User Comments, Suggestions, or Complaints | Privacy Policy | Terms of Service | Advertising
|
Demarini, Easton and TPX Baseball Bats
|
AllianceTickets.com has cheap MLB Tickets. Get all your Colorado Rockies Tickets, Seattle Mariners Tickets, San Francisco Giants Tickets and all your favorite baseball tickets here. We also carry cheap Denver Broncos Tickets, Seattle Seahawks Tickets and Denver Nuggets Tickets. |
For wholesale prices on baseball gifts and equipment, check these stores out! |
Page rendered in 0.4278 seconds
54 querie(s) executed
Reader Comments and Retorts
Go to end of page
Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.
1. Der Komminsk-sar Posted: January 10, 2008 at 05:25 PM (#2665443)I use Access 2003, with the data grouped by years into 7 databases. I take a "fielding extract" from each, with just the plays and fields I need to run this, and load those into a separate database. I takes lots of hours to set up and crunch the whole thing. I'm really glad its done for now - at least until I find 10-20 things I need to do to improve it, which I trust everyone here will help me identify.
Only had time to skim the article, so apologies if the article covered this but does this system factor in double plays for second basemen? I was surprised to see Maz in 3rd.
Man, the Orioles had a helluva infield defense for a while, didn't they?
Not only the infield defense, they had Paul Blair in center. Pitching in fron of that defense was a pitcher's wet dream.
It would be interesting to see how TotalZone correlates with standard ZR back through 1987.
I just discovered the retrosheet play-by-play data myself, and I'm starting to process the data using mysql on this website.
I'm at a very prelimanary stage, but would you mind if I steal some of your ideas when I get around to doing defense?
Surprised that both Sweet Lou and Frank White are ahead of Maz, even though the time period cuts off parts of their careers, while Maz gets his entire career in there.
I would ask to keep your stuff open source, but that's up to you.
Zone rating is the reason. I've calculated beyond 1986, but I feel zone rating is a better measure for those years.
Pitcher handedness - yes, I will add that to the list of improvements.
Bet counting the whole career could cost them.
How well do your results correlate to zone rating?
Can answer that tonight.
Thanks a bunch for doing this, Sean Smith, wherever you are. I was just talking about Don Buddin yesterday in the Red Sox newsgroup but had little to say about his defense (he seemed to have below-average fielding percentages and above-average range factors, according to bb-ref, but what that means, I don't know). This at least gives me something else to point to.
Of course, although I doubt I will add anything, I'm more looking to pull together everything in one place for my own analysis.
http://www.replacementlevel.com/stuff/sg/1987-2007 Defense.zip.
This is in MS Access format, but I can make a CSV if there's a demand for it.
For most of the best/worst list I could tell exactly who it was from the retro id, but I had to look him up. What happened to him? He was a pretty good hitter for a middle infielder, had 160 bad AB at age 28, and never played again. I would think that even if they didn't like his D at short, he could have played second.
On second thought, its happening all over again! Low average, lots of walks, medium power, good overall offense for his position, not so great on D, loses his job at first sign of a slump - D'Angelo Jimenez!
Harveys - what do you think about the best/worst list? Is there anyone you think is rated egregiously wrong?
Norm Cash
Reggie Smith
Frank Howard
Its a big file, a little over 2 MB.
I remember Smith doing very well and Howard pretty bad. Not sure on Cash.
AROM, I am flattered you ask but no, I think this does a fine job of validating what folks suspected but didn't "know". My only quibble is that I think it must be overstating left field in Fenway. Yaz was a good outfielder but not THAT good. Some quirk I imagine.
And yes, Al Kaline really was that good. Just one of those guys who always got where he needed to be.
And Greg Luzinski was just horrible. It's been stated before but it's almost impossible to decribe to folks today how bad Luzinski was in the outfield. It was like watching one of Howard Stern's drunken, angry dwarves (dwarfs?) play the field........
What would be a good Baltimore infield? Let's say 1973: -2, +25, +26, +21. Yikes!
In the Hall of Merit, we did elect Keith Hernandez, and I'm feeling good about that. We also elected Jim Palmer, and we should have - but if you don't look at this, you're overrating Palmer.
Frank Howard: -19 in RF, -45 in LF, -10 at 1B
Reggie Smith: +42 in CF, +27 in RF
Thank you. Thanks to everyone for the compliments.
