Stolen Thunder
« Grumpy Old Men | Main | Bullpen 2005 »I know, I know, Stolen Thunder sounds like the next crappy film to come out of the Project Greenlight franchise, but if the resulting movie has anything to do with the Cardinals reclaiming their place in baserunning history, then I'll be first in line to buy a ticket.
But the Cardinals are baserunning, you say. What about Enos Slaughter's "Mad Dash", Lou Brock (the all-time NL stolen base king), and the Running Redbirds of the 80s? I would have said there is no need to convince me, but after the great October Collapse and a pair of columns by ESPN.com columnist and Kansas City Royals fan, Rob Neyer, this week, I'm not so sure anymore.
Yes, Neyer seems to secretly relish taking the Redbirds down a notch whenever he has the chance. In fact, in his columns on baserunning this past week he managed to take jabs at Lou Brock's legacy AND question whether he really was much of an effective base-runner.
Wills played his last game in 1972 (and managed to steal only one base in 71 games). And while there really wasn't another player quite like him – Lou Brock stole a lot of bases, especially in 1974, but was he really that exciting? – a style of baseball was invented (or perhaps reinvented) by Pete Rose and his teammates in Cincinnati.
Neyer seems to have an inferiority complex when it comes to the Redbirds, possibly because everyone and their brother has told him that the Royals didn't deserve to win the 1985 World Series (thanks to Don Denkinger, the Redbirds' version of Steve Bartman), but I digress.
In this case he might have a point.
One my first memories as a baseball fan is attending Lou's last game in 1979. My devotion to Lou grew to the point that sometime within the next few years I would trade Rickey Henderson's rookie card for Bob Horner's because I believed in the power of The Franchise. Despite Rickey's prodigious accomplishments over the next two decades and his unwavering insistence on referring to himself in the third person, Sean never doubted that Lou was the King. After all, while Lou (12) and Rickey (13) are the only two players in Major League history with at least 12 seasons with 50 or more stolen bases, Lou did it 12 times in a row (Rickey only 7).
Then Sean saw this chart.
Name SB CS Pct. Rickey Henderson 1406 335 80.8% Lou Brock 938 307 75.3% Billy Hamilton 912 ??? Ty Cobb 892 ??? Tim Raines 808 146 84.7%
Suddenly, Lou didn't seem to compare to Rock Raines. Sure, he stole 100 more bases, but he was also caught stealing 150 more times, too.
So what's more important, the number of stolen bases or the number of times caught? If John Mabry had stolen 5 bases and been caught zero times (Sean knows, not likely), would you rather have him on 2nd with 2 outs than someone like Reggie Sanders who had stolen 25 bases and been caught 10 times? Probably not.
Fortunately, as Neyer points out, there's a measure for how important stolen bases are.
"Stolen Base Runs," a method devised by Pete Palmer and based on the assumption that 66 percent is roughly the break-even point, weights the steals and the caughts to arrive at the number of runs all those attempts really led to. From the 2004 edition of Total Baseball, here are the eight players with at least 60 career Stolen Base Runs:Name SB SBR Rickey Henderson 1406 159 Tim Raines 808 112 Willie Wilson 668 87 Vince Coleman 752 86 Joe Morgan 689 79 Davey Lopes 557 71 Lou Brock 938 68 Ozzie Smith 580 61
Wow. Lou drops to 7th! Vincent Van Go jumps to 4, just behind Willie Wilson. And there's the Wizard, just below Lou. Sean's mind is in a tizzy. All along he thought Lou was the pinnacle of Cardinal base-running, but it appears that not only do the members of the Running Redbirds have him trumped, but so does Davey Lopes.
There's only one problem. Neyer's numbers are wrong.
Oh, he's got the right numbers for stolen bases, but the Stolen Base Runs numbers are calculated incorrectly.
A little background on how Stolen Base Runs (SBR) are calculated, from Dan Fox at Dan Agonistes.
