I almost fell out of my chair when McCarver was talking about this. It's one thing to not understand/care about sabermetrics, but this is a complete inability to think intelligently. How someone could say/think something like this is simply beyond me.

"The gyroball is like a screwball, see how he opened the door the other way there?". Or he just threw a forkball. I can't listen to him.

When I first moved to New York, I was very impressed by McCarver as an announcer.

He's become an old, silly lady.

It's clear to everyone I know, it's time to replace that booth. We don't really need more threads on it.

I take complaints about announcers and sportswriters with a big grain of salt because (a) I focus on the event, not what being said; often I'm not listening anyway; (b) it's a tough job. If someone does it well, I might acknowledge it; if not, I shrug my shounders; (c) it's too easy to criticize what one cannot do oneself; and (d) it's subjective...the guy you like is the guy that gets on my nerves.

That being said, this comment really gave me pause...I actually backed it up twice to see if I heard it right.

Tim, Tim, Tim...what are you thinking?

If McCarver had gone on to say why the conventional wisdom was wrong, I'd've been a lot happier. But clearly it surprised him, too. Think, man!

I dunno, anything McCarver says that shows he's even aware there is such a thing as a run state is kind of surprising.

So that’s the math. But do you really need it? A home run is one guaranteed whole run that no one can take away. The lead off walk increases your odds of scoring exactly one run by roughly 14%. The lead off home run increases those odds by ONE HUNDRED FREAKING PERCENT.

I'm no math wiz myself, but I'm thinking the odds of scoring after a leadoff homerun are not increased by 100 percent.

Who would you like to see as a color analyst?

I have no idea if he'd be any good or not, but I maintain that someone needs to give Doug Glanville a shot either in the booth or on a show like Baseball Tonight.

Who would you like to see as a color analyst?


Shigetoshi Hasegawa

Yeah, I agree with No. 7. Whether or not McCarver is an idiot, that 100%>14% analysis is irrelevant. "[S]coring exactly one run" is not part of this discussion.

McCarver is saying the likelihood of someone else getting on base and scoring is higher after a homer than a walk -- that's what it takes to have a multi-run inning.

Yeah, I agree with No. 7. Whether or not McCarver is an idiot, that 100%>14% analysis is irrelevant. "[S]coring exactly one run" is not part of this discussion.

Doesn't it have to be since order to score more than one run, you must first score one run?

Who would you like to see as a color analyst?

I'm with Posnanski's recent blog post. Get a player who's been on the field within the past five years. Color guys aren't like Twinkies. They are perishable.

Doesn't it have to be since order to score more than one run, you must first score one run?

If that second run scores, the guy who walked to start the inning is going to score about 99 percent of the time, too.

If that second run scores, the guy who walked to start the inning is going to score about 99 percent of the time, too.

Ok, I'm getting confused now. Isn't the situation simply that in McCarver's head multi-run innings are more likely to occur when the leadoff guy walks rather than when he homers because home runs kill rallies when in reality a multi-run inning is more likely to occur when the leadoff guy homers because of the simple fact that one run has already been scored and now you just need to get the second. I'm not sure what exactly it is you were saying in the first post I quoted.

By the way, as I wrote my post, it occurred to me that since the "guys in the truck" have already made a fool of McCarver with this stat at least once this season...is it possible that they knew this would be his reaction and are actively attempting to make him look dumb...er?

I think his surprise comes from the relative frequency of the two events. Leadoff walks happen more often than leadoff HR. Thus, they're more on the mind and, rightly, a thing to be feared. When McCarver was playing, leadoff HR were an uncommon event and while they hurt you and, it should be obvious, more often lead to multi-run innings, they just aren't common enough to stick in your head.

It's one thing to not understand/care about sabermetrics, but this is a complete inability to think intelligently.

Thinking about baseball statistics critically has made me a LOT smarter about things like this. I'm not surprised that someone of average intelligence that isn't all nerdy about numbers won't grasp this concept. If we were able to sit McCarver down and explain it to him slowly, and he still didn't get it - then I call him unintelligent. Until then I just think he's lazy.

I think this whole discussion is silly, but I think there is a grain of a point hidden in what McCarver is trying to get to in making the comparison between lead-off walks and lead-off home runs, and whether there is something about the former that might produce more multi-run innings whereas lead-off home-runs produce more innings which stop with just that one run and ONLY that one run.

Now, obviously (for exactly the reason NJASDJDH points out), it shouldn't be surprising at all that you get more multi-run innings out of the lead-off HR scenario, simply because you get more ZERO run innings out of the lead-off BB scenario. You will NEVER get a zero out of the HR inning; never. Since you are starting with a fresh slate (one run in, nobody on, nobody out), you are going to get some (presumably normal) percentage of innings where you'd score each of three possibilities: zero, one, and more than one runs from that point onward. ONLY in the first category (zero-run innings) would you end up without a multi-run inning on the board.

