I hope that one day the factions will embrace each other and work together to explore the game to new heights of understanding.

Yes we can!

Game strategy debates often remind me of an issue in New York City back in the '80s. Subway crime had gotten so bad that the MTA started to put a uniformed cop on every train. Subway crime then dropped dramatically. Days would go by without an incident on most trains. So some genius at the MTA decided that there was no advantage to having cops on trains, because there was no crime on trains. And around it went ...

John, I think you're beating a strawman here, or maybe you just come across more dogmatic, simple-minded advocates of sabermetrics than I pay attention to.

Getting back to the running game (or for that matter—putting on plays) again, I know what sabermetrics says about the utility of such things, but it doesn’t add up when I compare it to the game on the field. In most sports, it is considered advantageous to make sure the opposing team is ignorant of how you plan on defending against them.

Win expectancy and game theory. Discussed in detail in The Book (Tango, MGL, Dolphin).

For the value of baserunning, basestealing + everything else, Dan Fox's numbers have the best baserunning seasons adding around 20 runs. Seems huge to me.

The Padres (especially the Chris Young/Josh Bard combo) are conducting an experiment for our amusement on letting runners steal at will.

I see that Creation of the Humanoids is out on DVD. Nobody ever tells me anything ...

Ted Turocy has done some work on game strategy which goes beyond the hardline position that John suggests many analysts hold; see this paper and this one.

-- MWE

Ted Turocy has done some work on game strategy which goes beyond the hardline position that John suggests many analysts hold; see this paper and this one.

Heh ... more like the "many analysts I have come across."

A small disclaimer--I'm dealing with what I have personally come across in my travels. I feel fairly confident that not all are like that, it's just that I have come across comparatively few moderates on either side of the fence.

Many thank yous for the links. Both interesting and promising.

Best Regards

John

Bill James' +/- baserunning stats are among the more interesting sabermetric stats devised recently IMO. I think it indicates that statistically oriented analysts are not oblivious to the benefits of base running.

What the hell is the point of any of these stats versus scouts arguments? Do they actual add to the body of understanding, or are they written so the writter can get words to paper?

What the hell is the point of any of these stats versus scouts arguments? Do they actual add to the body of understanding, or are they written so the writter can get words to paper?

That and they're fun to write. Besides, I always get tremendous feedback and new resources when I do them (check Mike's links).

Also, read the very first words in the column...

Best Regards

John

Do they actual add to the body of understanding, or are they written so the writter can get words to paper?

My guess is the hope that if both sides could understand why they so vociferously disagree about certain things in baseball and what the major things are which over there are disagreements, then both sides would be more willing to consider joint endevaors of analysis that would be better than what both sides can accomplish in isolation.

My guess is the hope that if both sides could understand why they so vociferously disagree about certain things in baseball and what the major things are which over there are disagreements, then both sides would be more willing to consider joint endevaors of analysis that would be better than what both sides can accomplish in isolation.

People who are firmly entrenched in dogma, do not change their opinions based on words. Thats what these knuckleheads are... those who dont' see the value that both sides add aren't going to see it. It's just a big circle jerk.

By the way, there was this recent scouts vs. stats roundtable that you guys might be interested in reading. Not sure if it was linked here...I'll see if I can find it...

I would note that the reason that most sabermetric-minded fans frown on the stolen base is not because getting a free base is bad, but precisely because the base isn't free. The way the game is currently played, stealing is a significant risk. If a team was going to adopt a strategy that pushed an average steal attempt into the 95% success range, then obviously stealing would be an optimal strategy. Thus, it behooves teams to defend against stolen bases and thus it behooves teams not to attempt too many stolen bases (or rather not to attempt high-risk, low-reward steals).

All this leads naturally to the point that other posters have made: game theory is of essential importance. For example, if a team did adopt a strategy that pushed stolen base attempts to 95%, other managers and players should steal often, but should not continually exploit this as it may cause the opposing manager to change his suboptimal strategy. They should exploit it exactly as much as they can without causing the strategy to change to a more optimal strategy.

The way the game is currently played, it may be the case (I think not, but perhaps) that teams are too worried about the stolen base and in their attempts to defend it have employed a suboptimal strategy. If this is the case, then teams would be well not to steal too often because they will likely not break even, but should still steal sometimes in order to encourage their opponent to continue employing his suboptimally over-aggressive defense of stolen bases.

In other words, game theory is awesome and is the real challenge for an MLB manager.

I posted this on Ballhype and I'll run it here since I evidently didn't explain myself well in the article:

Seriously though, while some of what I wrote was to get a reaction I think the central point of what I wrote is still being missed. The questions that I wish explored are these: How much value is there in the distraction effect of things like potential stolen bases and potential hit-and-runs and sac bunts.