Stolen bases and caught stealing can also be taken out of the Batting Runs formula and be calculated separately as Stolen Base Runs (SBR) or Base Stealing Runs (BSR) as (.30*SB)-(.60*CS). Originally, the value of the stolen base and caught stealing was set at around .20 and -.35 respectively. However, Palmer was convinced by Dave Smith of Retrosheet to increase both the positive and negative impacts of the stolen base on the basis that they occur in situations where games are more in question. In other words, stolen bases are strategically more important and so have a greater impact on wins and losses. Not many people seem to buy this argument since runs and not wins are what is being calculated. Apparently, Palmer agreed and so in The 2004 Baseball Encyclopedia BSR is simply calculated as (.22*SB)-(.38*CS).
So Neyer's chart should read something like this:
Name SB CS Pct. SBR SBR Rickey Henderson 1406 335 80.8% 182 159 Tim Raines 808 146 84.7% 122 112 Vince Coleman 752 177 80.9% 98 87 Willie Wilson 668 134 83.3% 96 86 Joe Morgan 689 162 81.0% 90 79 Lou Brock 938 307 75.3% 90 71 Davey Lopes 557 114 83.0% 79 68 Ozzie Smith 580 148 79.7% 71 61
Granted, Lou's success ratio didn't magically improve, but it bumps him up to tied for 5th with Joe Morgan in Stolen Base Runs. Ozzie clocks in at 7th.
But what about the rest of Cardinals?
Here all-time Cardinal basestealers (modern-day stats only, as Caught Stealing stats weren't consistently tracked until after WWII) ranked by career SBR.
CAREER CARDINAL Player SB CS Pct. SBR SB CS Pct. SBR V Coleman 752 177 80.9% 98 549 115 82.7% 77 L Brock 938 307 75.3% 90 888 285 75.7% 87 O Smith 580 148 79.7% 71 433 102 80.9% 57 A Van Slyke 245 59 80.6% 31 104 26 80.0% 13 W McGee 352 121 74.4% 31 301 97 75.6% 29 L Smith 370 140 72.5% 28 173 63 73.3% 14 E Renteria 237 89 72.7% 18 148 50 74.7% 14 T Herr 188 64 74.6% 17 152 51 74.9% 14 R Lankford 258 117 68.8% 12 250 115 68.5% 11 J Javier 135 63 68.2% 6 134 63 68.0% 6 T Pendleton 127 59 68.3% 6 99 48 67.3% 4
No big surprises here, and note how top heavy the list is with Running Redbirds. Seven out of the top 11 are from the 1980s teams (Vince Coleman, Ozzie Smith, Andy Van Slyke, Willie McGee, Lonnie Smith, Tommie Herr, and Terry Pendleton. And it appears El Birdos just let one of the organization's best basestealers in history go. Edgar, good luck to you in Beantown.
Your 2005 Cardinals are a far cry from 20 years ago, but the team did manage to generate 7 Stolen Base Runs last year (as opposed to say, the 33 SBR the 1985 Cards scraped together).
Career Cardinals Player SB CS Pct. SBR SB CS Pct. SBR R Cedeno 213 61 77.7% 24 5 1 83.3% 1 R Sanders 283 106 72.8% 22 21 5 80.8% 3 L Walker 228 75 75.2% 22 4 0 100.0% 1 M Grudzielanek 117 41 74.1% 10 0 0 0.0% 0 D Eckstein 82 27 75.2% 8 0 0 0.0% 0 S Rolen 91 33 73.4% 7 20 8 71.4% 1 S Taguchi 7 3 70.0% 0 7 3 70.0% 0 B Hart 3 1 75.0% 0 3 1 75.0% 0 H Luna 6 3 66.7% 0 6 3 66.7% 0 A Pujols 13 13 50.0% -2 13 13 50.0% -2 J Mabry 7 12 36.8% -3 3 9 25.0% -3 J Edmonds 54 41 56.8% -4 28 17 62.2% 0
Remember the other day? When Sean was bitching about how Albert Pujols makes the comman man look pathetic? Sean thinks he's finally found something most people are better at than King Albert.
