Baseball Strategy

[mc15]

Runners on 1st & 2nd, 0 out. Top of the 2nd and down down 1-0. It's a warm day with the ball jumping. Daryl George up, with James Beresford on deck. RHP on the mound

[matjaz27]

Interesting, George hitting 4 and Beresford 5..I'd bunt and let Beresford get 2 RBI's

[rhricho]

unless he was hitting $3.5+ then swing, but Bunt, Bunt! - but i thought the example was where George was 9 with beresford leading off! maybe i got that wrong!
Bunt everytime!

[mc15]

I didn't create the scenario... Starting to think it might not have been Beresford, but in fact Bedford. Anyway, we can leave it as it is for now

[mc15]

By ALAN SCHWARZ
Published: April 6, 2009. New York Times

You can learn a lot during a major league baseball game. Like Ukrainian, if it is a particularly slow nine innings.

As for the science of baseball strategy, one game teaches precious little. A well-timed sacrifice bunt can backfire and lose the game; a foolish steal can appear brilliant. The vagaries of randomness — the way Sandy Koufax got battered occasionally and a pipsqueak named Bucky Dent hit one of the most famous home runs ever — camouflage the game’s inner forces, which for 150 years have operated somewhere between fact and fable.

One game has little meaning. A thousand seasons can take a while. Thank goodness for quad-core processors.

“Computer simulations work pretty well in baseball for two reasons,” said Carl Morris, a professor of statistics at Harvard University who has written several papers that commingled baseball and formal statistical theory. “In general, they allow you to study fairly complicated processes that you can’t really get at with pure mathematics. But also, sports are great for simulations — you can play 10,000 seasons overnight.”

No one can afford to wait less than major league teams, which crave every extra run or victory they can wring from their $100 million rosters. John Abbamondi, the assistant general manager for the St. Louis Cardinals, says his team and about 10 others use simulations to evaluate potential trades and how they might affect the pennant race.

“It’s all part of the statistical analysis that complements the more traditional scouting we do,” he said.

Using computer simulations to explore in-game and other baseball strategies is by no means new. As early as 1958, a professor at the Massachusetts Institute of Technology programmed a behemoth I.B.M. 704 mainframe to investigate whether the sacrifice bunt was a smart play. (More on that later.) Simulators have since grown so complex that the most sophisticated one available to the public, called Diamond Mind, not only runs lickety-split on laptops but even considers minutiae like the effects of wind in individual ballparks.

Under what conditions is bunting advantageous? When does trying to steal make sense, and when does it decrease the chances of scoring? Questions like these turn out to be ideally suited to computer programs through which millions of iterations can smooth out the peaks and valleys of randomness, and converge toward a reliable approximation.

Known among formal statisticians as the Monte Carlo method, this approach takes spectacularly complex phenomena like weather patterns and stock performance and allows their behavior to be approximated, if not determined.

What are the chances of winning a game of solitaire? Rather than writing an equation that tries to take into account the trillions of trillions of possible hands and moves, a statistician can run a computer program that simply plays the game a few million times in minutes to see how often it wins. Dr. Morris says he has seen the Monte Carlo method used to improve computer graphics and explore gene sequences.

Like such competitors as Strat-O-Matic — which made its debut in 1961 with at-bats determined by cards and dice, and remains popular on the personal computer — Diamond Mind is designed to allow fans to play fictional games and seasons, exploring what-if scenarios that real life would be too slow and controversial to allow.

Take the age-old question of how much difference a team’s lineup order makes. This issue so vexed the former manager Billy Martin that he once literally picked his Detroit Tigers batting order out of a hat.

Luke Kraemer of Imagine Sports, which owns Diamond Mind, programmed the simulator to force the 2008 Yankees to bat their best hitter and cleanup man, Alex Rodriguez, ninth — to see how scoring was affected. Mr. Kraemer got the run total not for just one season, which can fluctuate as much as 80 runs in each direction from simple randomness, but for 100 seasons — more than 16,000 Yankees games in all.

The result? The Yankees scored 747 runs per season, 40 fewer than their real-life 787. (Diamond Mind was so accurate that 100 seasons with A-Rod batting fourth averaged 789, almost dead-on.) Most research suggests that those 40 runs would mean only about four fewer victories, for a strategy no manager would ever consider; so the difference with Rodriguez batting third or fifth would be insignificant, and nowhere near worth the forests of trees that would give their lives to the ensuing sports-page debate.

