Baseball Toaster was unplugged on February 4, 2009.
Stolen waters are sweet, and bread eaten in secret is pleasant.
—Speech of “a foolish woman” in the Bible: Proverbs ix. 17
MLB.com has an article that proclaims:
[L]ike so many trends that cycle through life and baseball, the running game is coming back.
They site the Marlins' World Series win from two seasons ago, the Padres' acquisition of ALCS stolen base hero Dave Roberts, Ozzie Guillen's re-imaging of the Chisox, and Omar Vizquel's ill-conceived signing in San Francisco.
I read the article and thought, "Hmm, could the stolen base be making a comeback unbeknownst to me?" Hey, if Terri Hatcher, bell bottoms, and Aerosmith can make comebacks, why not the stolen base?
So I brushed aside the anecdotal evidence and the annual "Dave Roberts is an All-Star" rhetoric and looked at the numbers. I ran a query to compare stolen bases and stolen base percentage along with batting ratios to a given team's runs per game ratio for every year since 1990. Here's what I found (the best statistic for correlating to R/G is bolded each year):
Yr | SB | SB% | BA | OBP | SLUG | OPS | HR/PA |
1990 | -0.344 | -0.121 | 0.533 | 0.455 | 0.885 | 0.874 | 0.619 |
1991 | -0.155 | 0.061 | 0.645 | 0.855 | 0.866 | 0.924 | 0.632 |
1992 | 0.038 | 0.338 | 0.676 | 0.854 | 0.846 | 0.915 | 0.638 |
1993 | -0.007 | 0.102 | 0.778 | 0.820 | 0.890 | 0.938 | 0.654 |
1994 | 0.267 | 0.171 | 0.737 | 0.867 | 0.908 | 0.940 | 0.663 |
1995 | 0.242 | 0.037 | 0.843 | 0.913 | 0.894 | 0.953 | 0.629 |
1996 | -0.062 | 0.026 | 0.833 | 0.906 | 0.938 | 0.961 | 0.677 |
1997 | -0.276 | -0.214 | 0.871 | 0.919 | 0.869 | 0.943 | 0.718 |
1998 | 0.073 | 0.149 | 0.841 | 0.946 | 0.890 | 0.944 | 0.720 |
1999 | 0.188 | 0.408 | 0.761 | 0.879 | 0.879 | 0.958 | 0.646 |
2000 | 0.052 | 0.206 | 0.797 | 0.941 | 0.869 | 0.929 | 0.599 |
2001 | 0.068 | 0.336 | 0.839 | 0.929 | 0.879 | 0.934 | 0.655 |
2002 | -0.053 | 0.368 | 0.832 | 0.833 | 0.914 | 0.935 | 0.638 |
2003 | -0.182 | 0.237 | 0.886 | 0.916 | 0.953 | 0.974 | 0.704 |
2004 | -0.167 | -0.151 | 0.802 | 0.875 | 0.928 | 0.970 | 0.716 |
The only trend that I see in the data is with the ratio stats. OPS took over as the power surge of the early Nineties took hold. By the end of the Nineties, Billy Beane's favorite stat OBP seemed to be eclipsing OPS, but then OPS took over again and Slugging quickly passed OBP.
Meanwhile, stolen bases and stolen base percentage have had very little to do with scoring over the last fifteen years. They had a slight correlation (esp. SB%) from 1999-2003, but slide to actually a very slight negative correlation last year.
If anything, the stolen base had less to do with scoring in 2004 than they had in the previous six seasons.
But maybe looking at the team stats gives an incomplete picture. What about leadoff men? Maybe having one that can steal an extra base from time to time leads to higher scoring?