Sean's Career Stolen Base Runs = 0. Albert's = -2.
Anyway, much is made of the Cardinals' loss of defensive presence up the middle this offseason, but they also lost two very good baserunners in E-Rent (Career 237/89/72.7%/18) and T-Dub (Career 335/68/83.1%/48). Yes, Eck and Grudz are decent baserunners in their own right, and having Booger around for the better part of the season should help, but of the team's 7 SBR last year, 4 of them were created by Tony Womack.
So is the 2005 team going to be a better baserunning squad? Are they going to avoid the mistakes that may have cost them the Series in 2004 and go on to take their first title in 23 years?
These are questions Sean poses to you, dear reader. Tony La Russa does seem to have adjusted his mentality to more of a small ball/National League approach, but that doesn't mean the spirit of Wave 'Em Home Wendell isn't somehow inhabiting Jose Oquendo's body.
As for Mr. Neyer, he does rank two Cardinals in his list of all-time basestealers...
1. Rickey Henderson
2. Tim Raines
3. Vince Coleman
4. Joe Morgan
5. Max Carey
6. Willie Wilson
7. Ty Cobb
8. Lou Brock
9. Davey Lopes
10. Billy Hamilton
So, maybe he isn't so much pissed about the Don Denkinger fiasco but instead all of the money he lost investing in the Brock-a-brella.
More fun with charts...
Cardinals Team Stolen Base Records Since Last World Series Appearance
Year SB CS Pct. SBR 1982* 200 91 68.7% 9 1983 207 89 69.9% 12 1984 220 71 75.6% 21 1985* 314 96 76.6% 33 1986 262 78 77.1% 28 1987* 248 72 77.5% 27 1988 234 64 78.5% 27 1989 155 54 74.2% 14 1990 221 74 74.9% 21 1991 202 110 64.7% 3 1992 208 118 63.8% 1 1993 153 72 68.0% 6 1994 76 46 62.3% -1 1995 79 46 63.2% 0 1996* 149 58 72.0% 11 1997 164 60 73.2% 13 1998 133 41 76.4% 14 1999 134 48 73.6% 11 2000* 87 51 63.0% 0 2001+ 91 35 72.2% 7 2002* 86 42 67.2% 3 2003 82 32 71.9% 6 2004* 111 47 70.3% 7*1st Place +Tied for 1st/Wildcard
Best Teams by Stolen Bases Per Game (SB/G) in History of MLB
Team G SB CS SB/G SBR Cincinnati Reds 18294 16086 --- 0.88 St. Louis Cardinals 18296 15840 --- 0.87 Kansas City Royals 5695 4857 2146 0.85 253 Los Angeles Dodgers 18271 15434 --- 0.84 San Francisco Giants 18246 15008 --- 0.82 Washington Nationals 5702 4564 2058 0.80 222 Chicago Cubs 18261 14121 --- 0.77 Houston Astros 6842 5267 2409 0.77 243 Pittsburgh Pirates 18254 14004 --- 0.77 Florida Marlins 1877 1414 627 0.75 73 Chicago White Sox 16170 11857 --- 0.73 Seattle Mariners 4411 3129 1480 0.71 126 Philadelphia Phillies 18263 12907 --- 0.71 San Diego Padres 5706 3983 1927 0.70 144 Colorado Rockies 1881 1310 661 0.70 37 Toronto Blue Jays 4413 3050 1405 0.69 137 Atlanta Braves 18223 12574 --- 0.69 Tampa Bay Devil Rays 1131 774 349 0.68 38 Milwaukee Brewers 5702 3766 2035 0.66 55 Oakland Athletics 16137 10456 --- 0.65 Minnesota Twins 16182 10445 --- 0.65 New York Yankees 16150 10138 --- 0.63 Detroit Tigers 16200 9931 --- 0.61 Cleveland Indians 16176 9894 --- 0.61 New York Mets 6833 3900 2115 0.57 54 Anaheim Angels 7000 3967 2309 0.57 -5 Baltimore Orioles 16176 9023 --- 0.56 Texas Rangers 6987 3877 2149 0.55 36 Arizona Diamondbacks 1133 599 275 0.53 27 Boston Red Sox 16162 7872 --- 0.49Posted by Sean at January 23, 2005 01:45 AM
Nice post. Those all-time SBR lists basically are limited to the 70's and 80's. It makes me wonder if the recent home run binge will pass as well.