We know the statistical value of things like the stolen base, SB/CS ratios etc. but is there value above that?

I’m curious about the human element. We know which pitchers have the best fastballs/curveballs/sliders etc. What pitchers are easily distracted and what distracts them? A possible hit-and-run, a burner on first, a potential bunt? Which ones are not bothered by such things? What team's infielders have the best anticipation/reaction when defending plays that are put on? Statistically, what are the most common sources of distraction? What are the most valuable sources for distraction? How can a team employ the most effective and avoid using things that are of minimal distraction.

For example, were I playing a team like the Marlins with the defensively challenged Hanley Ramirez/Dan Uggla or the 2006 Blue Jays that opened the season with a keystone of the inexperienced Russ Adams/Aaron Hill I'd be doing everything within reason to challenge their poise, their reactions, their lack of aptitude etc.

For another example, suppose a team has an older pitcher (37 years old or higher) on the mound and the man playing third is struggling with nagging leg or back problems--would attempting to bunt on the left side of the infield with men on be a better option than swinging away with a league average batter up?

Those are the questions I hope to have answered--the indirect effects that impact run scoring/game situations. We know the raw run value of the stolen base. Now over the course of the season what is the run value of defensive uncertainty/distracted pitching/inexperienced infielders?

Say over the course of a year the SB/CS ratio cost a team five runs. How many runs did the team gain by distracting the pitcher with a man on first? How many more fastballs were thrown because of them? How many more hittable pitches did the baserunner give the guy in the batter's box? How did it affect pitch sequence as opposed to when first base is vacant? How many times did the throwing arm of a catcher cause a pitcher to change his mind as what to throw? Are pitch counts and stamina impacted by throws over to first?

We read about unearned runs--in tennis there is a stat that differentiates between errors and unforced errors; well, how many unearned runs were due to forced errors caused by defensive uncertainty and distraction?

Again--this is the sort of thing about which I’ve wondered. I appreciate the information I’m getting but it isn’t the question I am asking.

Best Regards

John

Part of the reason I think the debate gets so heated is that numbers can't describe everything:

1. As John posts in #16, context can mean alot. And some of that context can't be measured (as others have said, Game Theory also comes into play).
2. There are things we seem to be measuring poorly, like defense.
3. One of the things we keep talking about is sample size. I think it was Bill James who recently said that we sometimes confuse "doesn't exist outside of a small sample" with "the noise in a small sample may be covering up what's really there". Thus, clutch hitters may exist, but that existance can't be culled from the numbers because of the noise.
4. As they say in the stock market, "Past results do not guarantee future returns." Players are a year older, for good or bad. Every player ages differently. Opponents learn from a player's past performance/tendencies.

Now the vast majority of "stats" people accept these things. But on one side of the debate, I see people who ignore these things, and seem to think the numbers can (or eventually will) explain everything; on the other, I see people who think all "stats" people think this way.

Say over the course of a year the SB/CS ratio cost a team five runs. How many runs did the team gain by distracting the pitcher with a man on first? How many more fastballs were thrown because of them? How many more hittable pitches did the baserunner give the guy in the batter's box? How did it affect pitch sequence as opposed to when first base is vacant? How many times did the throwing arm of a catcher cause a pitcher to change his mind as what to throw? Are pitch counts and stamina impacted by throws over to first?

Good questions.

You'd also want to know how many hittable pitches the batter let go because he's giving the man on first a chance to steal. Does a runner moving on a pitch cause a distraction that negatively affects a hitter ? (I think Frank Thomas didn't want Tim Raines running when he was up for this reason) Does a hitter hurt himself by swinging at bad pitches on a hit & run?

You could have a positive SB/CS ratio and still be losing runs if the negatives outweigh the positives. As to whether the positive externalies outweigh the negative, it probably depends on the batter/pitcher matchup and looking at the league totals is not going to adequately answer the question.

It's not something sabermetrics can't answer, but it sure would be a ton of work.

Well getting back to #16. One of the first really interesting things to come up with play by play analysis available was that while the pitcher may well have been distracted by the threat of the running game, the impact was far greater on the hitter.

For the gory details see: Doug Drinen's study of base stealing and batting

Bottom line: Big drop in batting average and power. Partially compensated by a slight increase in OBP.

Seems that the effect currently isn't as dramatic as it was in the 80s, but there's also less running going on. Further, patient hitters as a group seemed to be far less affected (suggesting that the issue isn't so much distraction as hitters potentially ending up behind in the count. Some guys are better at hitting behind in the count than others)