Let's take a look at 2004. Here are the stats for all teams from the leadoff spot in 2004 sorted by Runs Per Game:
TEAM | BA | OBP | SLG | OPS | SB | CS | SB% | HR/PA | R/G |
Boston | .294 | .370 | .455 | .825 | 20 | 10 | 66.67% | 2.51% | .833 |
San Francisco | .276 | .359 | .480 | .839 | 12 | 7 | 63.16% | 3.29% | .827 |
NY Yankees | .276 | .347 | .446 | .792 | 9 | 11 | 45.00% | 2.57% | .759 |
Cleveland | .297 | .373 | .483 | .856 | 19 | 8 | 70.37% | 3.56% | .747 |
Philadelphia | .275 | .339 | .432 | .771 | 25 | 8 | 75.76% | 1.90% | .728 |
Atlanta | .260 | .320 | .386 | .706 | 29 | 10 | 74.36% | 1.94% | .722 |
Tampa Bay | .298 | .338 | .446 | .784 | 60 | 15 | 80.00% | 1.48% | .720 |
Baltimore | .275 | .341 | .388 | .730 | 28 | 13 | 68.29% | 0.76% | .716 |
Texas | .301 | .354 | .481 | .836 | 18 | 4 | 81.82% | 3.01% | .704 |
Montreal | .258 | .348 | .457 | .806 | 17 | 7 | 70.83% | 3.85% | .698 |
Anaheim | .291 | .350 | .377 | .726 | 23 | 7 | 76.67% | 0.52% | .691 |
Houston | .283 | .342 | .462 | .804 | 7 | 2 | 77.78% | 3.00% | .679 |
St. Louis | .293 | .335 | .377 | .712 | 27 | 7 | 79.41% | 0.92% | .654 |
San Diego | .288 | .334 | .395 | .729 | 11 | 5 | 68.75% | 1.30% | .654 |
Chicago Sox | .279 | .349 | .414 | .764 | 23 | 11 | 67.65% | 2.25% | .648 |
Seattle | .368 | .412 | .452 | .864 | 33 | 12 | 73.33% | 1.15% | .648 |
Chicago Cubs | .269 | .329 | .464 | .793 | 21 | 8 | 72.41% | 3.79% | .642 |
Los Angeles | .276 | .335 | .394 | .728 | 48 | 7 | 87.27% | 1.18% | .636 |
Oakland | .310 | .366 | .462 | .828 | 12 | 5 | 70.59% | 2.41% | .617 |
Florida | .337 | .383 | .429 | .812 | 39 | 21 | 65.00% | 0.67% | .617 |
Kansas City | .254 | .324 | .357 | .682 | 19 | 12 | 61.29% | 1.34% | .605 |
Cincinnati | .248 | .353 | .342 | .695 | 33 | 11 | 75.00% | 0.91% | .593 |
Milwaukee | .246 | .315 | .366 | .681 | 71 | 13 | 84.52% | 1.59% | .590 |
Colorado | .280 | .320 | .372 | .692 | 12 | 11 | 52.17% | 1.41% | .586 |
Minnesota | .293 | .362 | .424 | .786 | 12 | 5 | 70.59% | 2.22% | .568 |
Toronto | .281 | .335 | .419 | .754 | 6 | 4 | 60.00% | 2.52% | .553 |
Detroit | .277 | .309 | .377 | .687 | 28 | 17 | 62.22% | 1.56% | .549 |
NY Mets | .239 | .291 | .377 | .668 | 32 | 9 | 78.05% | 1.98% | .549 |
Pittsburgh | .301 | .364 | .382 | .745 | 12 | 8 | 60.00% | 0.53% | .540 |
Arizona | .247 | .296 | .374 | .670 | 11 | 12 | 47.83% | 1.73% | .488 |
OK, so how well do the key leadoff stats correlate to runs per game from the leadoff spot? Here goes:
BA | OBP | SLG | OPS | SB | SB% | HR/PA |
.173 | .393 | .592 | .583 | -.021 | .159 | .371 |
So, basically nothing correlates well to scoring from the leadoff spot. Surprisingly, slugging average correlates the best though not well at all.
Most important, stolen bases and stolen base percentage have apparently nothing to do with high scoring from the leadoff spot. They are the two lowest on the board. Besides the top stolen base total was for Milwaukee, one of the lowest scoring teams from the leadoff spot.
My conclusion? It was a slow news day at MLB.com. There's no empirical evidence that the stolen base is back. But I wouldn't mind if they did over the next few years if it helps the underrated Tim Raines get into the Hall.
OK, then "The SB is back!", I'll accept "The SB Is a Useful Strategy in Certain Situations". Can you get MLB.com to change it?
Secondly, the Rock should be in the Hall - maybe because I was so impressed by him watching all those Phils-Expos games in the '80's but he is, in my opinion, the most underrated "star" of m,m,my generation.
I'd agree. It's particularly galling given how overrated Maury Wills has become over time.
In fact, he was a key part of my favorite game that year. If I remember correctly, the Orioles had a sizable lead, but gave it all back. It was capped off when Williams homered, which prompted Armando Benitez to drill Tino Martinez in the upper back. All heck breaks loose. After everything settles down and Darryl is hauled off to the Bronx House of Detention (well, maybe just the home clubhouse), Tim Raines homered in the next at bat. It was hilarious.