Posted by: Rob at January 23, 2005 09:51 AMCan't stand Neyer, and not just because he seems to love knocking the Cardinals. Even when he's right, he's so smug, it makes me angry if I agree with him.
But I do remember one thing he said where it seems like baseball goes in cycles. Where home runs dominate, then pitching then baserunning, the back to home runs or something like that. I'd really like for the stolen base to come back. Doesn't look too promising though.
Posted by: CalvinPitt at January 23, 2005 10:57 AMSo how did Neyer compute his SBR's (or how did TB compute the numbers which he quoted)? I can't seem to reproduce them with any of the weights....
Posted by: Jon K at January 23, 2005 11:36 AMGreat Post. So how did Neyer compute his SBR's (or how did TB compute the numbers which he quoted)? I can't seem to reproduce them with any of the weights....
Posted by: Jon K at January 23, 2005 11:36 AMI don't see anywhere in your article that calls into mind the situation. A caught stealing for the first out is likely to be much more costly than a caught stealing for the third out because the chance of a big inning developing is much more likely with a runner at first and no outs as opposed to a runner at first with two outs. In the latter case, it will take two hits (or possibly a long hit) to score a run. In the no-outs situation, a lot of things can happen to help move the guy around.
This stat is sort of like the zone rating stats people are using to define defensive abilities. They represent a good start but don't seem to come up to substantial answers, at least in my mind.
Posted by: Steve T. at January 23, 2005 11:47 AMJon -
I hadven't heard back from Rob Neyer yet regarding where he pulled the numbers from; but judging from his busy schedule, I'm left to assume he just copied them from Total Baseball. I haven't read this year's TB, but perhaps in all of the haggling about the value of SBR (.20-.30 for SB and .35-.60 for CS) there was an arbitrary set of numbers used prior to publication.
Steve -
From what I understand, situations were taken into account when creating these averages; but as I mentioned above, they don't seem to have come to full agreement quite yet.
These stats, just like any, should be taken with a grain of salt. As you said, they are a nice start, but they don't reveal the whole picture. What would be nice is if we could somehow combine SBR with OBP and runs scored... then Billy Beane might have a conniption fit trying to figure out who to go after first during the 2005-06 offseason.
Sean
Posted by: Sean at January 23, 2005 12:21 PMI'd guess one of Billy Beane's minions, probably Huckabay, already has done something like that.
Posted by: Rob at January 23, 2005 12:48 PMI wonder if the composition of the list---everyone but Brock and Joe Morgan (who deserves and mention as among the best stealers, by any neasure, of all time) played all or almost all of their careers after Brock retired in 1979.
Eight times during Brock's 18 full season the NL team average in runs scored was at or below 4.01 runs per game, compared to just 4 times since then at or below that level. This suggests that perhaps outs were more to be gambled with, as they were "worth" fewer runs. Today, with team RPG at about 4.7, those outs "cost" about .174 runs (.174 x 27 = 4.7). From 1963 through 1969, the average NL team scored just 3.89 runs per game, or .144 runs per out. Getting thrown out in the mid- to late-'60s cost about 20% less, run-wise, than it does today, which makes sense when you consider the nature of today's game as exemplified by the 2004 Cardinals: with Larry Walker, Albert Pujols, Scott Rolen and Jim Edmonds batting 2-3-4-5, how much would it burn you to see the leadoff hitter caught stealing when those guys, on average as a group, will homer almost once per game?
