Stop laughing please.

I think you did a great job of explaining it here. I’ve been parroting the book on this for a long time; IMHO, Albert cannot ever become a TRULY great player, a transcendental player like Bonds was in 2001, until he starts walking 150-200 times a year.

And we WANT that. Everyone always underestimates how valuable it is to have one more baserunner on the field. Men on base are worth their weight in gold.

IMO the only time a player should ever be intentionally walked is with one out and the winning run on third, or with 2 outs and the winning run on second or third.Other than extreme situations like that, walking Pujols to face cesar izturis is a losing proposition more often than not.

(it’s also cowardly).

The only criticism I really found was that it didn’t take the inning (outside of the 9th with all others being equal) or the opposing pitcher or that that “league average” hitter your talking about was OPSing .519 over the last month (and struck out 16 times over that span) and the two guys behind him weren’t doing a whole lot better. But, you’re right. Who am I to question The Book? After reading your response, it’s now apparent this wasn’t up for discussion. Btw, didn’t we use to think getting 3 outs in the 9th inning was no different than getting 3 outs in any other inning? Hasn’t that thought started to be questioned a bit? Why is this any different?

Hey, he and I didn’t completely buy in to what you wrote as being a way to handle the situation 100% of the time. Sorry. I’ll try to do better next time and realize my questioning The Book is blasphemy.

And just because I want to read this again.

I don’t understand it. Therefore, it’s wrong and stupid. And no I can’t prove it b/c I don’t want to waste my time trying to read it and understand it.

Man, that makes me feel better about ever disagreeing with anything around here ever again.

aaaaand a ground ball to second. out at second – and out at first.

and could potentially turn into something very good. Best case scenario on DeRo is that he’s an average 3B (i.e. a good upgrade for us) for a couple of months. I’d like DeRosa but I’m broadly against giving up Motte or Perez. They can have Salas, Reifer, Kinney or maybe even Todd if they want a reliever but I think we need to keep Motte and Perez unless a real 2 or 3 win upgrade is on the table.

and is a free agent at the end of the season. It would be a horrendous trade from our perspective.

please stop Marky Mark. please.

Piss a guy off by saying “we’re gonna beat you by making sure you bat in a tough situation.”

if there’s a reason why a guy is in a slump – an injury, the wife is fooling around – then that makes it less likely he’ll hit the ball hard. It isn’t the “slump” that causes something, it’s the underlying cause of the slump that causes the hitter to be less likely to get a hit in his next at bat.

This assumes that the underlying condition is still present. Unfortunately, observers can rarely tell if a slump has a specific cause, or if it’s just chance. This makes identifying the removal of the condition even more tricky.

This is why I get so annoyed at the small sample sized fascists… In a vacuum, 123 at bats is not a sufficiently large sample size to declare that Rick Ankiel can’t function in close and late situations. But if we know he has a hard time under pressure, and he specifically mentions that he “tenses up” in these situations, then there’s more of a basis to start concluding certain things.

is very well explained, in fact — much better than most of the stuff BP puts out, for example. As to whether the samples are independent and identically distributed, I admit I don’t remember all the details of the methodology well enough to answer that question. You should read it, however. It is an excellent read and, in fact, corrected me on some of my own preconceived notions (re: sac bunts, for example) about the game.

i dig the IBB now

says 449 true distance, 437 standard distance.

It’s run expectancy, not win expectancy. But giving up ~.2 expected runs is always a bad idea. It’s more than the difference between albert and skip schumaker, and you’re giving it up for free and then facing the next guy, who is perfectly capable to being a little run generator himself.

there’s a table (p294 in the softcover copy) with the bases occupied states, outs, and wOBA {.400, .465, .530}. Walking pujols in that case increased the numbers of runs expected (RE) by 0.193 runs over what pitching to him would have increased RE by 0.193 runs.

IE, the RE before walking albert, had albert been league average, was 2.052. The fact that it was Albert at the plate, with his ridiculous .455 wOBA instead of a league-average hitter, increased our RE by 0.172, from the standard 2.052 to 2.224. Since RE after walking albert was 2.417, walking albert increased RE by the 0.172 runs that albert is worth over the average player, PLUS an addition 0.193 runs.

The only times that the pitching team breaks even in RE walking a .465 wOBA player, even considering his skill, is with 2 outs and runners on 2B or 2B+3B.