Diamond Mind took its cuts at several other baseball knucklers, running 100 full seasons of games for each:

•

The intentional walk. This frequently used defensive strategy avoids dangerous hitters and can set up a double play, but it also awards a free base, and even the best hitters usually make an out. So is it smart in the long run? Diamond Mind found that it was not, though the difference was only about five runs per team per season.

•

The stolen base. Advancing from first to second puts the runner in scoring position, but he — and the rest of your hitters — will have a hard time scoring if he gets thrown out. Mr. Kraemer looked at a recent team that ran wild (the 2008 Tampa Bay Rays) and one that barely stole at all (the 2005 Oakland A’s) and switched their mind-sets to see what happened. The A’s scored 20 runs fewer, which probably says more about their players’ inability to run in the first place. But when the speedy Rays stole sparingly, they increased their scoring by 47 runs per season — suggesting that perhaps the Rays were running too often in real life.

•

The sacrifice bunt. Is it worth making an out intentionally to move a runner from first to second? Forcing a team that hated that maneuver (the 2005 Boston Red Sox) to do it a lot cost them 19 runs per season. But making a bunting team (the 2008 New York Mets) avoid it also cost them — by 15 runs on average — suggesting that the Mets’ managers, Willie Randolph and Jerry Manuel, used it quite intelligently. (The 1958 M.I.T. statisticians found that the sacrifice was rarely a good move; major league managers paid little attention.)

One problem with computer simulations is that no matter how realistically they might be programmed, they can say more about the programmer than baseball itself. A computer, after all, cannot feel human emotions like pressure or the will to hit in the clutch.

“We can run the experiment in the simulation environment and think we’re measuring the effect of a great defense on a pitching staff, but it might tell us more about how we modeled defense,” said Tom Tippett, who wrote the original Diamond Mind code in the early 1980s. “The simulation is real close to real-life baseball, but in the end it isn’t real-life baseball.”

After developing Diamond Mind into the industry standard, Mr. Tippett was hired a few years ago by the Boston Red Sox — a sign of how much some teams have come to value simulation research. While none will discuss exactly what they model and how, Mr. Abbamondi, of the Cardinals, said they could provide objective insight into how an offense might be affected by trading for a hitter in midseason; how many games that might improve the team; and how that hitter might improve or deteriorate as he ages. Many of these measurements come in the form of scenarios of increasing uncertainty, not unlike the projection of hurricane paths.

As Mr. Tippett suggested, however, simulations have inherent limits, and probably will not ever model baseball’s vicissitudes of fate — how scrubs morph into all-stars and some teams just collapse. (Indeed, fans of the recent New York Mets would be relieved that some things defy re-creation.) Tony La Russa, the Cardinals’ manager, who is a sure bet for the Hall of Fame, said the value of computer simulations in baseball tended to stop at the dugout entrance.

“There’s way too much importance given to what you can produce from a machine,” he said. “These are human beings, and I don’t think any computer is going to model that close to what we deal with at this level.”

That can be as true now as it was 25 years ago, when a Tank McNamara cartoon captured it best. A downtrodden manager peered over his computer. He asked plaintively, “But will it take the blame?”

[theolderiget]

I love the quote attributed to the recently deceased Earl Weaver - "why don't you take the sacrifice bunt and shove it up someone's a s s and leave it there"

[lucadello]

Jan 24, 2013 15:51:21 GMT 10 theolderiget said:

I love the quote attributed to the recently deceased Earl Weaver - "why don't you take the sacrifice bunt and shove it up someone's a s s and leave it there"

Amen to that..................

[Chilli Wil]

Scenario is too light on info. How did the runners get on? How about some splits for George and whomever might be on deck? How's the guy in the hole with RISP and 2 out? How about the pitcher? Does he fare well with a runner on third less than two out? Does he throw to contact or can he dial up a K when he really needs it?

This is why we pay the guy calling the shots the big bucks!

FWIW if the two guys hit their way on it would be good to see something hyper aggressive like a hit and run in this spot.

[oldmanriver]

Chilli Wil, you are asking for way to much info, having seen George hit, bunt is definately the go, then you can merrily think about the next play, generally George batted 9 and Beresford when he came back batted 1. But, bunt is the best option for a light hitter in that situation.

[oldmanriver]

Another Weaver saying was, give me the three run bomb any day.