In the lead-off walk scenario, since your "starting point" is zero runs in, of the three categories I mentioned above of how the inning might proceed going forward (zero additional runs, one more, more than one more), it is in each of the first two (0 more, or 1 more) that you end up without a multi-run inning. ONLY if you score more than one do you end up with a multi-run inning. Granted, you have a runner on first (the lead-off walk) to get you started, but that's not a good enough head-start towards the multi-run inning to make a difference.

OK, so that explains why McCarver shouldn't be surprised. Duh.

But. I do think there is a grain of a point he is trying to make, which is that the home run leaves a situation (nobody on) that doesn't build towards anything more than it has produced all by itself. It has produced that run, already on the scoreboard, which is great and a much bigger advantage in whether you are likely to end up with more before you use up your three outs. The lead-off walk, however, does create on-going offense.

So here's the question: if we limit the universe of innings in which the lead-off walk occurs to those in which it DOES produce a single run (thus equalizing it to the lead-off home run situation in that respect), is the team then more likely to score additional runs? I think the answer is obviously yes. Why? Because we know that other things have to have happened to produce the first run -- more walks, more hits, etc. And almost all of those "more things" are themselves likely to lead to additional runs beyond the first. Indeed, at least one of those things could be a home run after the walk, which is automatically a multi-run inning right there.

So IF the walk eventually gets around to helping produce one run, it's also more likely than the solo, lead-off homer to produce a multi-run inning. But that's a huge if.

Ok, I'm getting confused now. Isn't the situation simply that in McCarver's head multi-run innings are more likely to occur when the leadoff guy walks rather than when he homers because home runs kill rallies when in reality a multi-run inning is more likely to occur when the leadoff guy homers because of the simple fact that one run has already been scored and now you just need to get the second.

It's very possible that I'm confused too, but here's how I see it. Say there's a leadoff homer. What needs to happen for this to become a multi-run inning? Someone else has to get on base and come around to score.

Now suppose there's a leadoff walk. What needs to happen for this to become a multi-run inning? Someone else needs to get on base behind the leadoff man, and both runners need to come around to score. If the second runner scores, the first runner is very likely to score too.

Is that less likely to happen than in the leadoff homer scenario? Yes, because of the possibility of the double play, and I suppose there's some issues of pitcher quality involved. But it doesn't seem extremely less likely to me than the leadoff-homer scenario.

Is that less likely to happen than in the leadoff homer scenario? Yes, because of the possibility of the double play, and I suppose there's some issues of pitcher quality involved.

Wouldn't one of the other key factors be that teams that put the leadoff runner on quite often attempt strategies specifically designed toward scoring that one run (stolen base, sac bunt, hitting behind the runner), all of which to varying degree lower the chances of scoring multiple runs.

By the way, my post in No. 7 was merely meant to note that the writer mistated his point. The odds of scoring one run after a leadoff home run are 100 percent. The odds of scoring one run don't "increase by" 100 percent. If you're going to mock McCarver's idiocy, then it's probably a good idea not to do it through idiocy.

Another reason I could see for McCarver's POV is that it is easy to rationalize a homer as one bad pitch, whereas a walk more likely involves several bad or misplaced pitches. If that's the case, maybe McCarver's thinking is that a leadoff walk means the guy is doing worse than a leadoff homer and because of that it is more likely the opposition continues to get to him.

Base out matrix:
No runners, none out (post-HR):
1.555

R1, none out (post-BB):
0.953

Derf.
Odds of the next guy getting a hit is good though: AL bases empty OPS: 0.742. R1: 0.788.

But, R12: 0.715 So who knows. I think the leadoff HR wins everytime, and it is pretty obvious to me.

I laughed out loud during the game last night when he said it, and I find it hilarious that we're even discussing this.

No runners, none out (post-HR):
1.555

R1, none out (post-BB):
0.953

I don't know how big a difference it makes, but this isn't exactly what McCarver was saying. It's not the total number of runs likely to score, but the number of times that more than one run scores.

A leadoff BB produces more ensuing dramatic tension than does a leadoff HR. My guess is that McCarver, in his muddle-headed way, thinks dramatic tension is better than runs. He's not alone in this -- didn't Tom Seaver once say that a double would sustain a rally but that a HR would kill it? The BB or double, with its greater follow-up suspense, gives the announcer more material to work with. Not surprising that announcers confuse their own needs with that of the team.

I don't know how big a difference it makes, but this isn't exactly what McCarver was saying. It's not the total number of runs likely to score, but the number of times that more than one run scores.