In Brock's era, with scoring down and runs less prevalent, the guys behind you were much less likely to be jacking doubles and homers, and the relative preciousness of a single run made 1-run strategies more worth pursuing. It's why a sac bunt by Tony Womack makes little sense in the 3rd inning of a close game, but might make terrific sense in the bottom of the 9th of a tie game, when a single run is worth pursuing.
Posted by: salvo at January 24, 2005 05:24 PMSteve, what you seem to be looking for is an expected run table. There are various such tables floating around... you can find one put together by David Nichols at http://www.nssl.noaa.gov/users/brooks/public_html/feda/datasets/expectedruns.html.
These types of tables are really useful. For example, the expected number of runs scored in an innning where there is a man on 1st, none out, is 0.88. Now, if your baserunner successfully steals 2nd, the expected runs becomes 1.13, adding 0.25. However, if your runner is thrown out, the expected runs decreases to 0.26, subtracing 0.62. So, in these types of situations, you need about 2.48 successful steals for every CS in order to just break even.
Now, in the other situation you mentioned, if you have a runner on 1st, 2 out, the exp runs is 0.23. A successful steal yields an exp run of 0.33, an increase of .1 runs (a surprisingly low addition to me). A failed attempt, of course, yields exp runs of 0, a decrease of .23. The break-even ratio of SB to CS is 2.30. So, your assumption about the differential impact in these two situations is correct, although the difference isn't as pronounced as I thought it might be.
Notice that these ratios are pretty extreme... much more so than any of the ratios that Sean discusses in his article. Don't really know what to make of that except that Nichols uses data from 1984 to 1994 for his table, so perhaps his data is more reflective of the current trends (just based on the fact that Palmer's book came out in 1984).
Nichols' table also speaks to your point, salvo. He includes a column that lists the probability that at least 1 run will score in the inning given the particular situation. For example, if there's a runner on 1st, none out, the prob that 1+ runs will score is 0.435. If this runner steals 2nd, the prob rises to 0.633. But if he is thrown out, the prob falls to 0.159. So, a SB increases the chance of a 1+ run inning by 45.5% (percent, not percentage points) and a CS decreases this chance by 63.4%. With 2 out instead of none out, a SB increases the scoring chance by 75% but a CS of course descreases the chances by 100%.
This kind of chart is also really helpful in evaluating the efficacy of bunting. You'll see that there is no situation when bunting runners over will increase the number of expected runs, but in some situations it will increase the prob that 1+ runs will score. Of course that assumes a 100% success rate on the bunt and doesn't take into account differential hitting quality.
Posted by: Jon K at January 24, 2005 08:47 PMJon,
Thanks for your input. Is there anything in those charts that allows for the speed of the runner? For example, in the case of a runner at second base with two outs, it seems to me that Tony Womack has a heckuva better chance of scoring than, say, Mike Matheny.
Posted by: Steve T. at January 24, 2005 11:27 PMI once theorized to myself about why baseball might go in such cycles (longball -> pitcher's duels -> base theievery). It makes sense that with so many players jacking homers left and right, teams would shift their focus to drafting and developing pitchers that can keep the ball in the park, and maybe work on getting solid hitters that can also field their position well. Then, when the offense calms down, and you've got all these speedy fielders, it makes sense to start sending runners all the time, because as salvo pointed out, outs are less costly in terms of runs. Finally, with all these great pitchers running around dominating the game, it would make sense for organizations to again shift their focus towards drafting/developing offensive stars, and the cycle would begin again. But that's just my theory. I guess you have to talk to a lot of old GMs to find out if there's any truth to it.
Posted by: John at January 25, 2005 10:46 AM