Friday, February 25, 2005

larry leads off? yes!

re the advisability of batting larry walker leadoff: the results are in.

james click of baseball prospectus generously ran a quick computer-simulation study on the cardinal batting order at my request. the data suggest that the cards might indeed score a few more runs with larry leading off — but only if the other big bats (pujols edmonds and rolen) bat immediately behind him in the two thru four slots. that’s the lineup josh schulz proposed at birdwatch on wednesday.

the advantage derived from this unorthodox sequence is not huge — a margin of 9 runs, on average, over the default setting (ie, eckstein, walkerpujolsrolenedmonds, grud’k/sanders or sanders/grud’k, molina, pitcher). that finding is consistent with what stewie13 has pointed out in comments here and at birdwatch, and what tangotiger said in a recent post at baseball musings: lineup manipulations can only net you, at best, an extra 5 to 15 runs. but tangotiger adds that 5 to 15 extra runs equals one extra win, and teams throw millions at ballplayers whose marginal value equals one win. if you can net a win for free just by moving guys around in the order, why wouldn’t you?

before we move on to the complete results of click’s quick study, i am compelled to advise you, per the blogger’s oath and the published regulations of the fcc and dept of homeland security, that the following exercise is completely theoretical and that the predicted 9-run gain may be as ephemeral as iraq’s weapons of mass destruction. moreover, click offers a caveat of his own: “Keep in mind that this isn’t meant to map to actual runs scored, but is just for comparative purposes between the various lineups. Finally, the standard deviation is around 40-45 runs, so while lineup C certainly stands out from the rest, the confidence level is low.”

ev’yone get it? this is for entertainment purposes only.

i asked click to run comparisons on five diff’nt batting orders. he ran each one for a thousand seasons, then sent me the low, mean, and high run totals (per 162 games) for each order. in reverse order of potency, with mean averages, the lineups were:

LINEUP D: 964 runs
1 eck’n / 2 grud’k / 3 pujols / 4 edmonds / 5 rolen / 6 walker / 7 sanders / 8 molina / 9 pitcher (low 826, high 1133)

LINEUP A: 970 runs
1 eck’n / 2 walker / 3 pujols / 4 edmonds / 5 rolen / 6 grud’k / 7 sanders / 8 molina / 9 pitcher (low 830, high 1120)

LINEUP B: 970 runs
1 walker / 2 eck’n / 3 pujols / 4 edmonds / 5 rolen / 6 grud’k / 7 sanders / 8 molina / 9 pitcher (low 827, high 1117)

LINEUP E: 975 runs
1 walker / 2 pujols / 3 edmonds / 4 rolen / 5 sanders / 6 molina / 7 pitcher / 8 grud’k / 9 eck’n (low 847, high 1101)

LINEUP C: 979 runs
1 walker / 2 pujols / 3 edmonds / 4 rolen / 5 grud’k / 6 sanders / 7 molina / 8 pitcher / 9 eck’n (low 840, high 1143)

i note, first of all, that the question that launched this inquiry — would the cards score more by flip-flopping walker with eckstein in the top two slots in the order? — is completely moot according to click’s study: both variations averaged 970 runs (lineups A and B). i note further that the “best” lineup, C, is in fact merely a renumbered carbon-copy of the “default” lineup, A. think daylight savings time — we follow the same cycle but arbitrarily displace the starting point by one unit. the “day” isn’t any longer, but by “springing forward” from david to larry we allocate the sunshine more advantageously . . . . . which in this case means (i think) that we allocate about 60 - 80 of eckstein’s at bats to walker pujols edmonds and/or rolen.

does that not sound like a complete no-brainer thing to do? even if it didn’t net you a single run, wouldn’t any baseball fan at a given ballgame rather see one of those four guys get an extra at-bat in the 9th inning, as opposed to eckstein?

add’l thoughts: pujols does not fit the ideal profile of a #2 hitter, as defined by tangotiger’s studies; he hits too many hrs and doesn’t draw enough walks. jimmy edmonds more closely matches the thumbnail, as does walker himself. but if you line walker and edmonds up 1-2, it leaves the cards vulnerable to lefty starters and left-left relief specialists (this pair of batters alone could keep jesse orosco and john franco on mlb payrolls until their iras mature.) might it be permissible to sneak, ahem, eck’n or grud’k into the leadoff hole vs a starting lhp? . . . . here is where we begin to run into the limitations of our study, which click himself would be the first to concede. the simulation program omits dozens of variables that might change the “optimal” lineup on a given day — platoon data, batter v pitcher data, ballpark considerations, left/right alternation, etc etc. it also does not account for pinch hitters, the double-switch (a la russa perennial), or even such quaint phenomena as “hot streaks” and “slumps” — which used to be very legitimate lineup-selection criteria back before the sabrevolution.

in any case, whatever lineup la russa employs today, we know he’s going to use a diff’nt one tomorrow, and yet another one the day after; that’s just tony. do i think he should use lineup C above as often as possible? hell yes. but if he does it even once, the cards damn well better win that day — ‘coz if they lose he’ll never hear the end of it.

and we’ll never see that lineup again.

