The Mockingbird

Mostly Whining

with 3 comments

In case you missed it, John Brattain mentioned the ‘Bird over at the Hardball times. Apparently that whole “if you don’t have anything nice to say…” line my mom fed me when I was too young to know any better wasn’t meant to be followed if your goal is to start interesting debates or finally see your name in the bright lights of a highly respected baseball publication. I wonder what other sage advice I should ignore. Maybe if I start running with scissors I could still make the big leagues??

Anyway, John has considerably softened his line on small ball, but as could be expected, I still disagree with a lot of what he says. Here’s my response to some of the points he makes in his latest article:

The “new testament” tells us, in effect, that you shouldn’t give away outs. Fine, so which situation constitutes giving away an out?

a) an out is made and no base runners advance.
b) an out is made, and one or two base runners are 90 feet closer to scoring.

What the “new testament” really says is that bunting is not a good way of scoring runs in the long run and should be avoided in more situations than most people think. Painting the Jays as a team that is dead set against the bunt, no matter what the situation is setting up a straw man. In fact the Jays are only a couple of bunts below league average. That’s not bad for a team that hasn’t had a lot of runners on base, and has a core that wasn’t made for small ball.

(Vernon/Overbay/Glaus/Rios/Thomas have a collective 9 bunts over their careers- and for good reason. I know, I know…every member of the team should be able to execute the fundamentals. But bunting is not easy- it’s an art and not something that can be picked up halfway through a rotten season. And if these guys are having anything but the worst seasons of their lives, you don’t want them laying one down no matter what the situation.)

That brings us to the bottom third of the lineup, which as Brattain points out, was a black hole this season.

However, while the regulars were batting .265/.335/.445, the scrubs were adding more injury to injury by batting .231/.275/.325 (55 OPS+). In other words, somebody cloned Cesar Izturis twice and the trio is at the bottom of your batting order. In baseball parlance, that’s called an “outmaker.” The Jays’ regulars were about league average (AL: .270/.338/.422) but the extras cost the Jays too many outs and got nothing in return.

It’s absolutely true that bunting becomes more appealing with hitters far below average. But the Jays weakest hitters haven’t been held back from bunting. In fact, John McDonald is 2nd in the major leagues in bunts per plate appearance (not including NL pitchers). Even guys like Phillips/Fasano/Clayton bunted at a high rate- over a full season (550 plate appearances) they would have 7/11/8 bunts respectively- eight is enough for top 10 in the American League.

And while it’s true that the worse your team is hitting, the more reasonable bunting gets, the Jays did not ignore that. In the first half, they bunted 14 times, 4 below average. In the second half, with weaker offensive numbers, they have already bunted 16 times (one above the AL average).

If the Jays are batting .224/.322/.362 w/RISP and two out in almost 700 plate appearances, then if you’ve got a man on third with less than two out, it’s close to now-or-never insofar as getting runs on the board. Gibbons should act accordingly.

As I’ve said before in these exchanges, I don’t think you can manage a team based on these kind of splits. And how would you, anyway? To compensate for their poor hitting with two outs, the Jays are hitting .310 with RISP and fewer than two (and a whopping .317 with a runner on third base). Even if you’re reacting to these kind of splits, why would we want to take the bat of their hands when it’s the only time they are hitting?

Once this particular epiphany has occurred, what do you do? Do you:

a) give up and concede defeat if the other team scores four runs?
b) refuse to apostatize from the GM’s philosophy and go down hacking with runners on base with two or three batters who aren’t within hailing distance of league average?
c) take some risks, put some pressure on the defense and try to sneak in a precious run? Is it a gamble? Sure, but no more so than letting consecutive batters with an average .233/.275/.325 line swing away, hoping that they’ll drive home base runners with a hit or, failing that, not give up an out.

This is a false choice frequently brought up around small ball. Either change to a new strategy, or concede defeat. The Jays came into the second half within striking distance and with a finally healthy lineup, thinking that it would finally score some runs. Letting it swing away was not accepting defeat. I think panicking and micromanaging and making changes just for the sake of change is really giving up on the team.

