Cleaning up the mailbox
Posted by Andy on August 9, 2011
I get a lot of suggestions from readers sent in by email, and many of them are quite good.
Here are a few recent ones:
Win Spread
Dan F. wrote in before yesterday's Phillies-Dodgers game noting that Roy Halladay (14-4) and Hiroki Kuroda (7-13) had a large difference in their records, with a "win spread" of 27 (the sum of Halladay's win total and Kuroda's losses). Dan wondered what the biggest win spreads in history are. There's no way to use the PI to find that.
Certainly there have been much larger ones at the end of seasons--for example in this game near the end of the 2003 season, Kyle Lohse (14-11) matched up with Mike Maroth (8-21) for a win spread of 35.
But--what are the largest win spreads for a game as early in the season as last night's?
Incidentally, last night came out true to form as Halladay picked up the win and Kuroda the loss.
Brewers on the road
John V. wrote in to talk about the Brewers' disparity in home and road records. Splits in records can be found right here. At the moment, the Brewers have 20 more losses on the road than at home. The Athletics (15) and Nationals (13) are the only other teams with a difference of at least 10.
The Brewers' home winning percentage is .732 while on the road they are .407, for a difference of .325.
Last season, the teams to finish with the biggest disparity in home and road losses were the Tigers (23) and Pirates (23), plus the Braves at 21 just missed made the playoffs. The Tigers win percentage at home was .642 and one the road it was .358, for a difference of .284. The Pirates at home were .494 and .210 on the road, for a difference of .284 as well. The Braves were .691 at home and .432 on the road, for a difference of .259.
So, the 2011 Brewers are really quite stilted...
Most homers since 1990 by stadium
Reader eorns wrote in with a list of most homers by stadium since 1990. I don't know how he figured it, but the top 10 are:
US Cellular 3908 SkyDome 3832 Wrigley Fld 3797 Coors Fld 3644 Camden Yards 3622 Rangers Bpk 3532 Angel Stad 3486 NetworkAssoc 3342 Fenway Pk 3336 Yankee Stad 3272
The stadia with the fewest are those that haven't been used the whole time--Fulton County, PetCo, Astrodome, etc.
It would be neat to calculate this on a per-game basis, eorns!
Young Astros
Reader Harrison L wrote in the talk about how inexperienced the Astros' recent starting lineups have been. For example in their August 3rd game, the rookies included Jose Altuve, J.D. Martinez, Jimmy Paredes, and Jordan Lyles.
Houston starting a very young lineup? This is nothing new. Think back to September 29, 1963.
August 9th, 2011 at 10:09 am
Ah, Chicago... a city that loves its home runs.
August 9th, 2011 at 10:18 am
And yet in April and May they say its too cold to hit them!
August 9th, 2011 at 10:18 am
If you don't have a cutoff date, the biggest win spreads undoubtedly came in the early years of baseball. If Old Hoss Radbourn plays anyone else in his last start, his win spread is 59 plus however many losses the other guy had -- his win total is so high that it wouldn't even matter. He'd blow away any 20th and 21st century candidate.
The same would probably be true almost no matter where you put the cutoff date. Pitcher wins have been slowly dropping over time for a really long time, so the winner in this metric would be at or near the cutoff date every time.
August 9th, 2011 at 10:21 am
The 1987 Twins won the World Series after an extreme home/road split: 56-25 home, 29-52 road.
The 1949 Red Sox also enjoyed Boston: 61-16 home, 35-42 road.
August 9th, 2011 at 10:23 am
"Recently", Denny McLain and Sonny Seibert combined for 32 on a game with yesterday's date (8-8) in 1968.
http://www.baseball-reference.com/boxes/DET/DET196808080.shtml
That was a lopsided contribution (23-9). With some balance, one could get quite a bit higher I would imagine. Don't know how to search efficiently though and I'm supposed to be working.
August 9th, 2011 at 10:46 am
Re: "Brewers on the Road," the Braves made the playoffs last season....
August 9th, 2011 at 10:47 am
Interesting that more home runs have been hit in U.S. Cellular Field than in Fenway or Wrigley since 1990. Wasn't U.S. Cellular a pitcher's park at one point?
August 9th, 2011 at 10:52 am
#6, thanks. I misread the chart I was looking at.
August 9th, 2011 at 10:56 am
Wilbuw Wood is a good choice for this since he made so many starts from 71-75. In 1972 his own record was 18-11 on August 8th, 20-16 in 1973 and 17-13 in 1974. I couldn't find any as high as 32 in that time frame but there are plenty that are high 20's or 30. For example:
- 8/11/73 - Wood is 20-16 & Mickey Lolich is 11-11 entering the game. Lolich gets the W, Wood the L so 20-17/12-11 after the game.
-
August 9th, 2011 at 11:02 am
Is there a place where we're supposed to post ideas for blog posts, if we have them. Of course, you're not obliged to use them.
August 9th, 2011 at 11:06 am
Michael, there is no place to post ideas, but you can email me or the other blog authors with ideas or questions. Our email addresses are available on the sidebar on the right under "Contact".
August 9th, 2011 at 11:29 am
Wasn't there another Colt 45 game in 63 which included all rookies? I am remembering Jay Dahl, Ron Davis or Brock Davis. No Aspromonte-all rookies I think.
August 9th, 2011 at 11:29 am
Pitching Win-Loss is pretty much a useless stat. A better comparison would be lifetime ERA+ or even WAR.
August 9th, 2011 at 11:40 am
Some live-ball era win spreads I found:
49: A's- Browns, Sept. 12, 1931, Grove (28-3) and Sam Gray (9-21)
45: Mets-Dodgers, Sept. 25, 1963: Koufax (24-5) and Roger Craig (5-21)
44: Reds-Cubs, Sept. 23, 1933: Guy Bush (18-12) and Paul Derringer (7-26)
43: A's-Red Sox, Sept. 7, 1928: Grove (21-6) and Ruffing (9-22)
42: A's-White Sox, Sept. 21, 1920: Scott Perry (11-23) and Cicotte (19-10)
For one this early in the season I found this:
33: Reds-Pirates, August 11, 1922: Luque (8-18) and Wilbur Cooper (15-9)
All records are at the start of the game, and so don't include that decision.
August 9th, 2011 at 11:55 am
On August 7, 1962 Don Drysdale 21-4 defeated Craig Anderson 3-13 of the Mets for a 34. I didn't go back further than that year.
Check out the Sept. 27, 1963 Houston game also. A lot of rookies there playing in their only game of the year.
August 9th, 2011 at 12:51 pm
@12, @15.
It wasn't just that game where the Colt .45s/Astros showed off their young guns. They had remarkably young teams in the early 60s, often including a teenager (or two or three).
The Angels at the same time seemd to have a similar philosophy.
Or, maybe it was just two expansion teams fielding their teams on the cheap?
August 9th, 2011 at 1:07 pm
#15 That is the game I remember. Three pretty good players in the middle of the lineup plus Jerry grote.
August 9th, 2011 at 2:14 pm
@12, @15, @16/ "... Colt 45 game in 63 which included all rookies..." it was also the LAST game of the year, (9/29/63) when managers frequently get "creative" with their lineups, which would explain the large number of very young players.
August 9th, 2011 at 2:46 pm
@18
Those Colts took that strategy to the extreme. Not only lots of rookies, but lots of teenagers. And a handful of guys called up from the single-A. Joe Morgan, John Paciorek and Sonny Jackson were teenagers called up from Modesto.
They went 11-3 down the stretch (6-1 against the Mets).
August 9th, 2011 at 4:25 pm
It's kind of scaring me that I can now recognize oblique references to the Paciorek debut and Shamsky's big game....
August 9th, 2011 at 6:26 pm
More win spreads.
This may be the biggest ever, albeit on the last day of the season.