Thursday, February 24, 2005

larry begs off . . . .

if this becomes more serious, i'll have to start my "reggie leads off" campaign

larry leads off . . . . pujol in two hole

over at the birdwatch, josh schultz offers a larry walker leadoff scenario that's even more unorthodox, and likely more potent, than the one i proposed t'other day: bat larry, albert, jimmy, and scotty in the first four slots in the order, with the pitcher hitting 8th and eckstein 9th. like this:

1. walker, rf
2. pujols, 1b
3. edmonds, cf
4. rolen, 3b
5. grud'k, 2b
6. sanders, rf
7. molina, c
8. mulder, p
9. eck'n, ss

after the 1st cycle through, the lineup would be functionally equivalent to the more traditional:

eck'n, ss
walker, rf
pujols, 1b
edmonds, cf
rolen, 3b
grud'k, 2b
sanders, rf
molina, c
mulder, p

with such a lineup, the first four cardinals to bat in ev'y game would have .400 obps or close thereto, and walker pujols and edmnds would get more late-inning at-bats over the course of the year while eckstein would get fewer. what sucks about that?


postscript: matthew leach addresses a question today at mlb.com about larry walker as cleanup hitter.

Wednesday, February 23, 2005

placido no like home

placido polanco wants out of philly --- would you give up jeff suppan to bring him back to st louis? so asks redbird reasoning in his post today. i for one wouldn't make the trade, even as skeptical as i am about the cards' middle infield --- because i'm more skeptical of the pitchers' durability. but the post is a good read, and it raises one of the more upbeat possibilities of 2005: if things break just right the cards could find themselves in the enviable position of having a proven mlb arm to dangle for trade bait at midseason.

mlb dot communism

parity in baseball? it's coming, predicts tim marchman in the new republic --- and he gives the credit/blame to bud selig, the internet, and mlb.com. "In 2003 [mlb.com] became profitable," marchman notes, "and this winter it made a series of deals that lay the foundation for it to become even more powerful in the years to come." also: "Increasingly, MLB.com has a finger in every online baseball pie, which is why CEO Bob Bowman's prediction of 30 percent to 50 percent revenue growth every year for the next five years doesn't look so out of line. Such rosy forecasts may have something to do with rumors, frequently cited over the last year in outlets such as Wired and USA Today, that MLB.com is going to make an IPO."

article makes selig sound downright visionary, which seriously undermines its credibility . . . . . worth a read nev'theless.

Monday, February 21, 2005

larry leads off

recent study at baseball prospectus examines lineup construction and apparently asks some unorthodox questions — viz., would the giants score more runs with barry bonds in the leadoff slot? you can’t read the article, less’n you have a subscription (which i don’t), but baseball musings has an excerpt and some commentary. ditto only baseball matters, with a followup email exchange between perricone and the study’s author.

which leads me into my annual pout about la russa’s underutilization of the leadoff hole, and an unorthodox question of my own: how would the cardinals fare with larry walker leading off?

i ask because walker excels at the one skill a leadoff man has to have: on-base ability. he has a career obp of .401, which i grant is inflated by 10 seasons at coors field. but his obp away from coors over the last 6 seasons (covering 1255 at-bats) is .382 — with a high of .416 and a low of .370. how good is .370? over the last 20 years only two stl leadoff men have beat it: frankie vina in 2001 (.380) and delino deshields in 1998 (.371).

though womack did a surprisingly nice job last year, he still only ranked 4th on the team in runs scored, tied for 4th in OBP, and fifth in times on base. it’s that last statistic that really tells the tale. like most cardinal leadoff men of recent memory, womack did not excel at his primary offensive task — not even while setting career highs in batting average and obp. he scored 91 runs, a good total — but with three mvp candidates batting behind him, wouldn’t a really good leadoff man have cracked 100 runs by mid-august?

eckstein is no worse than womack, but he’s no better either — last year his times on base (205) and runs scored (92) were nearly identical to tony’s. and his obps are trending downward, just .325 and .339 the last couple of seasons vs a career mark of .347.