Well, in a sense, these numbers come from the total times this situation occurred and han average of how many runs were scored. Since we are looking at 4 years worth of data, it is safe to assume we aren't looking at some skew due to 7-run innings versus three 2-run innings.

So, I'd say it does address what McCarver was saying.

So, I'd say it does address what McCarver was saying.

Then what's the answer? How often does a leadoff HR lead to a multirun inning, and how often does a leadoff walk do so?

Tony Muser said that a double is better than a HR for continuing a rally.

McCarver on "Family Guy"

How often does a leadoff HR lead to a multirun inning, and how often does a leadoff walk do so?


A ratio of 3 to 2 in favor of the HR.

HTH.

A ratio of 3 to 2 in favor of the HR.

Thanks. That's what I thought all along... McCarver was wrong to favor the walk, but it's not ridiculously out of line to think that. It's not 100 to 14 or anything.

"Tony Muser said that a double is better than a HR for continuing a rally."

This I could almost see. If you're talking about a specific situation where you need multiple runs (like the bottom of the ninth). The double makes a DP unlikely, and hitters do better with runners on base.

HE THROWS 6 DIFFERENT PITCHES!!! HOW CAN HE DO THAT IF THE CATCHER ONLY HAS 5 FINGERS?!?!?!?#@!*)!&@!?!?!ONEONEONE

Is a pitcher who gives up a leadoff HR more likely to be struggling than one who gives up a leadoff walk?

and don't get me started on mccarver's side by side comparison of indians pitchers windups.

i'd love to see joe buck respond with "so, what you're saying is that the pitcher should just balk home the runner?"

How often does a leadoff HR lead to a multirun inning, and how often does a leadoff walk do so?

And if you count only those innings in which the leadoff walk pays off in at least one run, I bet those innings are more likely than the innings with a leadoff HR to end up with additional runs scoring as well. McCarver's problem is that he attributes this to the leadoff walk, rather than to the additional offense that came afterwards that allowed the walk to turn into a run. The BEST one could say for this is that it might be possible that a leadoff walk helps to bring about the additional offense, to a greater degree than the home run. But I'd need to see some evidence of this -- some data showing that additional offense is actually more likely to follow a leadoff walk than a leadoff home run. Even if such data was shown, it wouldn't be enough to make up for the many innings where the leadoff walk led to squadoosh.

McCarver does have a point.

I believe the data he checked were only the multi-run innings from this past season. He then found more of those innings were started by a lead off home run than a lead off walk. I don't think this is either strongly intuitive or counter intuitive. Remember, lead off walks occur much more frequently than lead off homers.

His mistake was to confuse these odds with the likelihood of a given event leading to multi-run inning. In which case, lead off home runs obviously rank first.

"Tony Muser said that a double is better than a HR for continuing a rally."

This I could almost see. If you're talking about a specific situation where you need multiple runs (like the bottom of the ninth). The double makes a DP unlikely, and hitters do better with runners on base.

Boy, you must not've been here when this was going around. Did Muser ever come in for some snark. I thought it was rather unfair, but since Neyer made no bones about hating Muser, it was pointless to stick up for him in these parts.

i'd love to see joe buck respond with "so, what you're saying is that the pitcher should just balk home the runner?"

Even Joe Buck wouldn't ask that, mainly because of squadoosh.

How often does a leadoff HR lead to a multirun inning, and how often does a leadoff walk do so?


A ratio of 3 to 2 in favor of the HR.

HTH.

I think this is wrong, unless you are using actual data from Retrosheet or something.

If you look at the run frequency matrix, the frequency of getting 1 or more runs with none out, none on is about .29. This is the chance of scoring multiple runs after leading off with a home run.
The frequency of scoring 2 or more runs with a runner on first and none out is about .26. The ratio to me seems to be about 10:9.

The main difference, I'm sure, is the double play. Perhaps McCarver has heard of it.

Well, in a sense, these numbers come from the total times this situation occurred and han average of how many runs were scored. Since we are looking at 4 years worth of data, it is safe to assume we aren't looking at some skew due to 7-run innings versus three 2-run innings.

So, I'd say it does address what McCarver was saying.

No it doesn't. The skew isn't in 7 run vs. 2 run innings. The skew is in 1 run vs. 0 run innings in which the solo HR has a massive advantage. However these situation are, by definition, irrelevant to the question.

Yet another example of saberists not understanding the meaning of the numbers they generate.


The walk leads to fewer multi-run innings only because of the possibility of an out on base of which pointed out the double play is the most common situation.

Edited for clarity and politeness.

From the historical run frequency matrix, the chance of scoring multiple runs with a man on 2nd, no out, is .28. You can still pick the guy off, or Ortiz might run into the ball.