- 52: Cards-Reds, Sep 30, 1934, Dizzy Dean (30-7), Si Johnson (7-22)
Big one for a mid-August game.
- 37: Dogers-Cubs, Aug 13, 1966, Sandy Koufax (19-6), Dick Ellsworth (5-18)
August 9th, 2011 at 6:42 pm
WIN SPREAD. Never heard of it before this. My first reaction is that it would be more meaningful today, considering how quickly pitchers are removed, as compared to decades ago when complete games were expected and 20-game winners racked up losses in the teens, and "win spread" doesn't seem to balance the scale there. Whose "win spread" applies when two 20-game winners are the starters, one has lost 4 and the other has lost 13? Of what value is "win spread" when we're talking about two 20-game winners? It might tell you something about which pitcher perhaps runs out of gas sooner than the other, but they both still hang around long enough to win 20 games. ??? These days "win spread" seems to be an indicator of which pitcher will get yanked first, because he more frequently leaves the game while his team is trailing and never catches up to tie the game, which the team goes on to lose. Ergo, Win Spread has a slight tendency to predict which team will win, I would think.
Am I thinking straight so far?
In the above-cited example, I submit a more concise measuring tool: the
HALLADAY SPREAD. Never mind the W-L, Roy Halladay is the advantage no matter who else is pitching. 🙂
Since we're electing them to the Hall of Fame now, let's not forget relievers.
SAVES SPREAD: When Mariano Rivera enters the game, compare his Saves to the Saves Blown stat of the other team's reliever, and what does that tell you?
August 9th, 2011 at 6:46 pm
If Old Hoss Radbourn plays anyone else in his last start, his win spread is 59 plus however many losses the other guy had -- his win total is so high that it wouldn't even matter.
Well, I just had to check. October 11, 1884, Providence vs. Cleveland, Radbourn starting against John Harkins. Radbourn came into the game with a record of 57-12, Harkins with a record of 12-31 — an 88-game win spread. Radbourn won the game, 8-1, and then made one more start against a Philadelphia pitcher making his only pitching appearance of the season.
August 9th, 2011 at 7:02 pm
When I first looked at the term win spread, I thought it would be the pitcher with the better record's excess wins over losses (in Halladay's case, 10) + the worse pitcher's excess losses over wins (in Kuroda's case 6). This gives a more, um, useful number (16 in this case).
This clears up the issue where a Wilbur Wood, Gaylord Perry-style pitcher with a 15-15 record doesn't look worse (by Win spread) then Kuroda does.
The original way would have a win spread of 29, implying that Halladay would prefer the 15-15 pitcher over the 7-13 pitcher, while the new way would have a spread of 10 (which is fewer than 16 and more representative of the narrower advantage Halladay would have).
August 9th, 2011 at 7:23 pm
@22.
Phil in Indiana,
I think win spreads was just a frivolity. Just having some fun with box score trolling, nothing more.
August 9th, 2011 at 8:24 pm
In 1990, the ChiSox played at Old Comiskey Park. And the Cell still has more HR than anybody else.
August 9th, 2011 at 8:46 pm
Yes, Win Spread is absolutely a frivolity....a curiosity and nothing more. The term was coined, as far as I know, by Dan F., who emailed me the idea.
August 9th, 2011 at 9:48 pm
@13...the day that winning the game is not the most important thing for a team/player is the day they may as well close the stadiums up. Maybe they should keep score by saying whichever team is the lower WHiP wins.
August 9th, 2011 at 9:53 pm
Thanks, Andy for giving some reader-e-mails the light of day in BR.
"Reader eorns wrote in with a list of most homers by stadium since 1990."
Eorns, I'll direct this question to you, rather than Andy. Have the dimensions and configuration of the parks on the list remained exactly the same since 1990 despite the name changes?
I know no changes were made to the Skydome when it became the Rogers' Centre (Canadian spelling) but what about the Rangers' ball park? Didn't they try to reduce wind gusts blowing out by installing high-level billboards in the outfield in the modern park?