what’s the diff’nce between .340 and .380 in the leadoff hole? over 650 plate appearances, it’s about 25 times on base. assuming equivalence among all other stats (steals, extra-base power, etc), the 25 extra times on base translates into roughly 10 to 15 runs. that alone is reason enough to lead larry off, imho. but the case gets stronger yet when you factor in walker’s superior extra-base power. that is, "on-base" usually means first base for eckstein, whereas for walker it often means scoring position or, better yet, a trot around the bags. consider: last year in 608 plate appearances, eckstein propelled himself into scoring position via stolen base or extra-base hit 43 times. walker (away from coors only) did it 34 times — in only 229 plate appearances. true up the plate appearances and walker would propel himself into scoring position 90 times to eckstein’s 43 — which tells us the diff’nce between eckstein’s .340 obp and walker’s .380 obp is going to be a lot more than 10 to 15 runs. more likely we’re talking 25 to 35 runs. and a figure that big has got to show up in the standings.

those of you who’ve read my concerns about eckstein’s glove are now going to accuse me of denigrating the poor bastard’s bat, too. but i’m really not. i’m saying eckstein’s probably going to give you in the neighborhood of 90 runs out of the leadoff hole — but walker’s going to give you 120. i’m saying 90 runs is nothing to brag about when you’ve got murderer’s row batting behind you. . . . . . really, all i’m saying is that la russa should flip-flop his first two hitters, bat walker in front of eckstein instead of the other way around. one advantage you gain by doing so is that you occasionally get some value out of one of eckstein’s outs. in the one case, larry hits a leadoff double, scoots to third on eckstein’s chopper to 2d, scores on a fly to center. in the other, eckstein leads off with a chopper to 2d, walker hits a one-out double, and then gets stranded. same double by walker, same groundout by eckstein — but a diff’t result.

we’ll never get to test this theory in real life, because la russa is about as likely to lead off with walker as he is to confess that he personally jabbed a steroid syringe into mcgwire’s ass. but if anybody out there has access to one of those high-powered simulations and wouldn't mind running a buncha seasons — half with walker batting leadoff, half with eckstein — i’d be curious to see what the computer showed.

Friday, February 18, 2005

reading list

too bizzy to write, but i can still read.

a quick followup to my post on bing devine / brock-for-broglio: mike’s baseball rants rates lou brock as one of the three best in-season acquisitions of all time, counting only trades made after june 1. brock contributed 22 win shares to the ’64 cards . . . .

enlightening post at the birdwatch about the cards’ geezer outfielders — writer named mo boiler looks at similar players at corresponding ages to set some likely parameters for sander/edmond/walker performance this season.

another worthwhile exercise at redbird reasoning — he’s looked at the top 3 nl central contenders, using projected win shares to get a gauge. he sez: cards and cubs 95 wins, astros 88.

item: john sickels’ new blogsite has daric barton as the a’s #2 prospect, with a solid A grade.

Sunday, February 13, 2005

swap me before i kill again

there’s a fun post at hardball times about trading - rates the winners and losers, franchise by franchise, going all the way back to 1960. the interesting thing about the cardinals is not where they rank — by this data they're middle of the pack, have basically broken even — but rather the sheer number of deals they have transacted. by this research, since 1960 the cardinals have made more trades than any team in the majors except the mets — 463 swaps, which works out to just over 10 a season.

can that be right? 10 trades seems high for a single year, let alone for a yearly average. so i decided to visit retrosheet, the source of hardball times’ data, to do a quick spot-check. chose at random a year from my youth — 1975, when i was 12 — and counted ’em up. and i’ll be damned if the cardinals didn’t make 17 trades that year — eight of them during the season. here they are (players to be named later in parens):

Feb 14 Stan Papi to Montreal for Craig Caskey
Mar 29 Danny Godby to Boston for Danny Cater
Apr 4 (Jim Willoughby) for Mario Guerrero
May 9 Larry Herndon and Luis Gonzalez to San Francisco for Ron Bryant
May 18 Ted Martinez to Oakland Athletics for Steve Staniland and (Mike Barlow)
May 27 (Doug Howard) to Cleveland for Luis Alvarado
May 28 Elias Sosa and Ray Sadecki to Atlanta for Ron Reed and (Wayne Nordhagen)
June 4 Ed Brinkman and Tommy Moore to Texas for Willie Davis
June 25 (Mike Barlow) to Houston for Mike Easler
June 30 (Bill Parsons) and cash to ChiS for Buddy Bradford
July 25 Jim Dwyer to Montreal for Larry Lintz
Oct 20 Willie Davis to San Diego for Dick Sharon
Oct 28 Larry Lintz to Oakland for Charlie Chant
Oct 28 Mike Garman and (Bobby Hrapmann) to ChiC for Don Kessinger
Dec 8 Ken Reitz to San Francisco for Pete Falcone
Dec 9 Ron Reed to Philadelphia for Mike Anderson
Dec 12 Buddy Bradford and Greg Terlecky to ChiS for Lee Richard.
Dec 22 Mick Kelleher to ChiC for Vic Harris

imho, this list of transactions appears to be the product of a diseased mind. consider first the number of players who were acquired and dealt away within the twelvemonth — five on’m (barlow, reed, davis, bradford, lintz). consider further the pointless, fevered traffic in bad utility infielders— out with stan papi, ted martinez, ed brinkman, and mick kelleher, in with mario guerrero, luis alvarado, lee richard, and vic harris. is there a peso’s worth of difference among any of these guys? it looks like bing had some kind of compulsive need to roil the roster — kind of like a sick kid has a compulsive need to pick at his chicken pox. surely he couldn’t have thought he was improving the team . . . . ?

