Stolen base weirdness in 2011
Posted by Andy on May 2, 2011
Some observations about team stolen base totals in 2011 through Friday's games:
- The Orioles have attempted only 6 stolen bases so far this season. That puts them on a pace of about 40 attempts for a full season. The last time a team was even close to so few attempts was 2008, when the Padres attempted only 53 steals. No team had fewer the 53 going all the way back to 1990.
- The Braves have also attempted a fairly small number at just 13, and they have succeeded only 5 times (the same number as the Orioles). The low success rate of 38% is horrible. The worst team last year (in terms of stolen base percentage) was the Giants at 63% on 87 attempts. Of course, that didn't prevent them from winning the World Series...
- The team with the best success rate--with 11 steals and just 1 caught stealing--is the Twins. And the Twins have a very disappointing 9-16 record and are in last place.
- At the other end of the spectrum, the Padres have the second-most steals so far, with 32. However, they have struggled badly to score runs with just 2.77 per game, by far the worst mark in the majors (an an unsusntainably low number.) The steals don't seem to be helping them push runners around, or maybe the only guys getting on base for them are the base-stealers.
- The Kansas City Royals have 33 steals with just 5 time caught. They have the most steals and one of the best success rates. This has probably helped them to their 13-13 record
May 2nd, 2011 at 8:05 am
how about the Jays? The most HR's last year to one of the top base stealing teams this year...
May 2nd, 2011 at 8:12 am
Jays lead with 8 steals of 3rd, but also have 3CS at 3rd. Don't think the return there is very good.
May 2nd, 2011 at 8:26 am
Add to this the Wall Street Journal story about Chris Snyder closing in on the record for stolen base futility and you have an interesting list to ponder so early in the season.
http://pinetarandbrickbats.blogspot.com/2011/04/braves-and-scandal.html
May 2nd, 2011 at 8:56 am
The Kansas Royals lead the AL in Hits, Doubles; are second on Total Bases; fourth in OPS. I suspect that has something to do with their 15-13 record.
May 2nd, 2011 at 9:46 am
Don't I remember reading that stolen bases actually has a negative correlation with winning? Or did I dream that?
May 2nd, 2011 at 9:48 am
Ctorg, I think they do have a slight negative correlation, and that's because the league totals for SB% is usually slightly lower than the break-even point for what makes SB valuable. But any given team with an above-average SB success rate may be benefiting from it in terms of winning games.
May 2nd, 2011 at 10:21 am
According tot he NL batting splits; NL Pitchers have 4 stolen bases. 1 each by HOU, STL, COL, and SD,
However looking at the team stat pages, I can't find a player on any of those teams with an SB. As far as I know no position players have pitched yet this season, so there may be a bug in the stats.
Or it may be I am missing something obvious.
May 2nd, 2011 at 10:23 am
I was also skeptical of the role of base stealing in KC's (now) 15-13 record.
But after looking at the game logs, I'm slightly less skeptical.
In their 15 wins, the Royals have 24 SB and 2 CS.
In their 13 losses, they have 11 SB and 3 CS.
They have 7 wins by 1 run. In those games, they are 14-0 in steals.
I'll try to look at some SB WPAs in a bit.
May 2nd, 2011 at 11:15 am
Re: impact of base stealing on KC's wins:
I'm not willing to slog through all their games, but here's a look at SB in their seven 1-run wins, plus a 6-SB game they won by 3 runs. Instead of WPA, I will use describe in words the steals' ultimate effect on the game outcome, comparing the actual results of SB to innings reconstructed without the SB as best I can. I've put them in order of impact, from most to least, with a big unknown at the bottom.
Note that there were no CS in any of these games, which makes the accounting much easier.
April 29 -- 1 SB in a 4-3 win. Huge impact: Trailing 3-2 in the 8th, KC got runners on the corners with 1 out. Jarrod Dyson pinch ran at 1B and stole 2nd; the throw got away, allowing the tying run to score and Dyson to reach 3rd, whence he scored the ultimate winning run on a shallow sac fly; no further events in the inning would have scored those runners.
April 5 -- 3 SB in a 7-6 win. Big impact: a 2-out double-steal produced an extra run when the next batter singled them home.
April 15 -- 1 SB in a 6-5 win. Impact: Ahead 5-1 in the 6th, another their 6th run as a direct result of a SB by Alcides Escobar. Seattle scored 4 times in the last 2 innings, but the Royals held on, thanks to that (seemingly innocuous) insurance run.
April 21 -- 2 SB in a 3-2 win. No direct impact; but speculatively, a possible big impact: Trailing 2-0 in the 8th, Getz drew a 1-out walk and stole 2nd; after another out, he stole 3rd. Alex Gordon and Billy Butler drew walks; it's possible that they were pitched more carefully with bases open. Francoeur then singled home a run. KC scored 2 in the bottom of the 9th to win.
April 19 -- 5 SB in a 5-4 win. Minimal or no impact. Three SB came in the 1st inning, in which they did not score. Two came in the 4th, when leading 2-0; one of the base stealers scored, but he probably would have scored without the SB.
April 2 -- 1 SB in a 5-4 win. No impact.
April 1 -- 1 SB in a 2-1 win. No impact, though there was potential for a big impact: With the game tied in the bottom of the 8th, Mike Aviles walked; after 2 outs, he stole 2nd and took 3rd on a throwing error, but Billy Butler struck out. They won in the 9th on Kila Ka'aihue's HR.
April 3 -- 6 SB in a 12-9 win. Possible big impact, but impossible to say for sure: Five of the steals had no discernible effect on scoring; one of them scored, but would have scored without the SB. However, the last SB might have had a big impact. Leading off the bottom of the 12th, Chris Getz reached on an error and then stole 2nd. The next batter, Francoeur, grounded to 3B, which -- as far as we can tell from the play-by-play -- could have been a DP if Getz had been at 1st base. As the inning progressed, Getz made the 2nd out at home on a GB-FC, but Matt Treanor then homered for the win. The play-by-play gives no clear indication whether Francoeur's GB was a potential DP; the description says "Weak 3B," and Getz took 3rd on the play, which initially makes me think "slow roller" -- but the account of the previous inning also has a GB-5 described as "Weak 3B" that went for a 5-4-3 DP.
May 2nd, 2011 at 11:17 am
(Sorry about the phrasing errors -- hasty rewrite!)
May 2nd, 2011 at 11:25 am
Nice research, JA. It's probably fair to say that overall, Royals' SB have been worth maybe 1 to 1.5 full wins so far this season, which is a huge amount.
May 2nd, 2011 at 12:20 pm
By the way, have the Royals resurrected the "designated pinch runner" concept?
Jarrod Dyson already has pinch-run 7 times this year. He has 5 SB and 4 runs in that exact role.
Unfortunately, we may not see him doing that for a little while, as he injured his ankle yesterday while trying to stretch for a double.
P.S. In one game that Dyson entered as a pinch-runner, he scored twice in the same inning, as the Royals batted around and he came to bat, got a hit and scored again. As a matter of notation, how would we describe that plate appearance? He didn't pinch-hit, strictly speaking.
P.P.S. Unless I'm mistaken, pinch-running appearances are one of the few things not clearly accounted for in the B-R splits; they seem to be lumped into a mysterious "Other" category in the Defensive Positions splits, and I can't find a specific explanation of what "Other" includes. Also, pinch-running appearances for the DH apparently go into the "As DH" category; on the one hand, this treatment is logical, since by rule a pinch-runner for the DH is the new DH, but on the other hand, it's unfortunate that all pinch-running appearances can't be found under one heading.
May 2nd, 2011 at 12:55 pm
@3, according to my PI search the top ten in lifetime PA's without a SB are:
1. Russ Nixon, 2714
2. Johnny Estrada, 2244
3. Aaron Robinson, 2189
4. Chris Snyder, 2102*
5. Ryan Garko, 1752
6. Ted Lyons, 1726
7. Phil Niekro, 1707
8. Javier Valentin, 1663
9. David Ross, 1622*
10. Al Ferrara, 1573
* Active
Nixon also has the most CS without a SB, 7.
So Snyder may take over second place this year, but has a ways to go to catch Nixon. Ross will likely move up a few spots this year.
The most CS without a SB in a single season was Pete Runnels with 10 in 1952.
May 2nd, 2011 at 1:25 pm
MLB always tends to fall below the break-even point when looking at all SB attempts, but these have a large flaw in them. Botched hit and runs are counted as steal attempts. I'm not sure if these have ever been counted systematically, but lots of players get "caught stealing" on plays when they actually weren't trying to steal a base. Especially for the low total guys, who might steal 3-4 bases and get caught 3-4 times a year, the bulk in both categories might be made up of these. The problem with identifying the CS in this situation is that sometimes they are successful, since bad throws/catches are made often enough to give them a decent shot. There almost needs to be another sat - managerial CS and managerial Steals.
I think if you take the top 20-30 guys in each league and look at their success rate, that's a fairer summary of what the true success rate is. Some of the top volume guys still fall below the break even point, but they do give a more reasonable guess as to the chances of a "straight steal" than league totals.
May 2nd, 2011 at 2:11 pm
the league totals for SB% is usually slightly lower than the break-even point for what makes SB valuable.
In recent years, MLB SB% has been around 73%, which is well ahead of whatever the break-even point may be.
May 2nd, 2011 at 2:24 pm
My understanding is the break-even point is between 75 and 80%.
May 2nd, 2011 at 4:25 pm
Where did you get that? I've never heard of anything that high.
Here's one run-expectancy table from the peak of the offensive era: http://www.tangotiger.net/RE9902.html
If you look at the difference in run-expectancies based on whether a guy on 1st with 0 out successfully steals second or is caught, you get a break-even rate of about 73%. Stealing other bases or with more outs have different rates. But if you look at the situations when most SB attempts actually occur (close games, when playing for one run), I believe the break-even point actually drops based on the win-expectancy.
