm that, make some requests for catchers whose careers spanned the period where that SB/CS data suddenly becomes available... >

                    Ooh, like Sherm Lollar, Paul Richards, maybe Phil Masi?

                    <DOes Hornsby's fielding significantly dongrade him...

                    Well...he's a below average fielder by PCA...scored a C- letter grade for his fielding, but I don't think it's so bad that it relaly has a massive impact on his overall rating.

                    He earned about 34 dewfensive wins while the great defensive second basemen were earning as many as 75 defensive wins (72 wins for Bid McPhee)...so there is a difference...but it's rare to get a good defensive second baseman who can also hit...Eddie Collins...that's about it...and Collins wasn't as far ahead of Hornsby on defense as McPhee.>

                    Boy, do the History viewers need to hear this

                    <A complete list of PCA gold glovers...

                    I have a file made with all of the PCA gold glovers in each year and league (some of these may change as the new data gets processed obviously)....

                    I make that file available to anyone who wants it...can't really post it all here because it would take al ot of work to format it correctly but if you want to leave an e-mail address I'll send you this file...it also contains a complete listing of defensive, pitching, and offensive letter grades based on mean/standard deviation of era adjusted scoring rates...and a complete listing of League MVPs and Cy Youngs.>

                    I'll PM you my e-mail addy, mucho gracias
                    Mythical SF Chronicle scouting report: "That Jeff runs like a deer. Unfortunately, he also hits AND throws like one." I am Venus DeMilo - NO ARM! I can play like a big leaguer, I can field like Luzinski, run like Lombardi. The secret to managing is keeping the ones who hate you away from the undecided ones. I am a triumph of quantity over quality. I'm almost useful, every village needs an idiot.
                    Good traders: MadHatter(2), BoofBonser26, StormSurge

                    Comment


                    • #11
                      Sherm Lollar's season by season era adjusted defensive wins:
                      Code:
                      Age	Yr	Full G	ADWC
                      21	1946	15	0.07
                      22	1947	8	0.04
                      23	1948	7	0.04
                      24	1949	76	0.40
                      25	1950	99	0.29
                      26	1951	82	0.54
                      27	1952	103	1.66
                      28	1953	91	1.44
                      29	1954	77	1.23
                      30	1955	125	1.87
                      31	1956	126	0.61
                      32	1957	93	1.69
                      33	1958	111	2.45
                      34	1959	129	2.38
                      35	1960	108	1.77
                      36	1961	86	0.94
                      37	1962	53	0.32
                      38	1963	16	0.18
                      Career W/162: 2.07, 50th of 148 eligible catchers, C+ Fielding Rating

                      Paul Richards' Catching Wins:
                      Code:
                      Age	Yr	PRG	ADWC
                      23	1932	5	0.07
                      24	1933	18	0.22
                      25	1934	21	0.36
                      26	1935	72	0.54
                      34	1943	98	5.48
                      35	1944	90	1.98
                      36	1945	76	1.84
                      37	1946	48	0.68
                      Wow...we have an obviously fluky 1943 here but even without it, he was an excellent back-up catcher...never got to player regularly though...

                      You managed to pick guys who did NOT straddle the line between the 1960s and the 1970s when catching data because avaiable...LOL Pick again Ruth.

                      Comment


                      • #12
                        I'd like to know where Sean Casey stacks up in PCA from 1998-2004 among NL first basemen.
                        "It is a simple matter to erect a Hall of Fame, but difficult to select the tenants." -- Ken Smith
                        "I am led to suspect that some of the electorate is very dumb." -- Henry P. Edwards
                        "You have a Hall of Fame to put people in, not keep people out." -- Brian Kenny
                        "There's no such thing as a perfect ballot." -- Jay Jaffe

                        Comment


                        • #13
                          Matt,

                          Couple of questions of evaluation of catcher's.

                          From The Ferrell Brothers of Baseball, page 163, from the 1935 chapter.

                          "New York - in town for a four-games series that opened on August 5 - pounded six Boston hurlers in a 10-2 abbreviated game that was halted after five innings due to a heavy rain. The fifth inning was farcical. 'With another downpour imminent,' noted The Sporting News, 'the Yanks endeavored to make three outs, but the Sox endeavored just as much not to retire them. During this period the Yanks stole four bases.' The Boston Globe used a photograph of Rick Ferrell purposely juggling the ball at home plate, and Myril Hoag, who stole both second and home in the inning, waiting just short of home plate for Rick to tag him out. 'Finally Ferrell had his way and they gave Hoag a stolen base, of all things!' Both managers - Cronin and McCarthy - were fined $100 for their stalling tactics."

                          So, Matt, how would your system allow for this? I realize you do not have catchers' SB caught/attempt for that period, but if you had, how would you adjust for this? Held against Ferrell or not?

                          How also do you allow for doubleheaders in the period?

                          Page 14 of the same book. Also 1935.