How could Coors' be behind Skydome/Rogers as a homerdome? If we're not comparing the same number of seasons in use, eorns, then what is the point of the list?
August 9th, 2011 at 11:03 pm
The 1970 Orioles made all five of their World Series errors in the four games that they won. Their opponents, the Reds, made all three of their Series errors in the only game that they won.
Discuss.
August 10th, 2011 at 1:05 am
@30.
One thing I remember from the '70 series is a missed play by the home plate umpire that made a notable contribution to the Reds only win in game 4. Somehow the umpire stumbled and ended up on his duff with his back to home plate just as Brooks Robinson slid in, almost certainly safe, ahead of Bench's tag. With his back still to the plate, the umpire emphatically called Robinson out to end the 3rd inning. The Orioles had already scored 3 that inning and could have broken the game wide open. A Reds team down 3-0 in the series may have struggled to mount the comeback that did see them eke out a 6-5 victory that day.
As to why errors may or may not be a factor in the game, context does play a big role, of course. But, the specific scenario you described does seem unusual. But, let's chant together, "sample size".
August 10th, 2011 at 1:13 am
But, let's chant together, "sample size".
Goodness, in a best-of-seven series, you don't even have to scan the libretto for that disclaimer. It's right on the cover of the program.
August 10th, 2011 at 1:40 am
@32.
I admit "sample size" doesn't add much to the discussion. Sorry. 🙂
One thing about the umpire's botched call @31, that was what "prevented" the Reds' error from costing them. Robinson went to second when the centerfielder misplayed his single. So, without the blown call, that error would have made a contribution to Robinson's run scoring. The Orioles did score (once) on the Reds second error in the 6th, but couldn't do anything with a 2-out booted grounder in the 9th.
August 10th, 2011 at 8:53 am
The Orioles gave up 3 unearned runs in the Series, but they still outscored the Reds 33-20 over 5 games with 50 vs 33 hits and 10 vs 5 HR. The Orioles had 1/9 pitcher in the Series with a WHIP > 1.28. The Reds had 8/9 above 1.28. Even with that imbalance, 3 of the 5 games were decided by 1 run. The Orioles were down 0-3 and 0-4 in the first two games, but won each by a single run 4-3 and 6-5, despite 4 of their 5 errors in the Series. Cincinnatti was outscored 16-2 in the middle innings in the 5 games (10-2 in the 3 one run games).
August 10th, 2011 at 11:31 am
@31, I read about that as a six-year old, in the Step-Up book "Baseball Players Do the Strangest Things." Wasn't Bernie Carbo involved in that play somehow?
August 10th, 2011 at 4:18 pm
re: replies to my POST #22 and responses that the "Win Spread" concept is more or less a "frivolity....."
as if my post #22 is to be taken seriously? ? ? ? ? ?
🙂
August 10th, 2011 at 4:40 pm
Phil, I think of you as a lite version of Jack Keefe, who posts periodically, parodically and hilariously on BaseballThinkFactory.com. Your posts have smiling, winking emoticons strewn around just behind the words. It's just that not everyone here understands that.
Here, reader [*reaching*]. Take the salt shaker.
August 11th, 2011 at 7:04 pm
re: post 37, thanks KaHuna. I love this domain, but I learned long ago to take statistics with a grain of salt.
I began to understand this as a kid when I would round up the numbers from baseball magazines and design cards for a baseball board game I liked to play a lot, and I saw how On Base Percentage, a non-factor back then, is sometimes a better indicator than Batting Average. Most mags didn't even list OBP or OBA. I also got pretty good at working out Earned Run Averages in my head (this was in the day before pocket calculators, and I was using a slide rule).
I also began to see the limitations in statistics' ability to predict anything or tell you how valuable a player is, for example, the chances that Al Weis will beat you with his home run power. Going into 1969, he had a grand total of 4 in more than 1300 at bats. Then in the midst of a pennant race, BAM-BAM, two days in a row he goes yard as the Mets beat the Cubs. What were the odds? Who'd'a'thunkit??