[mc15]

There is a great Earl Weaver story where he was asked at a conference what his formula for winning was. He walked over to a blackboard and wrote:
GP+3RH=W
Turned to the room and said "Good pitching plus three run homers equals wins."
I like it!

[oldmanriver]

Interesting reading about the Heat v Aces game on the Vic league website.
The Aces regained the lead in the eighth, when Josh Hendricks led off with a hit, went to second on an errant throw, advanced on a Darryl George sacrifice bunt before scooting home when Beresford laid down a sacrifice bunt with runners at the corners.
The bunt probably wins this monir debate.

[bobby]

OMR this is what a few of have been saying all along, it is fine to bunt in the later part of the game but asking to do it in the 2nd is a different story. The "debate"about about calling to bunt in the 2nd inning not the 8th.

[mc15]

Ha! I did read that Beresford squeezed George in. Not sure of the details...
What was/can someone post the play by play for that innings?

[oldmanriver]

Bobby, there isn't a great deal of difference, one you are bunting to advance the runners to get a better hitter an oppurtunity to convert rather relying a weak hitter who would have had an extremely magnificent chance of hitting into a double, now, wouldn't that make evrybody happy. Just go the bunt and make Daryl George feel useful and contribute to a possible bigger rally. There is nothing in anybodys play book(except Earl Weaver's) that says you can't bunt in the second innings.

[Goblin]

Melbourne Top of the 8th

Pitching Change: Brendan Wise replaces Scott Mitchinson.

Joshua Hendricks singles on a ground ball to shortstop Ryan Khoury. Joshua Hendricks advances to 2nd, on a throwing error by shortstop Ryan Khoury.

Darryl George out on a sacrifice bunt, pitcher Brendan Wise to second baseman Aaron Bonomi. Joshua Hendricks to 3rd.

James Brooks singles on a ground ball to pitcher Brendan Wise.

James Beresford out on a sacrifice bunt, pitcher Brendan Wise to first baseman Allan de San Miguel. Joshua Hendricks scores. James Brooks to 2nd.

Carlo Testa walks.

Brad Harman lines out to third baseman Carter Bell.

[deadfish]

Jan 25, 2013 14:00:50 GMT 10 oldmanriver said:

Bobby, there isn't a great deal of difference, one you are bunting to advance the runners to get a better hitter an oppurtunity to convert rather relying a weak hitter who would have had an extremely magnificent chance of hitting into a double, now, wouldn't that make evrybody happy. Just go the bunt and make Daryl George feel useful and contribute to a possible bigger rally. There is nothing in anybodys play book(except Earl Weaver's) that says you can't bunt in the second innings.

Except there is a difference.
The Perth game was late in a low scoring affair where it looked like one run could be difference. Absolutely I'd bunt.

The scenario under discussion here is early in a game that looks likely to have plenty of runs scored - "the ball is jumping", Canberra have already scored in their only dig, if George (number nine hitter presumably, if Beresford is following) then Melbourne have ALREADY had 5 runners reach base.

The one extra run the bunt may get you probably won't be the difference. The potential big inning and early shower for the Canberra starter (who is obviously struggling) might. No way I'd bunt.

[oldmanriver]

This is what makes baseball interesting, different playing approaches to situations, they way it should be, unpredictable and not regimently out of a play book. Imagination will always make life interesting regardless of what innings it maybe.

[bobby]

very true OMR, but in same peoples eyes the imagination = Bush league play.....

[mc15]

Thanks for posting the play by play Goblin. The only thing I would add that makes the play from last night correct in my mind was that I believe the score was 2-2 at the time. 2-1 down and I would not bunt. With only six outs to play with I would be happy to try and knock the guy in from scoring position.

Leave the bunts to softball

[gm27]

Jan 24, 2013 13:10:23 GMT 10 mc15 said:

Runners on 1st & 2nd, 0 out. Top of the 2nd and down down 1-0. It's a warm day with the ball jumping. Daryl George up, with James Beresford on deck. RHP on the mound

bunt for a base hit

[mc15]

Jan 25, 2013 15:05:39 GMT 10 gm27 said:

Jan 24, 2013 13:10:23 GMT 10 mc15 said:

Runners on 1st & 2nd, 0 out. Top of the 2nd and down down 1-0. It's a warm day with the ball jumping. Daryl George up, with James Beresford on deck. RHP on the mound

bunt for a base hit

Yep I like it!