Joe Torre, 1971: -22 (dims some of the luster of hitting .363)
Joel Youngblood, 1984: -20
Frank Thomas, 1958: -16 (this would be the white Frank Thomas)
Johnny Bench, 1982: -15
Mike Shannon, 1970: -14 (in a short season)
Hal Smith, 1959: -13
Pete Rose, 1975: -12
Pedro Guerrero, 1984: -9 (hey, a lot of "real" 3B are worse than that)
On the other hand, we gave Jimmy Wynn a lot of peak credit for 1969. This has him at -26 in CF. (But is the Dome another really weird park to watch out for?)
...
Sarge Matthews -83
Greg Luzinski -94
A Philadelphia tradition for a while :)
Edit: Greg Gross, caddy for both those gentlemen, was -16 for his mostly part-time career.
Bob Dernier was +19 for his brief career, I am happy to say :)
I was thinking that too. But then, maybe its Palmer's greatness that is overrating Brooks, Blair, and Belanger? I'm not sure how to best approach that but it's worth studying. I don't consider this a finished project, and I'm perfectly happy if somebody else takes this and brings it to the next level.
I'm going to assume this is wrong. Retrosheet has a lot of missing plays in 1969, mostly concentrated on the Astros, Braves, Pirates, and Padres. The first run I had an amazing number of plays credited to the rightfielder, but there are probably bugs messing up ratings for all those teams. Probably best to throw out the ratings for 1969 and 1970 for those teams.
That's a huge park split.
geigg101 Geiger Gary CLE 1958 cf 167 10geigg101 Geiger Gary CLE 1958 rf 13 -1
geigg101 Geiger Gary BOS 1959 cf 202 -2
geigg101 Geiger Gary BOS 1959 lf 75 3
geigg101 Geiger Gary BOS 1960 cf 21 -1
geigg101 Geiger Gary BOS 1960 rf 169 -7
geigg101 Geiger Gary BOS 1961 cf 485 -4
geigg101 Geiger Gary BOS 1962 cf 445 1
geigg101 Geiger Gary BOS 1963 cf 349 0
geigg101 Geiger Gary BOS 1965 cf 40 1
geigg101 Geiger Gary ATL 1966 cf 56 -6
geigg101 Geiger Gary ATL 1966 rf 16 0
geigg101 Geiger Gary ATL 1967 cf 51 -3
geigg101 Geiger Gary ATL 1967 lf 17 0
geigg101 Geiger Gary ATL 1967 rf 13 -3
geigg101 Geiger Gary HOU 1969 lf 56 -1
geigg101 Geiger Gary HOU 1969 rf 25 -5
In the Hall of Merit, we did elect Keith Hernandez, and I'm feeling good about that. We also elected Jim Palmer, and we should have - but if you don't look at this, you're overrating Palmer.
A few months ago, someone made a pretty convincing case in a BBTF thread that Palmer pitched to contact, and could dial it up and get strikeouts when he needed them. So, while his defense obviously helped him, it's not clear that the great defense should be held against him.
The correlation is also a bit low at 1B, which is also true for Fox's SFR. One thought there is that 1Bmen may vary a lot in how much they defer to 2Bmen on grounders that either player can handle. It doesn't really matter at the team level who makes the play, but could create the appearance of large differences. (I suppose that 1Bman could also vary a lot in terms of how much they defer to RFs on shallow FBs.)
http://www.hardballtimes.com/main/article/the-all-time-best-pitchers/
(he's No. 19)
Well sure, I agree there's SOME difference. Of course, as that says, he was much worse than average when compared with other LFs who played at Fenway, and he was still -13 per 162 on the road (your system puts him as above average in 2007. It stretches credibility, if you take the road -4 runs as a given, to believe he was +6 at home after park adjustment in 75/80 games (which is what he would need to be to bring his overall average in line with the total zone rating), considering all other LF were so much better than he was at Fenway.
Don't get me wrong, Sean, I think this work is fantastic. Of course you are going to have some guys where the data looks questionable, for a variety of reasons. That doesn't invalidate the system. But I think it does raise questions, and one of those is whether the Fenway park factor is correct. I think both Manny's and Yaz's ratings suggest it's too forgiving.
Or that general perception is too harsh.
For Yaz, its important to read this the right way. He may have the highest positive value of any LF during that time period, but he's no Willie Mays. I have him at +11 runs per full season of chances (500 for LF). That suggests an excellent corner outfielder (but not the best on a per season rate - check out Willie Wilson) who would be slightly above average if he were playing center. Over his career Yaz did play a season's worth in center, 159 games, and is +5 there.
The way I read the data he was a centerfield - quality defender (though not an elite one) who played a long career in left. Sound about right?
- per 500 chance rating of the composite Fenway left fielder, not including those named Carl, Jim, Mike, or ManRam (oh yeah, and Ted - got his last 5 years).