Look at the Yankees- back in May, the exact same arguments for small ball were being thrown around and Torre was being blamed for not doing anything. Sometimes the courageous thing to do is stay the course, even if in hindsight the results never came.

The Blue Jays’ brain trust had a large enough sample size to know that the offensive philosophy they opened the 2007 season with lacked the personnel to execute it due to injuries and slumps. Being in third place, they had little to lose at this point but adopted a risk-averse approach to the final 61 games.

This is another thing that station-to-station is often accused of being- conservative and boring, unwilling to take risks. But what the standard run-expectancy argument tells us about bunting is that it isn’t a risk, giving you a small chance of scoring a lot, it’s a strategy that overall gives you bad odds and leaks runs.

For example, playing high-stakes poker is a big risk, but it might pay off if you do it well. Bunting for the most part is like playing slot machines – a flat out bad gamble. Every time you take your chances, on average you don’t get as many dollars/runs as you started with. If you make a bad gamble often enough, it doesn’t matter how smart you are- you will lose your money.

One person I was debating with on a blog discounted the relative importance of hitting with RISP/two out (a major Blue Jays bugaboo) over any other RISP situations. However, there is a difference: With less than two out and a runner on third, you still can cash him in via a number of means. He can be driven in with a hit, or a sac fly, or a sac bunt, or a wild pitch, or a passed ball. With two out, you have fewer options; generally, you have to get a base hit or your half of the inning is over.

Yeah, that was me…I still don’t buy it. True, with two outs, you need a hit or you’re not going to score, whereas with fewer you can score other ways. But that doesn’t have make hits with two outs worth more, it just means there’s more value to deep fly balls, etc, when there aren’t two away.

Let’s say I told you that you could go back over a season and change all your hits in the year with runners in scoring position to hits with two outs and RISP. Same number of run producing hits overall, but now everybody whiffs with 0 or 1 out but comes through in the “clutch”. Would you expect the team to score more runs? Win more close games? But…your RISP with two outs would be AWESOME! It just doesn’t matter to anyone other than a fan getting disappointed because the inning is over and the team didn’t score.

In the Jays’ case, the 15 games under consideration were chosen because they wasted a good effort by the pitchers (3.91 ERA in the 15 games). Obviously, an extra run or two in those contests would have been huge. According to the new testament, you don’t give away outs to score runs, yet in a lot of these games (as well as games in the first half), a single run was worth more to the Blue Jays than any other club.

Getting back to those 15 games lost by one or two runs with the pitching staff averaging less than four earned runs given up, here’s the thing: The Jays were—to state the obvious—0-15 in those games. Suppose the Jays took some risks. They certainly would’ve run into some outs, in which case they’d still lose. On the other hand, if a few of those gambles pay off, you end up with a few more wins.

First, I think it’s misleading to break things down and look only at games that the Jays didn’t score and lost by a slim margin. It’s the same sort of begging the question that Brattain does back in his original article when he drops the bombshell that the Jays grounded into DP’s a lot in situations where they didn’t score. Yes, you’re going to see a lot of runners stranded in those games, but that’s because…wait for it…those are the games where the Jays didn’t score.

If there were some way to go back to only those games and take chances with nothing to lose, that would be great, but no…you would have to do that every close game, including the times we let Aaron Hill swing away and instead of grounding into one of his club-leading DP’s, he hit one of his best-Jay-second-baseman-of-all-time doubles to win the game.

They Jays played a lot of close games this season, and if there was some way to win significantly more of them, it could make a big difference. But teams that bunt more don’t win considerably more close games (it’s mostly luck). 4/6 of the top bunting teams in the AL this season were worse in one-run games than they were overall, while the Blue Jays were better (with a record of 28-24). It wasn’t a big problem this season, and no team or strategy wins them at a very high percentage.

Finally, even the most bunt-happy team doesn’t find that many situations to bunt. You have to pick your shots. It needs to be late in the game, a runner on third, at least less than two out, and a relatively weak hitter at the plate. Even the least SABR of the teams has to pick their shots. The Texas Rangers lead the American League in bunts. And they bunt once more every 6 games than the Jays. So over those games the biggest bunters in the league would have bunted about twice more often.

That makes every run precious.