with ten minutes’ further digging, i found that 1975 was anything but an unusual year for bing devine. the following year he made another 17 trades, most of them as pointless as his ’75 deals, then tacked on 19 useless swaps in 1977. his "career high" came in 1973 — a mind-numbing 21 trades in that calendar year. four of those 21 deals included a pitcher named mike nagy, whom devine acquired twice and dealt away twice in the course of 1973.

between 1972 and 1977, devine authored a staggering 104 trades — 17 a year. during the full run of his second term as cardinals’ gm (1968-1977), he made 147 trades — 14.7 a year. that’s not normal. per the hardball times article, the per-team average in the 1970s was about 3 trades a season. stan musial, whose term as stl’s gm (1964-1967) separates the two bing devine eras, made only 9 trades during his three-year reign (including the deals that brought roger maris and orlando cepeda). and after devine left in early 1978 and john claiborne took over, the diarrheal spewing of transactions abruptly stopped. whitey herzog, considered a great wheeler-dealer, made only 22 trades during four years-plus as the cardinals’ gm.

so der bingle — what gives? here’s a theory, completely off the cuff and worth exactly what you paid for it. devine’s last trade before he got fired in august 1964 was arguably the most famous trade of all time: brock for broglio. other trades may have been more lopsided (see mike’s baseball rants for a top-20 list), but brock-for-broglio became enshrined as a symbol of the cubs’ futility — and since cub futility is one of the longest-running and most popular shows in the major leagues, brock-for-broglio occupies the same niche as an iconic broadway tune. you know — "yankee doodle" isn’t the best song ever written, but everybody knows it. . . .

anyway, i digress. when the cardinals rallied to win the 1964 pennant, the deposed devine became something of a martyr — and brock-for-broglio his badge of vindication. by the time devine returned the stl front office in november 1967, brock-for-broglio had played a central role in two cardinal world series triumphs. even at that early date, it was part of baseball lore — and that, i think, drove bing devine partially insane. perhaps he thought he had a special genius for talent evaluation and was determined to prove it by finding another brock, even if it meant sifting through a haystack of vic harrises and lee richards and mario guerreros. or, conversely, maybe brock-broglio burdened devine with a reputation for genius he could never live up to, and so he played the role as best he could by trading a lot, if not particularly well.

but most likely devine was just an old-school gm who made trades because that’s all gms had to do in those days. they didn’t have free agents to woo or arbitration hearings to prepare for; they spent their time smoking cigars and chomping red meat and drinking martinis with other gms, and --- sometimes for lack of anything better to do -- making trades. what, you wanna swap ballplayers? sure, what the hell, you take this guy and i’ll take that one. . . . like flipping baseball cards. something to pass the time.

for the record, trader jock has averaged a healthy 6 trades a year as the cardinals’ gm.

Thursday, February 10, 2005

deranged: eckstein, chapter 5

not to sound like chicken little . . . but the more i delve into david eckstein’s fielding stats, the more i find not to like.

this despite some reassuring words from black hawk waterloo, author of the excellent angels blog chronicles of the lads. per the ongoing discussion of eckstein herein (see posts of feb 8, feb 6, feb 1, and jan 9), waterloo writes: "I was very surprised to see Eckstein fare so poorly by David Pinto's method [ie, Probability Method of Range]. Most others have him right around average last year, and he looks pretty good to the naked eye (which of course can be deceiving). He doesn't have a lot of lateral range, but in the past he has positioned himself well, and despite his lousy arm gets the outs that it seems he should. . . . I think Cardinals fans will be more than pleased with his effort afield." adds waterloo: "The Angel pitching staff has been *heavily* flyball oriented the past few years, so there just aren't as many groundballs for their infielders to field. So range factor will undervalue the Angel IFs just a bit ..."

rob haneberg makes a similar point over at the outstanding birdwatch blogplex, in a comment to the feb 8 post: "MGL's UZR and Tom Tippett's Diamond Mind ranking say Eckstein should be at least average in 2005. . . .Eck's range factor is below average but the Angels have had a flyball staff, something like 28th in the majors in GB/FB ratio in 2004, according to ESPN.com."