                          "Boston dropped the next two games to St. Louis before traveling to Philadelphia where Cronin started Grove and Ferrell in a doubleheader on July 27. The first game was a 7-6 Philadelphia victory in 15 innings. Grove went the distance, allowing 21 hits and 6 walks....Johnny Marcum, the Athletics' ace, faced Wes in the second game, and Mack's only lineup change was a different catcher. Rick Ferrell, who caught all 24 innings, handled his brother's 2-0 three hit shutout. Wes was dominant, allowing only an infield single through the first seven innings."

                          I am wondering because most of the anecdotal data from the era claimed Rick Ferrell to be an incredible fielding catcher.

                          TSN in the Spring of 1933: "Ferrell's catching form in near perfection. There isn't a better thrower in either big league, and at bat, he's a line drive batter of the most dangerous type."

                          Ben Chapman in the Spring of 1933: "Well, everybody knows you don't steal on the catchers, as a rule, but the fellow I find the hardest to steal against is Rick Ferrell of the Browns. He gets the ball down there where the second baseman wants it."

                          Most observers of the day had similar views of Rick Ferrell, yet Bill James WS formula has Ferrell a B- defender while Dickey and Cochrane get an A. Does Ferrell's handling of the knuckleballers post 1938 hurt his evaluation. Shirley Povich, plus several of knuckleballers, said that the reason they were so successful was the Ferrell was not concerned about his own defensive reputation and thus was willing to repeatedly call for the knuckler, despite the fact that it often made him look bad; something other catchers wouldn't do.

                          Johnny Niggling on Dutch Leonard: "He was lucky enough to have a fellow like Ferrell who wasn't afraid to call for it. There were other catchers in the big leagues, supposed to be good catchers, who wouldn't risk their reputations by calling for the knuckler because it was so hard to handle. But when Leonard came up from Atlanta and found out he could pitch to Ferrell and win, the whole league got knuckle-ball conscious."

                          Povich on Leonard: "...may never have made good in the majors except for the help he got from Rick, who, unlike the catchers Leonard had in his trial with Brooklyn, kept calling for Leonard to throw his knuckler. Ferrell didn't care about his own catching record. In fact, in the same years he was acknowledged as one of the great catchers of the league, he was charged with the most passed balls, due chiefly to that butterfly ball of Leonards."

                          Povich on Niggling: "It's more that a coincidence, too, that Johnny Niggeling developed as a knuckleballer with the Browns when Ferrellwas the Browns' catcher these two seasons. He couldn't win in his previous trials in the majors, but when he found a catcher who could handle his knuckler and wasn't afraid to call for it, Niggeling blossomed as one of the best pitches in the league."

                          Matt, I have been impressed with your thoughts about how difficult it is to separate a player's stats from the team, but how can I believe that a 24 year-old is any more knowledgable and has a better system than the rest of the sabrmetric wizards who evalute history by a couple of computer strokes?

                          How do you adjust for team and managerial desicions. Both Ferrell's played on very middle of the road teams and thus did not have the advantages of players on better teams; IE more rest and platooning. Rick often was hitting .330 into mid to late August for the years 1931-1938 only to tail off badly in the last six weeks.

                          And how would you adjust for managerial decisions like this one. In 1935 and 1936, Cronin started Wes Ferrell and Grove on days of rest between starts this many times. (subtract two, one of each's first start of the seasom).

                          0 days of rest: Ferrell 1, Grove 0
                          2 days of rest: Ferrell 11, Grove 1
                          3 days of rest: Ferrell 36, Grove 27
                          4 days of rest: Ferrell 19, Grove 11
                          5 days of rest: Ferrell 4, Grove 9
                          6 and (+) days of rest: Ferrell: 3, Grove 10.


                          Just wondering?

                          Comment


                          • #14
                            I do think Ferrell's exceedingly high passed ball counts are the primary reason for his defensive demise statistically...some sabermetricians are attempting to find ways to make adjustments for caatching uncatchable pitchers like Wakefield and Sparks in the modern era (quoth Bob Euker: "The best way to catch a knuckleball is to wait for it stop rolling and pick it up"), but that is something I'm not sure how to alter without subjective adjustments...I would need a complete database of pitch types caught by every catcher to adjust for pitching difficulty....either than or I'd need complete wild pitch data...(if a pitcher is throwing a lot of wild pitches then he's probably not making it easy on the catcher)which dosn't exist until 1955.

                            Cochrane and Dickey were both excellent defensive catchers by PCA...I made the mistake of not looking up their letter grades and calling Dickey a poor fielder in another conversation...it is apparent that I was wrong.

                            Cochrane: A
                            Dickey: A-
                            Ferrell: C

                            BTW had I had stolen base data available, those four steals would probably not have gone into the books as steals but Fielder Indifferences which don't go against a catcher's record.

                            Comment


                            • #15
                              Matt,

                              I was wondering:
                              Who has the most walk-off home runs?
                              Can you make a top ten list for the best baseball players and top ten pitchers?

                              Comment

                              Ad Widget

                              Collapse
                              Working...
                              X