Hey Mike, I know this has nothing to do with the topic but what's the status of Cole Hamels? My fantasy draft is less than a week away and I need expert advice.
Secondly, in MLB.com's defense, I think they were simply stating that several teams seem to be transitioning toward speed (i.e. White Sox). Your numbers certainly indicate that generally that strategy is less effective in scoring runs, but it doesn't say anything about whether teams are doing it. Wouldn't a better analysis have just looked at raw stolen base numbers, seeing if attempts have gone up overall, or at least for an increasing number of teams? That seemed to be what the article was saying, not that "stolen bases are effective again." I don't think MLB.com ever makes anything but quantitative statements.
"Wouldn't a better analysis have..."--Now, who's beating up on whom?
OK, you asked for it, you got it. Here are the average team numbers for SB, CS, and SB attempts since 1990, with the annual percent increase. I wouldn't say that era of Vince Coleman is returning any time soon:
Yr Lg SB CS Attempts % Increase
1990 NL 148.92 60.58 209.50 12%
1991 AL 104.93 54.14 159.07 -3%
1991 NL 137.58 67.42 205.00 -2%
1992 AL 121.71 61.43 183.14 15%
1992 NL 130.00 61.75 191.75 -6%
1993 AL 110.64 62.29 172.93 -6%
1993 NL 122.43 56.29 178.71 -7%
1994 AL 79.79 35.93 115.71 -33%
1994 NL 81.50 37.79 119.29 -33%
1995 AL 95.07 41.86 136.93 18%
1995 NL 114.43 47.93 162.36 36%
1996 AL 103.86 45.29 149.14 9%
1996 NL 127.50 50.64 178.14 10%
1997 AL 106.50 51.64 158.14 6%
1997 NL 129.79 60.07 189.86 7%
1998 AL 119.64 53.86 173.50 10%
1998 NL 100.56 46.94 147.50 -22%
1999 AL 104.43 49.21 153.64 -11%
1999 NL 122.44 51.88 174.31 18%
2000 AL 92.64 41.93 134.57 -12%
2000 NL 101.63 46.00 147.63 -15%
2001 AL 117.64 48.07 165.71 23%
2001 NL 91.00 45.94 136.94 -7%
2002 AL 88.29 41.36 129.64 -22%
2002 NL 94.63 43.94 138.56 1%
2003 AL 91.36 39.07 130.43 1%
2003 NL 80.88 36.56 117.44 -15%
2004 AL 89.50 40.93 130.43 0%
2004 NL 83.50 32.94 116.44 -1%
Cole Hamels won't make the team until mid-season unless they again pick up Paul Abbott--then it'll be later like Floyd this year so that the can forestall the inevitable arbitration even longer. They are already 6 deep in the rotation and it looks like Myers will again be given a spot, incredibly. Floyd is ahead of Hamels. And if the Phils' brass had half a brain among them, adson would be given a shot, but they don't so he won't.
Sorry, it was just an idea -- the MLB article didn't seem to claim stolen bases correlated to runs scored or anything, just that more teams were running again.
The numbers you posted are interesting. Attempts have declined, but SB% appears to be fairly constant, hovering around the "break-even" mark at 70%, meaning, as a whole, these stolen base attempts probably aren't costing any runs (regardless of whether the raw totals correlate to runs). An average or above-average stealer should be just as valuable now as 15 years ago, correct? There just aren't as many of them.
And thumping their offense is with an overall SLG of .458 (second in the NL in 2004). As BP shows, the had the third most homers in the majors last year when with 235 they hit a franchise record number of offerings out of the park. Yet BP criticizes them for only scoring so much runs on their balls hit out of the park - because 141 of those homers were solo.
Their logic is that by having a better team OBP there would be more multiple run homers hit. Statistically that is true.
However, your research here states that the Cubs were doing everything right when they focused on SLG - because SLG correlates better with R/G than OBP (but OPS is even better).
The reason is probably that the number of solo homers and multiple run homers hit correlates with OBP, but only so much. The rest is fluke. That of course could be an interesting study. Or not?
On another topic, Mike: Its that time of the year where my heart feels baseball in it. I think there could be something coming up of ... you know what. Something we had done before. Together. At your blog. Still interested?
Comment status: comments have been closed. Baseball Toaster is now out of business.