I shared an A’s game a couple of years ago with Markos Moulitsas, and he asked me if I ever participated in the Daily Kos discussions, and I said no, and he asked me why, and I don’t quite remember what I said, I think I made up some lame excuse about focusing on baseball blogging. The truth was that I didn’t really feel like my political views had a solid philosophy behind them that I really believed in, so arguing about political details felt like a pointless waste of time to me, like arguing about wallpaper patterns before you have any sort of blueprint to your house, but I was afraid that if I tried to explain this to Markos that it would come out wrong (your blog is a pointless waste of time!) so I left the truth unsaid.
Lately, though, I find more and more that I am starting to have a general philosophy of things, and that I am getting closer and closer to being able to articulate my beliefs. I feel like I am circling around the same themes, firing bullets at some central target which I keep getting closer and closer to hitting.
And as I get closer to having my own philosophical legs to stand on, I feel like I am now more ready and willing to argue the wallpaper patterns, so to speak.
Here’s another bullet fired around that target. Yesterday, Steven Goldman of Baseball Prospectus made a very political argument in discussing Michael Barrett’s suckerpunch of A.J. Pierzynski. An excerpt from Goldman, with a quote at the end from Richard Hofstadter’s Anti-Intellectualism in American Life:
It’s not that Americans lack the skills for a good rhetorical bout, but that the art of negotiation is something that the culture doesn’t prize as highly as the sudden stroke, the force majeure. We like to hit people, or at the very least fantasize that hitting someone cuts a problem to the quick in a way that talking can’t do. Americans rejected the League of Nations and to this day many of them hate the United Nations. Membership in diplomatic organizations restricts our ability to unload a good haymaker when that irresistible urge arises. There is a streak of primitivism in American culture, “a persistent preference for the ‘wisdom’ of intuition, which is deemed to be natural or God-given, over rationality, which is cultivated and artificial.”
This paragraph is, if I may read between the lines a bit, criticizing three groups of people:
1. Michael Barrett
2. Iraq war supporters
3. Anti-statheads
It implies, by carefully selecting words such as “fantasize” and “primitivism” and by placing quotation marks around the word ‘wisdom’, that intuition is inferior, and that rationality should be the preferred, superior choice whenever possible.
And that’s where I’d choose a different wallpaper.
Let me start by choosing a few selected words of my own. First, I’d like to kill the word “intuition”. It has a negative connotation that puts it at a disadvantage in any argument against rationality. A decision made by intuition runs through a pattern recognition algorithm in our brains. So let’s replace “intuition” with the phrase “pattern recognition”.
By choosing the words “pattern recognition”, we can also get rid of the word “primitivism”. Because the pattern recognition algorithms in our brains are anything but primitive; they are extremely complex. We can easily program a computer to follow a rational algorithm, but nobody has even remotely figured out how get a computer to match a human brain’s pattern recognition ability. Rationality is far, far simpler (dare I say, primitive?) than pattern recognition.
Goldman then goes on to quote the BP book Mind Game, where they conclude via statistical analysis of teams pre- and post-fights, that baseball fights do not benefit the fighters. To which I say, of course they don’t.
Fights begin out of anger, and anger is an emotion that has evolved over millions of years. What evolutionary purpose does anger serve? To make a creature willing to overcome his self-preservation instincts, and risk physical harm to itself in order to communicate to another creature that it is behaving inappropriately. Anger is supposed to be costly.
Ever had a bird attack you when you get too close to its nest? You’re 20 times bigger than the bird, and you could probably kill it with one blow. But it doesn’t care; it’s angry at you. And it works, too. You’re not hanging around that nest to get pecked at; you’re gonna skedaddle away. Anger is a complex, effective interspecies communications tool, evolved over hundreds of thousands of generations of animals.
Rationality, on the other hand, is a brand new tool in the evolutionary chain. Only humans have it. It hasn’t been tested by hundreds of thousands of species over hundreds of millions of years. It’s been tested by one species over maybe a hundred thousand years.
Being skeptical of rational choices is the rational thing to do. I believe that our pattern recognition algorithms are often so much more sophisticated than our rational algorithms, that when they disagree, the rational argument is wrong more often than not. The rational argument is always missing something: some assumption, some variable, some pattern that the sophisticated pattern recognition algorithms don’t miss. Over time, after further analysis, and years and years of study, when all the variables are finally in, the rational analysis often ends up at or near the same place the pattern recognition algorithm started out with in the first place.
Now, don’t mistake me. I’m not saying pattern recognition is always better than rationality. Humans have both, and there’s a reason we have evolved both. Rationality has given us a huge advantage over other animals. There’s probably a time and a place where communicating with A.J. Pierzynski with a fist would be more effective than using a more rational communications tool, but Michael Barrett probably didn’t pick the right one. Given the context, Barrett’s anger didn’t seem appropriate or justified. What I am arguing is that we should not simply dismiss our intuition and emotions as primitive and inferior out of hand.
If you ask me, this is the reason the Afghanistan war has been (viewed as) more successful than the Iraq war. Americans were angry at the Afghan government after 9/11. Anger makes you willing to risk personal suffering. Iraq, on the other hand, was invaded based more on rational arguments than anger. WMDs, therefore blah blah blah. Americans weren’t really all that angry at Iraq. Which had two effects: (1) the decision was more likely to be flawed, because the rational mechanisms for making the decision to invade Iraq were less sophisticated than the complex, emotional mechanisms used to decide to invade Afghanistan, and (2) the lack of anger made Americans less willing to endure the physical suffering that the war would entail, making success even that much less likely.
To make a long point short: to maximize your odds of success, make sure your logic and your intuitions/emotions are in full agreement before making a decision.
* * *
All of which brings me around to the reason I started writing this blog entry in the first place, which was that I was angry with Ken Macha about today’s loss to the Rangers. The grand slam to Rod Barajas when the A’s had a 7-0 lead was infuriating. I can’t communicate my anger with Macha by throwing a good haymaker at him, so instead, at the risk of being ridiculed in public with my arguments, I am issuing this longwinded complaint instead. My anger must out!
The A’s are infamous, thanks to Moneyball, for being rational about their decision-making. Take the emotions out of it, Billy Beane likes to say. To which I say, that’s just wrong.
Sometimes Ken Macha drives me nuts, and sometimes it’s because I think he’s making an irrational decision, but I think the ones that drive me the most nuts are the ones that seem too rational. It’s like Macha won’t trust his pattern recognition tools at all, and requires rational, empirical proof that X is Y before he’ll act on it.
This manifests itself in the worst way when Macha is trying to decide whether to yank a pitcher or not. He seems unable to trust his eyes that a pitcher has run out of gas. He has some logical algorithm: if the pitcher:
(1) hasn’t maxed out his pitch count, and
(2) hasn’t yielded over five runs yet, and
(3a) hasn’t gone five innings yet, or
(3b) has gone five innings and still hasn’t given up a run this inning,
then
(4) leave him in the game.
Meanwhile, anybody with eyes can see that Brad Halsey has completely run out of gas. He loads the bases, but since no one has scored yet, there is logically, I suppose, insufficient evidence that Halsey is done. Whatever. Halsey serves up the grand slam to Barajas. Suddenly, a game the A’s should win by a blowout becomes a huge Texas comeback. Thank you, Ken “One Batter Too Late” Macha!
The human brain is constructed with the emotional center in charge of decisions, not the rational system. That is exactly as it should be. Let the rational inform your decisions, of course, but in the end, trust your pattern recognition system.
Nature has evolved over millions of years a persistent preference for the wisdom of intuition. This wisdom needs no quotation marks.
1. Ken, I just think you're a great writer.
2. Barrett's pattern recognition system told him to take a swing at Pierzyski. I'm guessing that Jim Bowden's pattern recognition system told him that spending money on Christian Guzman was a good investment. In the past, the pattern recognition system of a lot of folks told them that homosexuality was perverse, that black were inferior, and women should just have babies. You're right about the need to be skeptical about our rational system. But I think we should be equally skeptical about our pattern recognition system. I guess it's got an evolutionary head start of millions of years but I'm still not sure we should trust it more than our rational system. Especially, when it comes to matters of morality. Life is all about making choices, though. Choose we must, whether we depend more on our rational system or pattern recognition system. It's an existential dilemma.
3. Great post Ken (absolutely serious). I think I agree with you 95% (well, about macha 105%, which I guess will even out to a 100% in the end). Socially, in such a complex society, I don't think pattern recognition skills should outwiegh rationality almost ever. "Im gonna kick the crap out of this 6'6" 220 pound guy who hit on my wife while I was in the bathroom", "man, I can see the answers on that guys LSAT answer sheet" and "Man, my girlfriend says I don't love her if I go out of state for college" are good examples of when your intuition will be the strongest, the emotions will nearly consume you, but falling to them will be almost assuredly the wrong decision. Those situations, which face humans every day, were not tested by millions of years of evolution.
However, baseball is far more simple. Rationally, if a guy walks 3 straight batters, you should go up there thinking "take til he gives me a strike" until he grooves that 80mph fastball that your pattern recognition skills recognize as "meatball". Umm, yeah, swing. You are spot on about trusting your instincts, cause like they say, whatver has happened on a baseball field has happened a thousand times before. Pattern recognition at its most useful.
And yeah, rationality is not what I would call more primitive than intuition. Properly used, in USES intuition ALONG with all other information available to make the most sound decision. It is the Dual G4 processor compared to the celeron of intuition. Now, just like dual G4 processors, not everyone has that capability. But teh best use both. Beer and tacos baby.
4. 2 Are you sure it was the pattern recognition system that told them these things, and not a faulty logical argument? Eugenics, for example, is a rational idea based on a very faulty premise.
5. 3 Like I said, my point is that your instincts and your logic should be in agreement. If they're not, you should train yourself to recognize that one or the other is mistaken. My point is, however, don't always assume that the mistaken one is the emotional one (or for that matter, the rational one). They are both great tools. Use 'em wisely.
6. Excellent point...
7. I try not to get too angry. I find out in a real heated argument with someone my hands shake a lot. I look awful. I take it as a sign that something other than my logical mind is running my head because my body can't take it.
8. 4 I'm sure if your ask folks why they believe the things they believe they may reply a with faulty logical argument. Much of what we believe has to do with where we grew up, how we grew up, and our DNA. I guess it may be more accurate to say we are inculcated with certain beliefs and then we develop or are supplied with "logical" arguments to support the beliefs. But a rational system might be the most effective tool to show us where our "logical" arguments are faulty. Our pattern recognition system doesn't seem to have that same capability (or does it)? I agree they're both very useful tools and that we should not discount the pattern recognition system as many are want to do, but I think my pattern recognition system is telling me that we need to be clearer about what the rational system can do and what the pattern recognition system can do.
9. Here's another shot at the target: emotions are communications tools. If the rational conflicts with the emotional, then there's a strong likelihood that the rational is missing some data, and the emotions are trying to communicate something. You need to listen to them, and figure out what they're saying.
However, emotions are not precision instruments. This is especially noticeable when you're a parent. Kids will have all sorts of emotional outbursts, and often what they're communicating to you isn't exactly what they're saying on the surface. They may be crying about something happening with a toy, but the real message that the emotion is conveying is that the kid is tired or hungry or bored.
10. 8 Right. And that's why I say I'm circling the target, but I haven't hit it yet. I haven't figured out how to explain clearly what each system can do. When to use one system, and when to use the other, for maximum effectiveness.
And then, once we hit the target, figuring out how to design systems (political, economic, business, technological) based on that knowledge, to take advantage of that knowledge.
11. At the end you say rational should inform, but in the end, trust your pattern recognition. I guess I feel such a blanket statement isn't exactly true. When you bring up Ken Macha, it seems you infer that he is trapped in some kind of algorithm that doesn't allow his intuition to override. That isn't rationality either, thats what computers do. Rationality is "Okay, my pitcher is throwing a shutout, but right now, he has loaded the bases, and from my experience in baseball, he looks like he is gassed. I was caught off gaurd by this, and I don't have anyone warming up in the bullpen. At bat is one of the weakest hitters in the league, but a big league ballplayer nonetheless who is capable of hitting a bad pitch over the fence. The worst that could happen is he gives up a grand slam, I pull him, and we still have a 3 run lead, with our bullpen against theirs, and I think ours is better. The best thing that could happen is Halsey gets out of this, gains confidence...blah blah blah.What should I do?" That is rationality. "Danger, Danger, 100 pitches imminent, remove pitcher now" is not rationality. Trusting intuition even if the rationality doesnt line up is what Dusty Baker does. Ill agree that the best decisions are the ones where both line up. And your last comment says you believe that too. Your last line of your post doesn't.
And FYI, Sexual promiscuity is the tried and true method of evolution that is most benifical to species over the past billion years. Humans are one of the first in the history to be monogomous, and not even for their entire history. Today, in our society, when you are a bar and a hot chick walks by, and she acts in a way that your pattern recognition system recognizes as "She wants me" and your rationality says "Dude, my wife wile divorce me" I think the choice is obvious which one you should defer too, regardless of how long each mechanism has been around. In human society, deference should almost always lie with rationality.
But regardless, I still think this was a great post. If you mean more what you said in 5 than the last line of your post, my agreement rises to 99.99%, if not, well, I still think this was a great piece that is virtually spot on. Great job Ken.
12. 11 Well, I guess that's another reason I haven't hit that target yet. I need clear definitions of words like "rationality". If you think it means one thing, and I think it means another, then we're just going to talk past each other.
13. Interesting post. Starting with the situation in yesterday's game (this is an A's blog, after all), it's not even clear to me (as 11 is gettting at) what counts as "rational" and what counts as "intuitive" in this sort of situation. I was listening to the game on my way home from work, so didn't see how much Halsey was laboring, but I thought he should be taken out for, and for exactly the sort of logical reasons to which you ascribe the decision to leave him in: he's a back of the rotation guy, over 100 pitches, struggling to get outs, and the next batter was a right-hander with some power. Is this pure logic (if ((pitcher.pitch_count > 100) && (bases.areLoaded) && (pitcher.handedness != batter.handedness)) then getHimTheHellOutOfThere) or intution ("This is the kind of situation where I've seen things go very, very wrong before.")?
On the more general point: it seems to me that people for the most part have very poor intuition for statistics. Even well-educated people. In particular, there is a tendency to over-use our pattern-matching facility, and find order in things that are actually random. From the evolutionary point of view that you talk about this makes sense - failing to learn from experience can be deadly ("Well, my brother was eaten by a bear, but the sample size there is only one ... he's probably harmless"), while making spurious matches, resulting in superstition, is usually harmless.
When we carry this intuition over into areas which are quantifiable, it's easy to see it fail. Just look at all the people in Las Vegas, certain that they can identify a hot pair of dice, or a cold deck of cards. Baseball is more complicated than craps, of course. We can never know that a particular hitter's hot streak is as meaningless as that of a pair of dice; the best we can do is fail to find any predictive value in aggregate data. The stat-heads have been known to overstate their case at times, and it would be a mistake to reject all intuition out of hand. But it would be a bigger mistake to always let that intuition have the final word, even when rationality points decisively in the other direction.
14. 13 Well, that's kinda what I'm getting at. You can make a logical argument with algorithm X for leaving him in, and you can make a logical argument with algorithm Y for taking him out, and you can run some regressions and make a case that in this situation, you should choose algorithm Y.
The thing is, Macha should have been able to see with his own two eyes that Halsey was starting to miss on his location in exactly the sort of way that pitchers start to look when they get tired and start to miss location. He should have recognized the pattern, recognized that his default logical algorithm was now based on faulty assumptions, and brought in Steve Karsay right then and there. Karsay was ready in the pen.
In baseball and craps and economics and some other areas of life, yes, you defer to statistics/logic more often than intuition, because those are contexts were stats/logic work often better than intuition. Most situations, however, are not nearly as quantifyably simple. Even this one. The variables change from moment to moment: even if you know the equations, you have to adjust on the fly to the new context (Halsey's arm angle is now dropping), and that requires good intuition.
15. I think at the root of all logic is emotion. The logical, unemotional decisions made in life are to fulfill a very emotional need to achieve and be admired. The calm, collected decisions made in a baseball game are made in order to achieve the ultimate goal of uncorking a bottle of champagne on your teammate's head, in celebration of a World Series victory.
The ideal situation is where you get short gratification while working towards your long-term one. For Barrett, the need to satisfy that short-term gratification was just too great to resist.
16. 15 I think an evolutionary biologist would argue that the root of all logic is not emotion, but sex. The ultimate goal is to survive and reproduce. Everything else is just a tool towards that aim.
17. Haha, touche Ken, I think the evolultionary biologists would agree.
And I think your right, same ideas but using same words with different meanings.
I guess my overall point was intuition is a subset of rationality. A data point in the graph if you will. Sometimes that point will have almost overarching weight, sometimes it can be dismissed. Macha last night should have given his intuition (if he even posseses the capability to be inuitive) much heavier weight.
18. All due respect, but there was never anything rational about declaring war on Iraq; it was purely irrational, and based on fear -- mostly fear of being seen as unpatriotic.
19. Hmm, I guess I'm kind of stumped by the suggestion that intuition/pattern recognition is tied to emotion. That's... broken.
Furthermore, I think if you examine the history of progress of civilization, you'll find that the rational decision making process has a much better batting average than the intuitive decision making process (as for the emotional decision making process? its batting average is a good argument for a DH).
That said, the only guy in Boston who didn't know that Pedro was toast was the only one that really mattered - so I'm completely sympathetic to the emotional effect this has.
20. 18 Some pretty smart people, even some on the left like Thomas Freidman and Kevin Pollack, gave some compelling rational and moral reasons for the Iraq war. They may not have been right, but I don't think they were irrational or impulsive.
21. 18 I think we're stuck again on the fact that I haven't defined "rational" very well.
19 Sorry. The ties between intuition and emotion is something I've laid out in previous articles, but omitted here. The two are strongly intertwined in the mechanisms of the brain.
As for your second paragraph, I'm not buying it. Marxism and Communism and various forms of Fascism are all the result of a rational decision making processes. The (flawed) logic of these systems and their leaders have been used to justify some of the worst horrors in the history of mankind.
We have a much better batting average when direct human reasoning is removed from the equation, and automatic processes are put in its place: free markets and democracy are both more efficient and more humane. The collective intuitive decisions of the masses often work better than the best logical reasoning of the smartest people in the land.
22. I think I know what you are getting at, at least as far as baseball is concerned. One caveat I would add though is that patience is a virtue. I suspect patience is a rational component. Without patience a manager would be reduced to a mad scientest, flicking levers without thought to the bigger picture of the overall 162 game season. There have been managers like that and they usually don't last too long or have much more than fleeting success.
About two weeks ago I had much the same premonition that you had in watching Halsey pitch to the Rangers. Bobby Livingston was pitching for the Mariners and Vlad was at the plate. I have seen enough baseball to know that Livingston had no chance of getting Vlad out. At least Livingston held him to a single-- a line drive that would have killed any infielder foolish enough to get in front of it.
Stan from Tacoma
23. Sorry Ken, I think saying that Fascism is a result of the rational decision making process is pretty off base. Maybe the LEADER made a rational decision to take advantage of certain situations in an inhumane way, but it was the masses inability to overcome their emotions, intuitions, or pattern recognition skills that allowed those deeply flawed ideas to take hold. Emotions is what allows those very smart people in the bottom .1% of morality to to get the sheep chanting "4 legs good, two legs bad" and then to switch to "4 legs good, two legs better" without ever knowing the difference. If the Germans of 1940 were rational in their decisions at the polls, and not emotional about their incredible depressed state of their economy and status in the world, some nut spouting "Its all the jews fault" would have been sent to an asylum, not elected supreme chancellor.
Rationality, and not intuition, or pattern recognition skills, is what got the first humans to band together to grow agriculture, to find out that the earth revolves around the sun, and to find out that bunting with your #3 hitter with no out and a man on second is a bad idea. Pattern recognition skills are base animal instincts that allow the greatest atrocities on the planet, such as the sacrifice bunt.
And being a Poli Sci major, from what I remember Marxism and Communism are not flawed systems any more than democracy (they reflect what certain people feel their government should provide), though their implimentation in the USSR and various countries was. The US had to go through the Articles of COnfederation before they got the Constitution, and even with all the written records of how they came to get it, you still need 7 of the smartest, well trained (at least supposedly) people in the US to interpret it. Democracy in its purest form discriminates against minorities, and in a large one hugely favors the rich over the qualified. Now Im not communist, and I think democracy is the best form of government, but that's because I believe in keeping everything you earn economically and limited government oversight, and not total equality in economy and total government control. Thats an opinion though, not a fact.
I think you are trying to argue against robots in the dugout. I couldn't agree more. We need a human up there that knows when to override the system, but we have to remember that the system is there in the first place because its right most of the time.
24. Perhaps you're right, Ruben, I don't know. Obviously, I'm still not hitting the target, because y'all keep pointing out the flaws in my arguments. I appreciate that.
I still think the problem here is that I'm not defining things properly, or I'm combining concepts that should be separate, or I'm not defining the scope of what I'm aiming at.
I can sense where the target is, but I just haven't been able to find it. I'll keep shooting, and one of these days, I'll find it.
25. Yeah, I would definately agree. It seems these comments have been more about semantics, like if we got together, we would be agreeing after 5 minutes of "yeah, yeah, thats what I meant".
No matter, still was a great post, keep up the great work Ken. The great thing is, we all kind of know exactly where you are shooting at as well, hitting the yellow and blue, just waiting for the bullseye.