Isn’t the “feel” developed from experiece? And that experience is, in a sense, estimating prior probabilities? A poker player observes the behavior of other players (aggressive vs. passive), and if the hand is called down (bluff vs. real hand). Thus, the poker player is estimating a probability distribution for the other players even though he or she might not only be doing a rough estimate. i.e. the player is not actually recording the results from every hand and then entering that data on SAS or SPSS. I guess I would not want to say that experience and statistics are mutually exclusive. Also, there are other things captured by human experience that is not accounted for in statistical predictings. That’s the reason why we use fudge factors all the time in statistical modeling.

but when you account for those factors, it may lead to a decision different from what was the odds-on one to achieve the desired outcome. The question is whether a particular manager (TLR usually) can process and accurately measure this not-easily-measured information to make the best decision. I don’t think anyone is arguing that after weighing everything (the straight odds plus the hard-to-measure stuff) we want to the manager to take the path least likely to succeed.

See this is what confuses me.

His .230 average….his .309 obp….will beat me more often than not? He’s getting on base 30% of the time and making an out 70% of the time. Am I just completely dense, or does that statement not make any sense at all?

just started a couple days ago.

is pretty fun, though, isn’t it?

To be fairer, I should have calculated the RE for each situation and added it up, but I didn’t feel like taking the time to do that. Someone else can, if they want. i don’t think it’ll change much, given the size of some of the big innings that followed the albert IBB.

you can calculate the RE for albert, just like you can the average player, and that is what Tango did in the book, and that is what Chuckb quoted. My table just tries to illustrate the point in real-world terms.

The magnitude of some of the big innings we had post-IBB makes that pointless in my mind. We’ve been quite lucky after opposing teams walk albert this year- the table above shouldn’t be nearly as stacked in favor of “theory” as it is. Over 11 “bad idea” IBB’s, it should be about a 2.2 run advantage, and it’s clear that the results are much higher than that without doing the math.

Runner on 2nd, 1 out. Walk Pujols then 1 run scores.

Is that really a fail? Do we know for a fact that Pujols wouldn’t have driven in that run himself? Or hit a double, which he would have then "switched places" with that guy on 2nd (still 1 out) and then scored on whatever drove in the 1 run that actually happened equaling 2 runs scored? Or what if he hit a home run?

Yeah, but we know roughly how often AP will do all of those things (against an “average” pitcher) by looking at his career (or, if you prefer, stats over a shorter period). We can then work out how often (in % terms) AP drives in the guy on second (i.e. his BA), how often he drives in the guy on second and scores more runs himself (i.e. a HR, or, to a lesser extent, a 2b or 3b, which is obviously worth more than a walk) and work out the run expectancy of all those events. We can then compare that to how many runs score when he’s put on first base for free. There’s probably quite large error bars associated with every stage of this analysis, so, unless there’s a big difference between the results (in runs) of the two events, you could maybe take it with a pinch of salt…

Again, Albert Pujols is a .460ish wOBA player. You can calculate the odds of him generating runs given the bases occupied state. That’s what we are comparing to here (well, what I compared to in my spreadsheet; the table above was just a quick’n’dirty estimate).

the Cards need two good players that can get on base in front of Albert if he’s going to continue to hit 3rd. i’d actually rather see him hit 4th but i won’t get into that now. which is why every time Ank hits 2nd, many rightly have a fit. to say the dude is a crap shoot when he comes to the plate is an understatement.

as the roster stands now, outside of Colby, who can lead off? let me rephrase that, who should lead off? wasnt is you chuck that pointed out earlier this week that Skippy sucks so far this season, and he really sucks as a lead off guy? now it could be the jager talking, but why don’t the Cards have Colby lead off, Albert hit 2nd or 4th, then go with some combo of Luddy, Dunk, Yadda, Ank, Boog/Greene, Greene, P? makes sense to me.

one quick (if it can be) question about the book. does it factor in any human elements or current stats? like how the players in front of & directly behind the batter being IBB are hitting say for that month? does it factor in the same for the pitcher? those are things that most managers & pitching coaches use most to determine how said pitcher will pitch to said batter right? are they wrong then if they chose to pitch to a batter because the batters behind him are doing a lot better over the past month? because they would rather their pitcher face that batter because he’s “struggling” right now vs the guys behind him who are “hot”.

I wonder if the IBB totals in the NL are higher because managers pitch around batters to get to the pitcher

I would be astonished if the answer is not “yes”.