Each situation has its own break-even point based on various factors that we can't measure precisely (opposing pitcher, following hitters, etc). And no doubt there are some which are at least 80% (e.g. trying to steal 3rd with 2 outs). But it was my understanding that overall, break-even is somewhere around 67%.
May 2nd, 2011 at 4:38 pm
Not to make a big deal of this, but ... Chris Snyder isn't all that close to the no-stolen-base record.
He's moving up the ranks, but he's still almost 300 games and over 600 PAs behind Russ Nixon, who never swiped a bag in 906 games and 2,714 PAs.
May 2nd, 2011 at 5:56 pm
@ #7
I don't think you are missing anything. There have been 297 stolen bases so far in the NL, all other splits seem to reflect this, but stolen base splits per position exceed 340 stolen bases. Not sure what the problem is, but there definitely is a problem in the accrual of stats by position. I didn't total up the other stats, but suspect there could be some duplication there as well.
May 2nd, 2011 at 7:45 pm
Worst baserunner this year = Juan Pierre?
He's been caught 7 times in 12 attempts. Batting leadoff with a .581 OPS, if this continues, Guillen is a confirmed moron. For 7 CS, he probably needs to have at least 30 sb's for his running to have any positive value. Pierre's running game has been one of the most overvalued I've ever seen. All 3 years that he led the league in steals, he also led the league in CS. And if you include this year, he's been the league leader in CS 7 times!
All around the league I look at teams and constantly see some of the worst hitters batting 1st or 2nd. It blows my mind.
May 2nd, 2011 at 10:19 pm
@20, Jimbo -- I'm no fan of Juan Pierre or Ozzie Guillen, either. But leading the league in CS, even in 6 full seasons, is not necessarily an offensive sin. Rickey Henderson led the league in CS each of his first 4 full years; it doesn't say anything about his success rate or the net value of his steals. Many prominent base stealers have led the league in CS as often as they led in SB, or more often.
Pierre's running game is overvalued because his career success rate is just under 75%. In the offensive context he has played in, stealing lots of bases at a 75% clip hasn't added much to his teams' scoring -- certainly not enough to offset his modest OBP and total lack of power.
May 3rd, 2011 at 12:06 am
I bet no player has ever played as many career games, and earned as much salary money, as Pierre, and had such a low career WAR. He's been leading off full time for over 10 years and has been making 9 mill a year on his current contract, and for his career has only 13 WAR.
I'm guessing he's a good candidate for highest career hits/war for a player with over 1500 hits.
May 3rd, 2011 at 12:15 am
On top of the 5/12 stealing bases and the .581 OPS, Pierre has 4 errors in centerfield, and no assists.
It's tough to have a worse 30 games to start the year than that, across the board.
Of course, Pierre, and stolen bases, are fun to watch. I don't have time to watch many games, so I tend to watch the video highlights of every game. It always annoys me that non-run scoring plays are so de-emphasized in highlights. If a player hits a bases empty triple and then the next guy hits a sac fly, the highlight reel shows us the sac fly and omits the triple usually. Likewise, we almost never see stolen bases in the highlight reel, but I'm sure all of us generally find both the SB and the CS to be exciting plays. SB and CS should be shown in highlight reels more often IMO.
May 3rd, 2011 at 12:22 am
Pierre was thrown out again tonight, and is not 5/13 stealing. The only time he gets on base in the game, and he wipes himself off.
I wonder at what point he may have to shut his running game off. This isn't the 80's, when a player could steal 25/40 and have people think his running game is worthwhile.
May 3rd, 2011 at 12:45 am
Pierre is racking up an RBI for every CS so maybe Ozzie is just hoping he'll continue that trend. Pierre was actually "decent' last year. At least he got hit a lot and his SB% was well over 50% (68/18). I wonder if he is tipping off the catcher that he is going to run (which seems like every time he gets on base). And he already has more errors this year than in most seasons. I'm surprised every boxscore to see his name there in the one slot.
May 3rd, 2011 at 12:53 am
Hanley Ramirez stole a base today to improve to three steals and four times caught stealing.
It's rare for players to have a less than 50% success rate now but it used to be quite common in the "old days."
Pierre is also striking out more often than he used to, which is one aspect of his game I've always liked.
On the bright side he hasn't hit into a double play yet this year (Have I just jinxed him?)
May 3rd, 2011 at 6:14 pm
I'm guessing [Pierre]'s a good candidate for highest career hits/war for a player with over 1500 hits.
Alfredo Griffin had 1688 hits while compiling -2.4 WAR.
Pierre currently has about 140 hits per WAR for his career. There have been 69 players in MLB history with at least 1500 hits and more than 100 hits per WAR.
Pierre is the active leader in H/WAR among players with at least 1500 hits.
May 4th, 2011 at 2:48 am
Thanks for the info Johnny.
I'm guessing each previous offensive era had some slap hitters with many singles and not much WAR.
Didn't think it would be 69 though, that's far more than I expected.