- career home/road splits for those who do have those first names.
Mike Squires 1B -1
Rocky Colavito +38, but -3 in the year I saw a lot of him and thought he was very good (1964)
Aurelio Rodriguez 3B +73
Checking one my eyes said was average:
Zoilio Versailles -3, but +14! in 1965.
Checking one my eyes said was overrated:
Claudell Washington -49
Dwight Evans is +62.
Darrell Evans is +55.
Could be, Matt.
I look forward to stuff on deck, AROM. I do understand what you're saying - generally, LF is where you dump bad outfielders, so a LF who is historically elite at defense would not then become a historically elite defender in terms of all OFs. That makes sense.
When I was a kid, we lived in Philadelphia for three years during the late 1970s. I had a Greg Luzinski doll. When you pulled the string in its back, the doll said, "It's yours, Garry. It's yours, Garry. It's yours Garry."
Take the outfielders who can run and put them in center, take the ones who can throw and not so immobile to negate their arms and put them in right. Compare Luzinski to what's left - and he still has the worst negative run value of any player in that time period.
Yaz: Home +60 (+11/500)
Road: +54 (+10)
Rice Home: +13 (+3)
Road: +29 (+6)
Other RS: home +22 (+5)
Road: +0
Visiting players were +117 (+8 per 500) which is a red flag and means I have to look further.
Right now I don't have it set up to look at how opponents did against the Red Sox on the road.
I haven't gotten to 3B double plays yet, but he should get half credit for those too. Also, perhaps a 2B should get more credit than the SS for the pivot, but I'm not sure how much.
I looked at Kaline and Clemente and they had massive swings from year to year. Both go from excellence to mediocrity and, in some cases worse, one year to the next.
Since both played in the same park with the same team and were reasonably consistent in health (Clemente's ailments notwithstanding) and batting performance it seems strange that one year they're great and the next not so much.
There's no problem. Left field at Fenway is such a magical place that it makes all players above average.
Great article.
This includes Ted Williams' time; did he play evenly at home and on the road? He was a really, really bad outfielder by the end of his career, it looks like.
Kaline:
1956 +7
1957 +14
1958 +22
59-60 moved to center, was about average
1961 +17
1962 +4
1963 +1
1964 +13
After that he's in his decline phase, never above +10 but never below -3 either. I'd call that remarkably consistent. You should check the 2003-2007 UZR file, linked at the end of my article, to see what kind of year to year swings to expect. If you're expecting a great fielder to be +12 +13 +9 +14 +12 +15 before you say its consistent, I don't think you'll hardly ever find it.
Clemente:
You've got a point here, in an otherwise great career he's -7/-9 in 1960-1961, then -14 in 1969. The 1969 is likely a data error, as I said before there was a problem with missing plays in the retrosheet files that year. 60-61, I'm not sure. It does look a little weird but its not unprecedented for defensive measure to have fluctuations like that.
+31
+10
+2
+14
Craig Biggio
2005 -7
2006 +13
Its why you always want to look at multiyear data if you have it.
+12
0
-11
+10
Its all just sample data. The bigger your sample the better your estimate.
Obviously, he as one example is a tiny sample... maybe an issue with his data or something not yet defined.
Sal Bando +48 (better than reputation and other measures)
Bert Campaneris +79
Orlando Cepeda -21 (confirms)
Ron Cey +43 (strong until 80s decline)
Dave Concepcion +35 (only mid-70s were remarkable)
Nellie Fox +60 (partial data)
Jim Fregosi +15
Bobby Grich +79
Reggie Jackson +9
Harmon Killebrew -60
Chet Lemon +99
Eddie Mathews +38 (not a butcher over there)
Tony Perez -10
Willie Randolph +40
Ron Santo +26
Willie Stargell -110 (!)
Dave Winfield -55
Robin Yount (as a SS) +23
-- MWE
Right now I don't have it set up to look at how opponents did against the Red Sox on the road.
Why is this more relevant than how these people did against everyone on the road?
Because of the impact of facing Red Sox hitters.
The park factor is definitely too strong, if figured out why last night but still working on how to best fix it. Working on infield park factors as well, they make much smaller impacts than the outfield PF, but some of the beneficiaries are Concepcion and Maz, moving up about 15-20 runs.
On missing plays, I noticed a large number of plays coded as fielded by = 9, 1st assist = 9 and 1st putout = 9. Those are the plays I tried to remove. Are there entire games with no records at all as well?
You must be Registered and Logged In to post comments.
<< Back to main