To illustrate why not all runs are created equal despite being of identical numerical value: Suppose you had a homeless family and Bill Gates, standing side-by-side, and you had a $50 bill that you wished to give to one of them. To Bill Gates, it’s not going to mean a heck of a lot; he probably uses $50 bills to, er, “clean up” when the ol’ platinum bidet is being polished. To the homeless family, it means they get to eat that day and possibly tomorrow as well.

Now, if you let the numbers simply speak for themselves, they tell you that $50 and will purchase $50 worth of goods for both the homeless family and Mr. Gates. However, if you look at the situation at hand, you can see that the $50 is of larger value to the homeless family.

I agree that every run is precious when you’re not scoring a lot…but what does that have to do with bunting? The argument is that when you do “whatever it takes” to score, you actually end up scoring a lot less, and don’t even force across the one run that much more often. Take a look at this table, which shows the chance of scoring one run from every situation (from Baseball Between the Numbers, Jonah Keri, page 129).

Runners 0 Out 1 Out 2 Outs
Empty 28.0 16.5 7.1
1st 41.7 27.2 12.7
2nd 62.5 41.0 22.9
3rd 82.7 66.1 25.4
1st and 2nd 61.6 41.4 22.8
1st and 3rd 84.6 64.5 26.8
2nd and 3rd 86.1 67.4 26.6
Loaded 85.6 65.4 30.7

Some interesting results for a successful bunt:

  • 1st, 0 out -> -.7%
  • 1st, 1 out -> -4.3%
  • 1st and 2nd, 0 out -> +5.8%

So with a runner on first and an average hitter up, it’s never a good idea. And even if there is a successful bunt (and how many times have the Jays picked up a lead runner on an attempt this year?) in the best possible situation (runners on first and second, nobody out), the chance of going up only goes up 5%, while the average number scored drops considerably (.15 of a run).

So if the Jays bunted 20 times with runners on 1st and 2nd, nobody out, they would score one run about once more often. But over those 20 times, they would also score about 3 fewer runs. Ouch!

This team should be playing October baseball but it is not, because …

What really got me into this debate in the first place was what I thought was an exaggeration of the effect bunting could possibly have. And mentioning October is without a doubt an example of that. Pythagorean projection says that the Jays (even if they didn’t undershoot it by 4 wins as they have this year) would need 753 runs scored to keep pace with the Yankees this year. There is no way that any strategical change could “scratch out” 59 runs for for the Jays.

…it failed to scratch out runs and instead, chose to hope for big innings, which were few and far between.

And this is another misconception about station to station- that the whole idea is playing for a huge inning, and if that isn’t happening, then you should settle for one run. But you don’t have to be mashing home runs or scoring runs like the Yankees- bunting is a inefficient way of scoring even for an average team, it just gets worse if you’re putting up huge numbers. You don’t only cut down the chance of putting up big numbers, you turn 2-run innings into one, three into two, etc.

Ok, I’m finished. I hope I don’t sound too rabid, or like I’m saying that Brattain is totally wrong, because I agree with his argument against predictability and playing by routine or received wisdom. What I’m most firmly against is the idea that the Jays as missed out on a lot this year with a lack of strategy, or that they stubbornly refuse to change to meet circumstances. Small ball is a mirage that appeals in time of frustration. It’s very hard to effectively add a lot of it to a normal lineup, and doesn’t have the power to turn a bad team into a good one.

Written by halejon

September 23, 2007 at 2:09 am

Posted in Uncategorized

Tagged with

3 Responses

Subscribe to comments with RSS.

  1. Oh yeah, for what it’s worth, I don’t think a chance in strategy would’ve turned the Blue Jays from a poor offense to a good one–I felt it would turn a sucky offense into a less sucky one. To use an analogy: There are two things that can run over your living room carpet: a vacuum cleaner, and an F5 tornado after your roof has been torn off. Both suck, one sucks dangerously, one is somewhat useful.

    I think the Jays could’ve turned their offense from an F5 to an Electrolux. The difference could’ve been huge. Look at the Phillies this year: league average pitching probably would have given them a playoff spot by now. It would mean more to the Phillies than it would to any other team.

    Likewise with the Jays. They didn’t need a good offense, just one around the AL average–at least insofar as run scoring goes.