those are excellent points, from guys who are far stat-savvier than i. but i’m still concerned about eckstein’s range, and here’s why: in all four of his big-league seasons, eckstein has made fewer plays at shortstop than the guys backing him up at the position.

and it’s not even very close. over his career, eckstein has made 4.21 plays per 9 innings (vs a league average of 4.54 plays per 9). in that same four-year period, the angels’ other shortstops — playing behind the same flyball-oriented staff as eckstein — have made 4.81 plays per 9, a spread of more than half a play per game. here’s the data:

ANGELS SS, 2001-2004:
eckstn 4503 inn, 776 putouts 1332 assists = 4.21 plays per 9 innings
others* 1268 inn, 277 putouts 400 assists = 4.81 plays per 9 innings

* alfredo amezaga, chone figgins, shane halter, benji gil, wilson delgado, jose nieves, troy glaus

i know, i know — using range factor these days is sort of like using a compaq 286. ev’body’s moved on to vastly more powerful computational tools like uzr, pmr, defensive win shares, and so forth to measure defensive performance. the problem with those methods is that they are so often in conflict — a gold glover by uzr is too often doctor strangeglove according to pmr or dws — that you never know which one(s) you can trust. and they’re so new that it’s difficult to benchmark the numbers. range factor may be primitive, but at least it’s transparent; you don’t have to be a math ph.d. to understand how it works. i think as long as you work within its limitations, range factor can still tell a truth or two

one of the limitations is that you’re usually not comparing apples to apples — the fielding context (ballpark, pitching staff tendencies, teammates’ fielding range) varies from player to player at a given position, rendering comparisons suspect. but in this case we’re comparing a group of players who did their fielding within the same context — same team, same pitchers, same ballpark, same playing surface. and the comparison doesn’t favor eck —indeed, he has underperformed his understudies every single season of his career. the year-by-year breakdown:

2001 deficit: -0.5 pp9
eckstn 1051 inn, 178 po, 333 a, 4.37 pp9
others 387 inn, 87 po, 122 a, 4.87 pps

2002 deficit: -0.83 pp9
eckstn 1276 inn, 207 po, 397 a, 4.26 pp9
others 175 inn, 35 po, 64 a, 5.09 pp9

2003 deficit: -0.07 pp9
eckstn 985 inn, 193 po, 293 a, 4.44 pp9
others 445 inn, 95 po, 128 a, 4.51 pp9

2004 deficit: -1.19 pp9
eckstn 1191 inn, 198 po, 309 a, 3.83 pp9
others 262 inn, 60 po, 86 a, 5.02 pp9

that sure looks like a pattern . . . . true, the backups logged limited playing time in each season, so random chance might explain the disparity for any year taken in isolation. but over the four seasons the backups have amassed an entire season’s worth of innings — and their range factor over that "season" is .37 pp9 better than eckstein’s best season.

which leaves me a lot less persuaded that eckstein’s apparent lack of range can be explained by the angels’ flyball-oriented pitching staff. he may in fact be a more effective defender than his range factors suggest — but i’m still looking for evidence to support the notion.

Wednesday, February 09, 2005

mark down

re mcgwire and steroids: this column by bryan burwell of the post-dispatch sums up my position.

Tuesday, February 08, 2005

range: muse you can use

there’s a good debate about the cost of eckstein’s range deficit ongoing at baseball musings --- really smart guys who actually understand math. check out the february 7 and february 8 posts over there, and the attendant comments. the consensus seems to be that 44 singles should cost somewhere in the range (very broadly speaking) of 15 to 35 runs. baseball musing’s author, david pinto, used bill james’s runs created formula to peg the cost of 44 singles at 22 runs. "So estimates from 18 to 30 do appear to be in the correct range," he concludes.

however, both pinto and chronicles of the lads, whose calculations placed the alarming 33-run price tag on eckstein’s weak range, advise us to use the data with caution. says pinto: "Because Eckstein is getting 44 fewer outs doesn't mean those outs aren't being picked up by other fielders. . . . There are going to grounders where both the third baseman and the shortstop have non-zero probabilities of fielding the ball. If the third baseman, due to superior range, cuts in front of the shortstop often he'll take outs away from the middleman. In most cases these should even out, but in some situations the shortstop will appear not to get to balls that indeed are outs."

speaking of rangy 3d basemen, chronicles of the lads’ data shows scott rolen to be the most valuable cardinal gloveman, hands down. the chronicler calculates that rolen took 23 runs off the board, which i believe translates into roughly 3 wins in the standings. the chronicler, using pinto's data, has his calculations up for all positions except left field; here’s how the cards come out:

1b pujols +7
2b womack -2
ss edgar -3
3b scotty +23
lf ?????
cf edmonds +6
rf sanders -9

by this method (pending the lf calculations), the cards defense, as a collective, saved 22 runs — about what rolen saved by himself. in other words, the rest of the cards’ strong defenders were canceled out by commensurately weak ones. defense is so flipping hard to measure, i wouldn't take these conclusions as gospel --- but i wouldn't just shrug them off as meaningless either. the cards lost two gold glovers in the offseason; the defense is an enormous concern. pinto and chronicles' range/run estimates may have limitations, but they provide an excellent starting point for understanding what may be the 2005 cardinals' make-or-break issue.