[oldmanriver]

Imagination. Magnificent.

[Chilli Wil]

Jan 25, 2013 10:57:27 GMT 10 oldmanriver said:

Chilli Wil, you are asking for way to much info

That was kinda the point. All that info should (and hopefully is) available to the manager. Even at club ball you should have a book on most guys going around at the higher levels, so I think for a semi-pro league that isn't much of a stretch.

As aces fans we don't have that info (maybe Stu does he enjoys stats about as much as anyone) so we can second guess and postulate all we like but we will never have as complete a picture as the guy in the bench charged with making the calls.

As I said: HIT AND RUN!!

[theolderiget]

Earl Weaver on the Hit and Run:
I don’t have a hit-and-run sign, and I believe it’s the worst play in baseball… you often give the opposition an out on the hit-and-run play. That’s because you can’t trust the pitcher to throw a strike, so the hitter is often waving weakly at a ball that’s off the plate. That usually results in a weak grounder that gets the runner to second, but the hitter is easily retired at first. Hell, you may as well bunt! Over the course of the season, only a few guys actually get hits on the hit-and-run play, because everything must go right for it to work. About the only thing you can say for the hit-and-run is that it prevents the double-play grounder. But when you add up the caught stealing, the weak grounders, and the line-drive double plays, that advantage vanishes. I’ll take my chances with a normal swing anytime.

[stuartcapel]

Drryl George is hitting .196 with runners on this season.

James Beresford is hitting .200 with runners in scoring position.

[Deleted]

I must admit that I was always partial to the hit and run (or its variance in the run and hit), though it should always depend largely on the game circumstances.

For example, the capacity of the runner to turn this into a steal if no contact was made, the consistency of the pitcher in being around the strike zone, whether the pitcher was left or right-handed and throwing against a left or right-handed hitter.

Whether the player was known for hitting line drives into alleys or, in the case of a left-hander, was someone who consistently pulled the ball hard. I'd place trust in someone like James Beresford to deliver with a reasonable level of success.

A magnificent play when it comes off, and very hard to defend against.

More than that, if a hitter is in a slump or is down on confidence or pitch selection, it is a means of taking any indecision out of his hands. Jeez, I had a coach who used to put the hit and run on me even when there weren't any runners on!

High risk, nevertheless, especially if the defence opts to pitchout.

[theolderiget]

Thank God for Google - based on complete Major League statistics for he years 2003 - 2011 the following was the researchers findings.
"The hit-and-run is far from the worst play in baseball. For a small-ball tactic, it has been quite successful over the past nine seasons, increasing scoring by .06 runs per attempt on average. The value of the hole in the infield defense is real, adding about 27 points to the batting average of the hitter. The double plays avoided by executing the hit-and-run offset the runners caught stealing on the play, and the extra bases gained by the runner when the ball is put in play are enough to move the play into the plus column overall.

However, there are some situations where the hit-and-run attempt made less sense and was a barely positive or even a net negative play—with the fourth and fifth hitters in the lineup up, with one out, or in the popular ball-strike count of 2-1."

[mc15]

Jan 25, 2013 17:50:52 GMT 10 stuartcapel said:

Drryl George is hitting .196 with runners on this season.

James Beresford is hitting .200 with runners in scoring position.

So a 0.4% (1 hit in 250 at bats) difference between the two

[hank]

Aces, 2nd last in HR's and 2nd last in hits, Put your bat away and manufacture!

The hit and run is the worst play in baseball??? My goodness - you are kidding me, I'd love to see you tell Tony Larussa that!

mc you nominating the 0.4% shows your clearly still only thinking about hitting and not the situation!
You keep swinging the bat with a .240 team average and see how many finals series your team plays in.

I agree the 2nd does seem a little early to be pushing bunt especially with back to back hits in the second, but even after saying that, I would still be going from my seasons experience that told me perhaps bunt with my no.9 may work here??

MC your pushing this support thing ridiculously now and that fact that your noticing that ppl are changing posts etc says to me you need to log out for awhile and have a chill out session for 5mins! This is a forum that means discussions occur with difference of opinion, concluding someone's view with "howl at the moon" says little for your ability to handle objections! Good luck with that.
While I'm at it, we are so proud you have such an inside word!! Congratulations, may the rest of us be left in the cold not knowing a thing or having any confidence in the one's that are in charge of the teams WE TRY TO SUPPORT!!