    Also: you wrote that I’ve “softened” in my small-ball approach. I haven’t, because I’m not a “small-ball” kind of guy. All I was advocating was a more realistic outlook to run production and using the outs generated by the outmakers into something useful. Outs are never good (at least for an offense) but if you’re consistently in a position where outs are all but inevitable, then it might not hurt to get something from them because something is better than nothing. When your offense is as bad as the Jays is this year, then it’s important to any extra thing out of the lineup that you can–even if it’s 90’ of real estate.

    I did say that it might not hurt for Vernon/Overbay/Glaus/Rios/Thomas to drop the occasional bunt–but that’s simply for unpredictability’s sake. It’s a long term strategy to ensure that the opposition knows it can take nothing for granted. To illustrate: Ted Williams would occasionally intentionally make out. Why? Suppose you’re in a blowout game and there’s a rookie pitching on the mound. Williams would get to two strikes, and deliberately swing and miss for strike three noting the pitch thrown. Now, in a crucial game situation facing the same pitcher and the count gets to two strikes–any idea what pitch is coming? And Williams would be waiting for it.

    It’s a 162 game season and sometimes you’ve got to do things you might not ordinarily do simply to give the opposition one more data point they have to take into account.

    “As I’ve said before in these exchanges, I don’t think you can manage a team based on these kind of splits. And how would you, anyway? To compensate for their poor hitting with two outs, the Jays are hitting .310 with RISP and fewer than two (and a whopping .317 with a runner on third base). Even if you’re reacting to these kind of splits, why would we want to take the bat of their hands when it’s the only time they are hitting?”

    Here I was less than clear here so good catch. I should’ve specified that this applies more to the 7-8-9 hitters than to the run producers. Having said that, I don’t think it would’ve hurt for the big guns to shorten their swings a bit and simply look for contact than the long ball–the whole “swing for singles and the home runs will come.” I saw too many vicious cuts which appeared (I didn’t see many AB in slo-mo) to be longer swings.

    “Either change to a new strategy, or concede defeat.”

    I was using hyperbole there. I never expect a team to concede defeat. I was making the point that when the other team scores X amount of runs your club is now in a situation where the status quo generally hasn’t produced enough runs to get X + 1 run so why stick with the status quo when it has failed you when the opposition has put a league average amount of runs on the scoreboard.

    “This is another thing that station-to-station is often accused of being- conservative and boring, unwilling to take risks. But what the standard run-expectancy argument tells us about bunting is that it isn’t a risk, giving you a small chance of scoring a lot, it’s a strategy that overall gives you bad odds and leaks runs.”

    As with any statistical standard there are outliers and aberrations. The ol’ exception to the rule. The question that has to be asked, and which we’re ultimately going to have to agree-to disagree on is this: When you’re dealing with men on, none out, and three batters who are slightly better than 50% of league average which is the bigger gamble:

    a) bunting the runners over where man on third opens up several avenues for getting him home besides a base hit (PB/WP/balk/sac fly, ground out etc.)

    b) letting players with OBP/SLG of .275/.325 swing away hoping for walk or hit.

    You think (a) is the bigger gamble, I think (b) is. You use run expectancy, I use the stats of the players in question (7-8-9 hitters) in the decision making process.

    Getting back to the RISP/2 out issue, you’re discounting the human factor. There is more pressure on you with RISP/1 out than there is the RISP/2 out because if you fail with one out the guy behind you can pick you up. With two out–you’re the man. No second chances. I’m not stating that clutch hitting is a repeatable skill (just so we’re clear) but I think we both know that choking does occur with a batter. He can become more anxious, he can tighten up affecting his swing etc. I don’t view clutch hitting as being able to hit better in big situations, but rather being able to maintain your normal level of production when there’s more pressure.

    You have to remember that things do get into player’s heads and it affects their game. The Jays know they struggle with RISP/2 out this year. In close games this situation creates additional pressure because you know you haven’t produced well under this particular circumstance and it can become cyclical and a self-fulfilling prophecy. You start squeezing the sawdust out of your bat.