Sunday, February 06, 2005

when it range, it pours

re the cost of dave eckstein’s poor range, a poster calling him/herself "anolis" asks an intelligent question:

"If I read this correctly, Eckstein made 44 less outs than might have been expected, and the analysis suggests that this translated into allowing 33 more runs that expected. Can this be correct? Since the outs that Eckstein missed almost surely turned into singles, is the expected value of turning an out into a single 3/4 of a run? Seems high to me."

yeah, me too — at first. the 33-run estimate comes from the chronicles of the lads, an anaheim angels blogger who is bravely/foolishly attempting to draw meaningful conclusions from defensive statistics. the chronicler explicitly posits that an out by a shortstop prevents 3/4 of a run — i quote:

"I turned to Chris Dial's methodology, in which a run value is assigned to each out a defender makes. For instance, 98.7% of outs made by a SS prevent a single. A single is worth .47 runs. Each out is also just that: an out, which has a value of roughly .28 runs. So, 98.7% of the outs made by a SS prevent .75 runs. (The other 1.3% account for 1.06 runs each, because sometimes those screamers by the SS get through the gap and go to wall for extra bases.)"

in other words, a routine 6-3 groundout moves an inning one out closer to completion (preventing .28 runs) and wipes out the run-scoring potential of a single (.47 runs); add .28 and .47, and you get .75. i hear those gears turning, anolis — now you want to know who the hell chris dial is and whether his methodology makes any sense — and unfortunately, i can’t answer either question. but here's the link if you want to check his math.

i can however apply some common sense, and the exercise encourages me that .75 runs per single is in fact a reasonable estimate. let’s begin with an easy case: bases loaded, two outs, grounder up the middle — shortstop comes up a step slow and the ball trickles through. the hit results in two runs — and potentially more, because the inning continues. every run that scores subsequently would/could have been prevented by a rangier shortstop, so the cost of that particular grounder is over 2.0 runs.

here’s another case: men on first and second, one out, groundball up the middle. if the ss fields it, it’s a double play and the inning is over; if the ball gets through, one run scores, a runner winds up at third with one out, and the batter’s on at first base — the runner at third will score about .75 of the time, and the runner at first will score a certain percentage of the time as well. add it all up and the failure to reach that hypothetical grounder costs, on average, right around 2.0 runs.

of course, there will also be hits that dribble through with two outs and none on, and no harm will come of them. but if you average out a small number of high-cost cases with a large number of low-cost ones, it can still average out to .75. look at it this way: of the 44 singles that result from eckstein's range deficit, a certain portion will trigger big innings, a larger portion will trigger small rallies, and the lion’s share will have no impact at all. let’s make some conservative estimates. say only a third of the bad-range basehits (brbs) lead to any scoring at all — that’s 15 instances, or once every 10 games. let’s further suppose that five of those brbs cost just one run, another five cost two, and the last five cost three runs . . . . . well, that adds up to 30 runs, very close to chronicler’s 33-run estimate. does three runs seem an unreasonable impact to assign a lonely dribbler through the infield? i don’t think so. it’s a cherished cliché that you can’t give a team four outs in an inning; the extra out turns small rallies into large rallies, or non-scoring innings into one- and two-run innings. in that regard, a groundball that’s missed because of poor range is the same as a groundball that’s booted for an error. either way, it’s an extra out.

in fact, errors provide another useful "common-sense" check on the chronicler’s 0.75-run estimate. if we accept that errors are analagous to bad-range basehits — both are outs the defense fails to make — then we can learn much by quantifying the defensive cost of a single error. and that’s easy. we know how many errors are committed over the course of a season, and we know how many runs result from those defensive lapses — they’re called "unearned runs." divide the one into the other, and there’s your value. in the american league in 2004, there were 1520 errors resulting in 900 unearned runs — or .59 runs per error. in the national league, 1646 errors led to 1001 unearned runs, or .61 runs per error.

that’s not quite .75 runs per misplay, although it’s in the ballpark. but keep in mind that some errors don’t put a runner on base, and thus are not completely analogous to bad-range basehits. i’m talking about pickoff attempts that get thrown down the rightfield line, singles that roll through an outfielder’s legs and let the batter reach second or third, relay throws that end up in the dugout. those errors inflate the scoring potential of runners who’ve already reached safely, but they do not turn negative scoring potential (ie, outs) into positive scoring potential (ie, baserunners). the errors in this category — and i have no idea what fraction of the total error population they account for — are less damaging than either a booted groundball or a brb, and hence they lower the average cost of errors relative to brbs. if we adjust for this factor, we move some distance closer to .75 runs / error.

so i have convinced myself — if not you, anolis — that every ball that grazes eckstein's outstretched glove as it bounces past will cost the cardinals 3/4 of a run.

and btw anolis ---- thx for the birthday wishes.

Friday, February 04, 2005

the many and the few

i promise this’ll be the last of these “similar teams” posts for a while; i’m just cleaning out the file. today we look at teams that, like the 2004 cardinals, led the league both in runs scored and fewest runs allowed. the cards were the 10th such team since 1970; how did the previous nine fare the year after? let’s start by naming the teams, with their records:

2001 mariners, 116-46
1998 yankees, 114-48
1995 indians, 100-44 (113-49 extrapolated to 162 games)
1988 mets, 100-60
1984 tigers, 104-58
1978 dodgers, 95-67
1974 dodgers, 102-60
1971 orioles, 101-57
1970 orioles, 109-53

an illustrious group — eight won 100+ games, seven won pennants, and three won world titles. as a group they averaged 105 wins . . . but in the year after achieving this rare feat they notched on average only 90.5 wins. five of the nine teams won fewer than 90, and only one topped 100. six of the nine teams missed out on the playoffs the following year. their records, with playoff teams in bold:

2002 mariners, 93-69
1999 yankees, 98-64
1996 indians, 99-62
1989 mets, 87-75
1985 tigers, 84-77
1979 dodgers, 79-83
1975 dodgers, 88-74
1972 orioles, 80-74 (84-78 extrapolated to 162 games)
1971 orioles, 101-57

2002 mariners - 6th runs, 5th runs allowed
1999 yankees - 3rd runs, 2nd runs allowed
1996 indians - 2nd runs, 1st runs allowed
1989 mets - 3rd runs, 2nd runs allowed
1985 tigers - 6th runs, 4th runs allowed
1979 dodgers - 2nd runs, 9th runs allowed
1975 dodgers - 8th runs, 1st runs allowed
1972 orioles - 8th runs, 1st runs allowed
1971 orioles - 1st runs, 1st runs allowed

note that the three playoff returnees continued to excel in both phases of the game, while the teams that slumped — with one exception — slumped more in the batter’s box than on the mound. indeed, two teams (75 dodgers, 72 orioles) missed the playoffs despite leading again in least runs allowed, and only one team ranked lower than 5th the following year. the average rank for non-playoff repeaters was 6th place in runs, 4th in runs allowed; taking the nine teams as a whole, the avgs are 4th place in runs, 3d in pitching.

implications? for all the worry about the arms, it may be the cardinal bats that decide the team's fate. not a given that they're gonna produce 850 runs again . . . . and if they don't, look out.

Thursday, February 03, 2005

walking in their footsteps . . . .

left hanging from sunday’s primer on 100-win teams: do the 2004 cardinals bear any meaningful resemblance to any recent 100-game-winners? . . . . well yes, matterafact they do, if we judge "resemblance" on the following set of criteria:

1) wins
2) wins above/below pythagorean w-l
3) games ahead of 2d-place team
4) wins the year prior to 100-win season
5) improvement over previous year
6) order of finish previous year

by this gauge, the team most similar to the 2004 cardinals was the la russa-managed 1988 oakland athletics, tony’s first world series team. here’s how they match up:

1988 OAKLAND ATHLETICS
wins: stl 105, oak 104
actual vs pythagorean: stl +5, oak +4
games ahead: stl 13, oak 13
prev year wins: stl 85, oak 81
improvement: stl +20, oak +23
prev year finish: stl 3d, oak 3d

here are some more similarities.
· both teams excelled in hitting and pitching — cards led league in runs and finished 2d in era, a’s 1st in era and 2d in runs
· both had deep, shut-down bullpens
· both teams abruptly stopped hitting in the world series and lost in lopsided fashion.

and here’s a key diff’nce:
· the a’s had four everyday players aged 25 or younger (canseco, mcgwire, walt weiss, and stan javier) and got only 964 at-bats from players aged 34 or older. their position players averaged 28.1 years of age. the cardinals last year averaged 30.7 years of age and had only one regular 25 or younger (pujols). they got 1847 at-bats from players aged 34 or older.

what the a’s did next: returned to the playoffs three of the ensuing four years, winning two more pennants (’89-’90) and one world championship (’89).