No it doesn't. The skew isn't in 7 run vs. 2 run innings. The skew is in 1 run vs. 0 run innings in which the solo HR has a massive advantage. However these situation are, by definition, irrelevant to the question.

My mistake.

Yet another example of saberists not understanding the meaning of the numbers they generate.

Edited for clarity and politeness.

It wasn't either, IMO.

Yet another example of saberists acting like a superior ass needlessly.

Not edited for clarity and politeness.

I think this is wrong, unless you are using actual data from Retrosheet or something.

I wasn't. I was trying to do *exactly* what you did below, just from the wrong chart.

If you look at the run frequency matrix, the frequency of getting 1 or more runs with none out, none on is about .29. This is the chance of scoring multiple runs after leading off with a home run.
The frequency of scoring 2 or more runs with a runner on first and none out is about .26. The ratio to me seems to be about 10:9.

Closer than I'd think, but yes, this is what I was trying to do.

Yet another example of saberists acting like a superior ass needlessly.

Not edited for clarity and politeness.

Guilty. Though it was even less polite the first time. However since you admitted your mistake I'll admit mine.

However since you admitted your mistake I'll admit mine.

I couldn't admit mine until it was clarified how it was wrong, as Aunt Bea did (assuming Aunt Bea has it right)..

This I could almost see. If you're talking about a specific situation where you need multiple runs (like the bottom of the ninth). The double makes a DP unlikely, and hitters do better with runners on base.

Boy, you must not've been here when this was going around. Did Muser ever come in for some snark. I thought it was rather unfair, but since Neyer made no bones about hating Muser, it was pointless to stick up for him in these parts.

Yeah, but in the specific instance of the multi-run, ninth-inning deficit, lots of really stupid things become okay to do. A runner on third breaking for home on a tapper back to the pitcher creates some chance that the pitcher will panic and pass up the sure out; the possibility of preserving the out is worth the likelihood of giving up bases when you absolutely must have multiple runs. Any manager who said breaking for home was the preferred play without that qualifier, though, is a total dumbass and entirely deserving of Primate snark.

If you look at the run frequency matrix, the frequency of getting 1 or more runs with none out, none on is about .29. This is the chance of scoring multiple runs after leading off with a home run.
The frequency of scoring 2 or more runs with a runner on first and none out is about .26. The ratio to me seems to be about 10:9.

I'm not sure how the run frequency matrix is calculated, but I don't understand why the frequency of getting 1 or more runs with none out, none on would be exactly the same at the beginning of an inning, and after a HR. Wouldn't the HR be an indicator of an inferior pitcher, or a tiring pitcher or a pitcher whose mechanics are off or something? That's probably the angle McCarver was taking - maybe when he played, pitchers with poor control were more likely to be prone to big innings.

Same for runner on first, zero outs - I'm not buying the assumption that the expectation would be exactly the same after a single or a walk.

Yes, I think it's closer than one might immediately think due to the point made in 11.

Put another way, the following sequences of events lead to the same amount of runs:

HR, 2B, 2B
BB, 2B, 2B

HR, BB, 3B,
BB, BB, 3B

I don't really know why anyone would think the walk would produce MORE, but it's somewhat reasonable to think the walk would produce around the same. GIDPs screw it up.

That's probably the angle McCarver was taking - maybe when he played, pitchers with poor control were more likely to be prone to big innings.

Please don't make up excuses for him. This is almost certainly untrue.

Same for runner on first, zero outs - I'm not buying the assumption that the expectation would be exactly the same after a single or a walk.

It's not exactly the same (nothing is). But I have seen the numbers on this one and they are very very similar.

I'm not sure how the run frequency matrix is calculated, but I don't understand why the frequency of getting 1 or more runs with none out, none on would be exactly the same at the beginning of an inning, and after a HR. Wouldn't the HR be an indicator of an inferior pitcher, or a tiring pitcher or a pitcher whose mechanics are off or something?

I'm pretty sure the expectations are calculated from all none on none out situations, not just the start of an inning. You'd have to look at several seasons' worth of data on the following a leadoff HR situation to know if it's different.

Please don't make up excuses for him. This is almost certainly untrue.

I don't know if it is, but nobody else around here seems to know either, and that's the issue.

Please read the book thread by people who actually have done the research.

Leadoff walks are no more likely to score than leadoff singles.

I can't believe the thread has gone this far.

The only way McCarver could be right is if a pitcher giving up a leadoff home run actually is performing better than the average pitcher who gives up a walk or single to lead off. Sure sounds wrong to me.

I can't believe the thread has gone this far.

I can't believe its gone this far as a serious discussion without wandering off into a mock-fest of Joe Buck and McCarver.