    For the most part, over the course of a career everything will progress/regress to the mean. Just because a player/team struggles with RISP/2 out in 2007 doesn’t mean they did so in 2006 or will in 2008. However, we’re dealing with 2007 here. In 2007 the Jays have struggled all year with RISP/2 out. It has affected them. Will they improve in this regard? Of course, but it’s not likely to happen this year and that was the season/situation we’re dealing with.

    Run expectancy, statistical anomalies (RISP/2 out), regression/progression while true, doesn’t account for the situation at hand for the simple reason that they’re based on large sample sizes that don’t start in April and end in October. That’s why baseball decisions have to be made regarding the situation at hand based on what the team has been doing of late accounting for the personnel on hand at that particular moment. The fact that RISP/2 out and run expectancy will eventually return to the mean doesn’t mean it will do so when you want it to. Things might even out by next May or perhaps it’s simply evening out from a higher level that was achieved earlier but that doesn’t help you at this particular moment.

    Gotta scoot. I hope to have some more stuff later.

    Best Regards

    John

    John Brattain

    September 23, 2007 at 11:49 am

  2. Ok, but league average (which is around what projections say would be required to equal the Yankees) is still 54 runs away. And the league leader in bunting is 26 away. We’re not getting 54 runs with 26 bunts even if they were the best offensive tool of all time.

    That’s a cool story about Williams, but I don’t think it equates to sluggers bunting. One or two bunts a year is not going to get the corners cheating in or anything when Glaus comes to the plate. If he wants to try something that he’s terrible at and maybe leg out an infield single, nobody cares. It’s like the Ortiz shift. Go ahead Dave, bunt! It’s not worth it.

    The shorten swing argument is interesting…I’ve heard they’re pulling too much, swinging for the fences too much. I find it really hard to tell any of that without knowing the hitters or being right there, but that could be part of the problem. I’m not sure though, because the Jays power numbers are down just as much this season.

    “Either change to a new strategy, or concede defeat.”

    I know you don’t mean it literally. But the decision doesn’t work that way. Nobody knows that the other team is going to score X runs until of the game and just because your team hasn’t been scoring lately doesn’t mean that you should (or can) change strategy as soon as the other team puts up league average. More likely is you’re down by 1 in the 7th and the Blue Jays strategy is more likely to swing away because there are innings left. (And thats’ a decision that makes a marginal difference anyway.) The way you broke down the games into only the close ones paints the picture of the Jays just sitting around, knowing the status quo isn’t going to score, but sticking to a failing strategy. Really, a lot of those games they were down the whole game, didn’t have any high-percentage chances to bunt, or weren’t within 1 run late. Late in the game against a tough pitcher with a weak hitter up when obviously 1 run is crucial, the Jays did not avoid the bunt.

    I agree that there may be more situations to bunt with lousy hitters than most managers take into account. I just don’t think that it makes that much of a difference when it’s such a losing idea for average hitters and OBP .275 guys aren’t THAT much worse than average.

    As for the RISP, ok…so it’s true that it’s harder to hit with RISP/2 outs, for everyone, (and for reasons other than the purely psychological). But that doesn’t mean that those situations are any more important to winning. Who cares how a team did in mentally “clutch” situations this year if it doesn’t have any higher impact on scoring runs or winning games? And yeah, they’re generally within a standard deviation and will even out eventually.

    I agree that you can’t always manage by year-long averages, but neither can you hyper-react to short term splits. And sometimes there’s nothing you can do about it. If your team can’t hit with runners in scoring position, or with two outs, how does a strategy that gets you more in scoring position, or closer to two outs, help overcome that?

    halejon

    September 23, 2007 at 5:10 pm

  3. […] railing at real sports journalists instead of all pitch f/x all the time, and I wrote a couple of scathing articles about his bunt fetish. To my incredible surprise and embarrassment, he came onto the site […]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

RSS Recent Posts

Recent Comments

Don’t jump! |… on Hudson + Lester = 4ever
halejon on Blue Jays Blockbuster —…
Happ’s Grasp O… on New Blue Jay J.A. Happ is Over…
The Daily Duce: Tues… on More on the Wind vs. Dickey…
The Daily Duce: Tues… on Does closing the dome help R.A…

Blogroll

RSS bluejays mlb

%d bloggers like this: