I would consider this an anti-rationality quote because he’s refusing to actually Shut Up and Multiply. If a guy can beat you, you give him a free pass?
And on this you are simply mistaken. Many people refuse to shut up and multiply. They are unable to admit that they do have limits and that sometimes losing is the best thing to do.
It is quite easy for those doing math on the situation to see that intentional walks that aren’t considered automatic are usually a massive error.
Apart from not being math, your “math” is just wrong. You ought to be able to see why if you read the surrounding conversation here. You can potentially save multiple runs if you correctly evaluate your ability with regard to one particular conflict and concede.
This isn’t “anti-rationality”. It is anti conventional wisdom and common enforced exhortation.
I apologize, for I spoke far too strongly. I should not have made such a claim. Nor do I on reflection think this is the place to start actually doing said math, and your belief being so strong made me call up a friend to go over the problem. I would however strongly dispute that this thread makes it clear that this should run the other way, and continue to believe with p~.9 that there are in fact too many intentional walks (I would have said p~.99 before this thread, my friend after consideration said .95).
This thread seems to be saying as evidence for there being too few: There exists a cognitive bias that, all things being equal, will cause managers to be reluctant to walk a batter intentionally.
I agree that this bias exists. However, I think there’s a directly opposite bias that says “don’t lose to their best guy” regardless of whether it’s right to do so and a bias much stronger than either that says “do the thing that won’t get me hammered in the press if I lose.” I would guess there’s also Overconfidence Bias at work here: Managers think that better matchups are more distinct from worse matchups than they actually are. There are a lot of biases surrounding this decision, they run in both directions and only a small (if growing) number of teams are willing to sit down, do math and try and figure out the right answer.
The only way to actually know which way this runs is to observe what managers actually do and compare that to a well developed model that evaluates the chances of each team winning given each potential decision.
My observations over the years is that these are the categories of situations involving possible intentional walks:
1) Conventional wisdom automatic walks. You need to walk him, and you do.
2) Conventional wisdom “free” walks. CV says that the run doesn’t matter so put the guy on. Given the option value of being able to walk a guy, I think managers use this far too often; they essentially use it as long as the next guy is worse.
3) Getting to the pitcher. This is done at least as much as is reasonable given the lineup effects of doing this.
4) Walking the obviously more dangerous guy because there are men on base. This is the situation where it is possible they walk too rarely; I am willing to accept that some managers do this too rarely. Some clearly do it too often.
5) Walking the dangerous guy because you flat out won’t pitch to him. This requires such a strong hitter to be right due to the value of an out. I’d have a very hard time believing this is substantially underused.
6) Walking one guy to get to another somewhat similar guy. This is done way too often, many times in places that boggles the mind.
I suspect that what happened is that the quote comes from a time when the conventional wisdom was different and managers did in fact walk batters too rarely, especially due to issues of sportsmanlike conduct, but such considerations seem to be almost entirely gone.
So what does “can beat you” mean? I still don’t understand this either in the context of baseball or rationality.
If “can beat you” means “could, theoretically, beat you” then you walk everybody.
If “can beat you” means “is p>.5 to beat you” then you never intentionally walk anybody—if a pitcher is so tired he thinks the next batter is p>.5 to get a hit, he should ask to be relieved.
If “can beat you” means “is p>k to beat you” where k is some threshold, then Paige does seem to be saying “walk people more than you do currently”.
And on this you are simply mistaken. Many people refuse to shut up and multiply. They are unable to admit that they do have limits and that sometimes losing is the best thing to do.
Apart from not being math, your “math” is just wrong. You ought to be able to see why if you read the surrounding conversation here. You can potentially save multiple runs if you correctly evaluate your ability with regard to one particular conflict and concede.
This isn’t “anti-rationality”. It is anti conventional wisdom and common enforced exhortation.
I apologize, for I spoke far too strongly. I should not have made such a claim. Nor do I on reflection think this is the place to start actually doing said math, and your belief being so strong made me call up a friend to go over the problem. I would however strongly dispute that this thread makes it clear that this should run the other way, and continue to believe with p~.9 that there are in fact too many intentional walks (I would have said p~.99 before this thread, my friend after consideration said .95).
This thread seems to be saying as evidence for there being too few: There exists a cognitive bias that, all things being equal, will cause managers to be reluctant to walk a batter intentionally.
I agree that this bias exists. However, I think there’s a directly opposite bias that says “don’t lose to their best guy” regardless of whether it’s right to do so and a bias much stronger than either that says “do the thing that won’t get me hammered in the press if I lose.” I would guess there’s also Overconfidence Bias at work here: Managers think that better matchups are more distinct from worse matchups than they actually are. There are a lot of biases surrounding this decision, they run in both directions and only a small (if growing) number of teams are willing to sit down, do math and try and figure out the right answer.
The only way to actually know which way this runs is to observe what managers actually do and compare that to a well developed model that evaluates the chances of each team winning given each potential decision.
My observations over the years is that these are the categories of situations involving possible intentional walks: 1) Conventional wisdom automatic walks. You need to walk him, and you do. 2) Conventional wisdom “free” walks. CV says that the run doesn’t matter so put the guy on. Given the option value of being able to walk a guy, I think managers use this far too often; they essentially use it as long as the next guy is worse. 3) Getting to the pitcher. This is done at least as much as is reasonable given the lineup effects of doing this. 4) Walking the obviously more dangerous guy because there are men on base. This is the situation where it is possible they walk too rarely; I am willing to accept that some managers do this too rarely. Some clearly do it too often. 5) Walking the dangerous guy because you flat out won’t pitch to him. This requires such a strong hitter to be right due to the value of an out. I’d have a very hard time believing this is substantially underused. 6) Walking one guy to get to another somewhat similar guy. This is done way too often, many times in places that boggles the mind.
I suspect that what happened is that the quote comes from a time when the conventional wisdom was different and managers did in fact walk batters too rarely, especially due to issues of sportsmanlike conduct, but such considerations seem to be almost entirely gone.
“If a man can beat you, walk him” does not mean “walk people more than you do currently”.
So what does “can beat you” mean? I still don’t understand this either in the context of baseball or rationality.
If “can beat you” means “could, theoretically, beat you” then you walk everybody. If “can beat you” means “is p>.5 to beat you” then you never intentionally walk anybody—if a pitcher is so tired he thinks the next batter is p>.5 to get a hit, he should ask to be relieved. If “can beat you” means “is p>k to beat you” where k is some threshold, then Paige does seem to be saying “walk people more than you do currently”.
It isn’t about tiredness or general competence.
The other replies here explain the quote well. I must affirm the rational decision making principle that is illustrated.