1984 DETROIT TIGERS
wins: stl 105, det 104
actual vs pythagorean: stl +5, det +5
games ahead: stl 13, det 15
prev year wins: stl 85, det 92
improvement: stl +20, det +12
prev year finish: stl 3d, det 2d

additional similarities:
· the tigers led the league in runs scored and fewest runs allowed
· they had three gold glovers (lance parrish, lou whitaker, and alan trammell)
· they had a hall of fame manager (sparky)

key diff’nces:
· like the ’88 a’s, the tigers were much younger than the cards — position players averaged 28.5 years, only 577 at-bats from players aged 34 and up. their five best hitters all fell between 26 and 29 years old.
· they won the world series

what the tigers did next: finished 3rd the next two years, with 84 and 87 wins (15 games and 8.5 games out of first, respecitvely). in 1987 they won another division crown, beating toronto on the last day of the season in one of history’s best-ever pennant races. in 1988 they won 88 games and finished second by a game to boston. in 1989 they collapsed, finishing last at 59-103, 30 games out.


2002 ATLANTA BRAVES
wins: stl 105, atl 101
actual vs pythagorean: stl +5, atl +5
games ahead: stl 13, atl 19
prev year wins: stl 85, atl 88
improvement: stl +20, atl +13
prev year finish: stl 3d, atl 1st

add’l similarities:
· just as old as the cards — avg age nearly identical (30.6); just two regulars 25 or under, with the rest over 30; 1252 at-bats from the 34-and-up set
· like the cards, the 2002 braves had been deemed also-rans by preseason pundits. their main division rival (ie, mets) had made splashy offseason acquisitions (bob alomar, mo vaughan, j burnitz) while the braves essentially stood pat.
· hall of fame manager

key diff’nces:
· imbalanced: led the league in era but finished 10th in runs scored
· unstable: replaced four-fifths of their rotation the ensuing season, plus two position starters
· two hall of famers in rotation (glavine and maddux)

what the braves did next: won division titles in 03 and 04 with 101 and 96 wins, respectively. the braves got younger in 03, with four regulars under 30 years old, and younger still in 04, fielding seven 20somethings and one 32-year-old (chipper jones). the braves’ pitching corps experienced a similar turnover.


1980 NEW YORK YANKEES
wins: stl 105, nyy 103
actual vs pythagorean: stl +5, nyy +6
games ahead: stl 13, nyy 3
prev year wins: stl 85, nyy 89
improvement: stl +20, ny +14
prev year finish: stl 3d, nyy 4th

add’l similarity:
· old bats: only one regular 25 or under, four aged 34 and up, with the geezer cohort contributing 2001 at-bats

difference:
· old arms: the yankees got nearly half their starts (76) from pitchers aged 35 or older. one of the old guys (tommy john) won 22 games, and another (rudy may) won 15; the remaining two (el tiante and gaylord perry) went a combined 12-13.

what the yankees did next: went to the world series in strike-marred 1981, then settled in for a decade of above-averageness — four straight finishes of 2d or 3d place, seven straight years between 79 and 97 wins.



these comparisons only go so far, so i’m not drawing firm conclusions. but it is at least encouraging that all four of these similar 100-win teams remained very competitive for quite a few years. three returned to the postseason the following year, with two getting to the world series; and two kept themselves at or above the 100-win mark for several more seasons. i’m not expecting the cardinals to do that — they’re too old — but these four models suggest that 90-95 wins remains a realistic expectation, at least for this year.

Tuesday, February 01, 2005

range redux

a while back i guesstimated that david eckstein's limited range at shortstop "might cost the cardinals between 35 and 100 base hits — ie, 35-100 balls that edgar would have gloved."

in his post today at baseball musings, david pinto makes a similar guess. having examined minutely detailed data from the 2004 season, pinto estimates that eckstein cost the angels about 45 outs. further number-crunching at an angels blog called chronicles of the lads suggests that eck was worth negative 21 runs to the angel defense. i quote:

"David Eckstein had 3,562 balls in play against him. He made outs on 356 of these when about 400 would have been expected, based on the probabilities observed from 2002 through 2004. Using Dial's methodology, we find that he prevented 268.43 runs where 301.81 would have been expected; so he's -33.37 Runs Against Expected. We convert that to 4,000 BIP to get a number of -37.47. However, as the average SS was -16.22, that makes Eck -21.26 against average, which is pretty bad."

more than "pretty" bad, actually --- the only shortstops who cost their teams more runs were nomar garciaparra and felipe lopez. the same post shows edgar renteria in the middle of the pack at negative three runs, which means (if you buy all this analysis) that the switch to eckstein will cost the cards something like 18 runs ---- or 2 to 3 games in the standings.

if stl misses the playoffs by 2 games, we'll know whose fault it is.